From df149e1509d972b2d5bfe903531d9670c2fe83c7 Mon Sep 17 00:00:00 2001 From: Filippos Karapetis Date: Sat, 23 Jan 2010 17:55:54 +0000 Subject: Separated the parser code svn-id: r47480 --- engines/sci/console.cpp | 4 +- engines/sci/engine/grammar.cpp | 583 --------- engines/sci/engine/kernel.h | 1 - engines/sci/engine/kparse.cpp | 210 ++++ engines/sci/engine/kstring.cpp | 176 --- engines/sci/engine/said.cpp | 2483 ------------------------------------- engines/sci/engine/said.y | 838 ------------- engines/sci/engine/state.h | 4 +- engines/sci/module.mk | 6 +- engines/sci/parser/grammar.cpp | 583 +++++++++ engines/sci/parser/said.cpp | 2483 +++++++++++++++++++++++++++++++++++++ engines/sci/parser/said.y | 838 +++++++++++++ engines/sci/parser/vocabulary.cpp | 633 ++++++++++ engines/sci/parser/vocabulary.h | 338 +++++ engines/sci/resource.cpp | 2 +- engines/sci/vocabulary.cpp | 633 ---------- engines/sci/vocabulary.h | 338 ----- 17 files changed, 5095 insertions(+), 5058 deletions(-) delete mode 100644 engines/sci/engine/grammar.cpp create mode 100644 engines/sci/engine/kparse.cpp delete mode 100644 engines/sci/engine/said.cpp delete mode 100644 engines/sci/engine/said.y create mode 100644 engines/sci/parser/grammar.cpp create mode 100644 engines/sci/parser/said.cpp create mode 100644 engines/sci/parser/said.y create mode 100644 engines/sci/parser/vocabulary.cpp create mode 100644 engines/sci/parser/vocabulary.h delete mode 100644 engines/sci/vocabulary.cpp delete mode 100644 engines/sci/vocabulary.h (limited to 'engines') diff --git a/engines/sci/console.cpp b/engines/sci/console.cpp index 2a1adb3fcf..1e472839af 100644 --- a/engines/sci/console.cpp +++ b/engines/sci/console.cpp @@ -30,7 +30,6 @@ #include "sci/debug.h" #include "sci/event.h" #include "sci/resource.h" -#include "sci/vocabulary.h" #include "sci/engine/savegame.h" #include "sci/engine/state.h" #include "sci/engine/gc.h" @@ -42,10 +41,11 @@ #include "sci/sound/music.h" #endif #include "sci/sound/drivers/mididriver.h" -#include "sci/vocabulary.h" #include "sci/graphics/gui.h" #include "sci/graphics/cursor.h" +#include "sci/parser/vocabulary.h" + #include "graphics/video/avi_decoder.h" #include "sci/video/seq_decoder.h" #ifdef ENABLE_SCI32 diff --git a/engines/sci/engine/grammar.cpp b/engines/sci/engine/grammar.cpp deleted file mode 100644 index 8c6c93e583..0000000000 --- a/engines/sci/engine/grammar.cpp +++ /dev/null @@ -1,583 +0,0 @@ -/* ScummVM - Graphic Adventure Engine - * - * ScummVM is the legal property of its developers, whose names - * are too numerous to list here. Please refer to the COPYRIGHT - * file distributed with this source distribution. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - * - * $URL$ - * $Id$ - * - */ - -/* Functionality to transform the context-free SCI grammar rules into - * strict Greibach normal form (strict GNF), and to test SCI input against - * that grammar, writing an appropriate node tree if successful. - */ - -#include "sci/vocabulary.h" -#include "sci/console.h" -#include "common/array.h" - -namespace Sci { - -#define TOKEN_OPAREN 0xff000000 -#define TOKEN_CPAREN 0xfe000000 -#define TOKEN_TERMINAL_CLASS 0x10000 -#define TOKEN_TERMINAL_GROUP 0x20000 -#define TOKEN_STUFFING_WORD 0x40000 -#define TOKEN_NON_NT (TOKEN_OPAREN | TOKEN_TERMINAL_CLASS | TOKEN_TERMINAL_GROUP | TOKEN_STUFFING_WORD) -#define TOKEN_TERMINAL (TOKEN_TERMINAL_CLASS | TOKEN_TERMINAL_GROUP) - -static int _allocd_rules = 0; // FIXME: Avoid non-const global vars - -struct ParseRule { - int _id; /**< non-terminal ID */ - uint _firstSpecial; /**< first terminal or non-terminal */ - uint _numSpecials; /**< number of terminals and non-terminals */ - Common::Array _data; /**< actual data */ - - ~ParseRule() { - assert(_allocd_rules > 0); - --_allocd_rules; - } - - // FIXME remove this one again? - bool operator==(const ParseRule &other) const { - return _id == other._id && - _firstSpecial == other._firstSpecial && - _numSpecials == other._numSpecials && - _data == other._data; - } -}; - - -struct ParseRuleList { - int terminal; /**< Terminal character this rule matches against or 0 for a non-terminal rule */ - ParseRule *rule; - ParseRuleList *next; - - void print() const; - - ParseRuleList(ParseRule *r) : rule(r), next(0) { - int term = rule->_data[rule->_firstSpecial]; - terminal = ((term & TOKEN_TERMINAL) ? term : 0); - } - - ~ParseRuleList() { - delete rule; - delete next; - } -}; - - -static void vocab_print_rule(ParseRule *rule) { - int wspace = 0; - - if (!rule) { - warning("NULL rule"); - return; - } - - printf("[%03x] -> ", rule->_id); - - if (rule->_data.empty()) - printf("e"); - - for (uint i = 0; i < rule->_data.size(); i++) { - uint token = rule->_data[i]; - - if (token == TOKEN_OPAREN) { - if (i == rule->_firstSpecial) - printf("_"); - - printf("("); - wspace = 0; - } else if (token == TOKEN_CPAREN) { - if (i == rule->_firstSpecial) - printf("_"); - - printf(")"); - wspace = 0; - } else { - if (wspace) - printf(" "); - - if (i == rule->_firstSpecial) - printf("_"); - if (token & TOKEN_TERMINAL_CLASS) - printf("C(%04x)", token & 0xffff); - else if (token & TOKEN_TERMINAL_GROUP) - printf("G(%04x)", token & 0xffff); - else if (token & TOKEN_STUFFING_WORD) - printf("%03x", token & 0xffff); - else - printf("[%03x]", token); /* non-terminal */ - wspace = 1; - } - - if (i == rule->_firstSpecial) - printf("_"); - } - printf(" [%d specials]", rule->_numSpecials); -} - -static ParseRule *_vdup(ParseRule *a) { - ++_allocd_rules; - return new ParseRule(*a); -} - -static ParseRule *_vinsert(ParseRule *turkey, ParseRule *stuffing) { - uint firstnt = turkey->_firstSpecial; - - // Search for first TOKEN_NON_NT in 'turkey' - while ((firstnt < turkey->_data.size()) && (turkey->_data[firstnt] & TOKEN_NON_NT)) - firstnt++; - - // If no TOKEN_NON_NT found, or if it doesn't match the id of 'stuffing', abort. - if ((firstnt == turkey->_data.size()) || (turkey->_data[firstnt] != stuffing->_id)) - return NULL; - - // Create a new rule as a copy of 'turkey', where the token firstnt has been substituted - // by the rule 'stuffing'. - ++_allocd_rules; - - ParseRule *rule = new ParseRule(*turkey); - rule->_numSpecials += stuffing->_numSpecials - 1; - rule->_firstSpecial = firstnt + stuffing->_firstSpecial; - rule->_data.resize(turkey->_data.size() - 1 + stuffing->_data.size()); - - // Replace rule->_data[firstnt] by all of stuffing->_data - Common::copy(stuffing->_data.begin(), stuffing->_data.end(), rule->_data.begin() + firstnt); - - if (firstnt < turkey->_data.size() - 1) - Common::copy(turkey->_data.begin() + firstnt + 1, turkey->_data.end(), - rule->_data.begin() + firstnt + stuffing->_data.size()); - - return rule; -} - -static ParseRule *_vbuild_rule(const parse_tree_branch_t *branch) { - int tokens = 0, tokenpos = 0, i; - - while (tokenpos < 10 && branch->data[tokenpos]) { - int type = branch->data[tokenpos]; - tokenpos += 2; - - if ((type == VOCAB_TREE_NODE_COMPARE_TYPE) || (type == VOCAB_TREE_NODE_COMPARE_GROUP) || (type == VOCAB_TREE_NODE_FORCE_STORAGE)) - ++tokens; - else if (type > VOCAB_TREE_NODE_LAST_WORD_STORAGE) - tokens += 5; - else - return NULL; // invalid - } - - ParseRule *rule = new ParseRule(); - - ++_allocd_rules; - rule->_id = branch->id; - rule->_numSpecials = tokenpos >> 1; - rule->_data.resize(tokens); - rule->_firstSpecial = 0; - - tokens = 0; - for (i = 0; i < tokenpos; i += 2) { - int type = branch->data[i]; - int value = branch->data[i + 1]; - - if (type == VOCAB_TREE_NODE_COMPARE_TYPE) - rule->_data[tokens++] = value | TOKEN_TERMINAL_CLASS; - else if (type == VOCAB_TREE_NODE_COMPARE_GROUP) - rule->_data[tokens++] = value | TOKEN_TERMINAL_GROUP; - else if (type == VOCAB_TREE_NODE_FORCE_STORAGE) - rule->_data[tokens++] = value | TOKEN_STUFFING_WORD; - else { // normal inductive rule - rule->_data[tokens++] = TOKEN_OPAREN; - rule->_data[tokens++] = type | TOKEN_STUFFING_WORD; - rule->_data[tokens++] = value | TOKEN_STUFFING_WORD; - - if (i == 0) - rule->_firstSpecial = tokens; - - rule->_data[tokens++] = value; // The non-terminal - rule->_data[tokens++] = TOKEN_CPAREN; - } - } - - return rule; -} - -static ParseRule *_vsatisfy_rule(ParseRule *rule, const ResultWord &input) { - int dep; - - if (!rule->_numSpecials) - return NULL; - - dep = rule->_data[rule->_firstSpecial]; - - if (((dep & TOKEN_TERMINAL_CLASS) && ((dep & 0xffff) & input._class)) || - ((dep & TOKEN_TERMINAL_GROUP) && ((dep & 0xffff) & input._group))) { - ParseRule *retval = new ParseRule(*rule); - ++_allocd_rules; - retval->_data[rule->_firstSpecial] = TOKEN_STUFFING_WORD | input._group; - retval->_numSpecials--; - retval->_firstSpecial = 0; - - if (retval->_numSpecials) { // find first special, if it exists - for (uint i = rule->_firstSpecial; i < retval->_data.size(); ++i) { - int tmp = retval->_data[i]; - if (!(tmp & TOKEN_NON_NT) || (tmp & TOKEN_TERMINAL)) { - retval->_firstSpecial = i; - break; - } - } - } - - return retval; - } else - return NULL; -} - -void Vocabulary::freeRuleList(ParseRuleList *list) { - delete list; -} - -static ParseRuleList *_vocab_add_rule(ParseRuleList *list, ParseRule *rule) { - if (!rule) - return list; - - ParseRuleList *new_elem = new ParseRuleList(rule); - - if (list) { - const int term = new_elem->terminal; -/* if (term < list->terminal) { - new_elem->next = list; - return new_elem; - } else {*/ - ParseRuleList *seeker = list; - - while (seeker->next/* && seeker->next->terminal <= term*/) { - if (seeker->next->terminal == term) { - if (*(seeker->next->rule) == *rule) { - delete rule; - delete new_elem; - return list; // No duplicate rules - } - } - seeker = seeker->next; - } - - new_elem->next = seeker->next; - seeker->next = new_elem; - return list; - } else { - return new_elem; - } -} - -void ParseRuleList::print() const { - const ParseRuleList *list = this; - int pos = 0; - while (list) { - printf("R%03d: ", pos); - vocab_print_rule(list->rule); - printf("\n"); - list = list->next; - pos++; - } - printf("%d rules total.\n", pos); -} - -static ParseRuleList *_vocab_split_rule_list(ParseRuleList *list) { - assert(list); - if (!list->next || (list->next->terminal)) { - ParseRuleList *tmp = list->next; - list->next = NULL; - return tmp; - } else - return _vocab_split_rule_list(list->next); -} - -static void _vocab_free_empty_rule_list(ParseRuleList *list) { - assert(list); - if (list->next) - _vocab_free_empty_rule_list(list->next); - list->next = 0; - list->rule = 0; - delete list; -} - -static ParseRuleList *_vocab_merge_rule_lists(ParseRuleList *l1, ParseRuleList *l2) { - ParseRuleList *retval = l1, *seeker = l2; - while (seeker) { - retval = _vocab_add_rule(retval, seeker->rule); - seeker = seeker->next; - } - _vocab_free_empty_rule_list(l2); - - return retval; -} - -static int _vocab_rule_list_length(ParseRuleList *list) { - return ((list) ? _vocab_rule_list_length(list->next) + 1 : 0); -} - -static ParseRuleList *_vocab_clone_rule_list_by_id(ParseRuleList *list, int id) { - ParseRuleList *result = NULL; - ParseRuleList *seeker = list; - - while (seeker) { - if (seeker->rule->_id == id) { - result = _vocab_add_rule(result, _vdup(seeker->rule)); - } - seeker = seeker->next; - } - - return result; -} - -ParseRuleList *Vocabulary::buildGNF(bool verbose) { - int iterations = 0; - int last_termrules, termrules = 0; - int ntrules_nr; - ParseRuleList *ntlist = NULL; - ParseRuleList *tlist, *new_tlist; - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - - for (uint i = 1; i < _parserBranches.size(); i++) { // branch rule 0 is treated specially - ParseRule *rule = _vbuild_rule(&_parserBranches[i]); - if (!rule) - return NULL; - ntlist = _vocab_add_rule(ntlist, rule); - } - - tlist = _vocab_split_rule_list(ntlist); - ntrules_nr = _vocab_rule_list_length(ntlist); - - if (verbose) - con->DebugPrintf("Starting with %d rules\n", ntrules_nr); - - new_tlist = tlist; - tlist = NULL; - - do { - ParseRuleList *new_new_tlist = NULL; - ParseRuleList *ntseeker, *tseeker; - last_termrules = termrules; - - ntseeker = ntlist; - while (ntseeker) { - tseeker = new_tlist; - - while (tseeker) { - ParseRule *newrule = _vinsert(ntseeker->rule, tseeker->rule); - if (newrule) - new_new_tlist = _vocab_add_rule(new_new_tlist, newrule); - tseeker = tseeker->next; - } - - ntseeker = ntseeker->next; - } - - tlist = _vocab_merge_rule_lists(tlist, new_tlist); - new_tlist = new_new_tlist; - termrules = _vocab_rule_list_length(new_new_tlist); - - if (verbose) - con->DebugPrintf("After iteration #%d: %d new term rules\n", ++iterations, termrules); - - } while (termrules && (iterations < 30)); - - freeRuleList(ntlist); - - if (verbose) { - con->DebugPrintf("\nGNF rules:\n"); - tlist->print(); - con->DebugPrintf("%d allocd rules\n", _allocd_rules); - con->DebugPrintf("Freeing rule list...\n"); - freeRuleList(tlist); - return NULL; - } - - return tlist; -} - -static int _vbpt_pareno(parse_tree_node_t *nodes, int *pos, int base) { - // Opens parentheses - nodes[base].content.branches[0] = (*pos) + 1; - nodes[++(*pos)].type = kParseTreeBranchNode; - nodes[*pos].content.branches[0] = 0; - nodes[*pos].content.branches[1] = 0; - return *pos; -} - -static int _vbpt_parenc(parse_tree_node_t *nodes, int *pos, int paren) { - // Closes parentheses for appending - nodes[paren].content.branches[1] = ++(*pos); - nodes[*pos].type = kParseTreeBranchNode; - nodes[*pos].content.branches[0] = 0; - nodes[*pos].content.branches[1] = 0; - return *pos; -} - -static int _vbpt_append(parse_tree_node_t *nodes, int *pos, int base, int value) { - // writes one value to an existing base node and creates a successor node for writing - nodes[base].content.branches[0] = ++(*pos); - nodes[*pos].type = kParseTreeLeafNode; - nodes[*pos].content.value = value; - nodes[base].content.branches[1] = ++(*pos); - nodes[*pos].type = kParseTreeBranchNode; - nodes[*pos].content.branches[0] = 0; - nodes[*pos].content.branches[1] = 0; - return *pos; -} - -static int _vbpt_terminate(parse_tree_node_t *nodes, int *pos, int base, int value) { - // Terminates, overwriting a nextwrite forknode - nodes[base].type = kParseTreeLeafNode; - nodes[base].content.value = value; - return *pos; -} - -static int _vbpt_write_subexpression(parse_tree_node_t *nodes, int *pos, ParseRule *rule, uint rulepos, int writepos) { - uint token; - - while ((token = ((rulepos < rule->_data.size()) ? rule->_data[rulepos++] : TOKEN_CPAREN)) != TOKEN_CPAREN) { - uint nexttoken = (rulepos < rule->_data.size()) ? rule->_data[rulepos] : TOKEN_CPAREN; - if (token == TOKEN_OPAREN) { - int writepos2 = _vbpt_pareno(nodes, pos, writepos); - rulepos = _vbpt_write_subexpression(nodes, pos, rule, rulepos, writepos2); - nexttoken = (rulepos < rule->_data.size()) ? rule->_data[rulepos] : TOKEN_CPAREN; - if (nexttoken != TOKEN_CPAREN) - writepos = _vbpt_parenc(nodes, pos, writepos); - } else if (token & TOKEN_STUFFING_WORD) { - if (nexttoken == TOKEN_CPAREN) - writepos = _vbpt_terminate(nodes, pos, writepos, token & 0xffff); - else - writepos = _vbpt_append(nodes, pos, writepos, token & 0xffff); - } else { - printf("\nError in parser (grammar.cpp, _vbpt_write_subexpression()): Rule data broken in rule "); - vocab_print_rule(rule); - printf(", at token position %d\n", *pos); - return rulepos; - } - } - - return rulepos; -} - -int Vocabulary::parseGNF(const ResultWordList &words, bool verbose) { - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - // Get the start rules: - ParseRuleList *work = _vocab_clone_rule_list_by_id(_parserRules, _parserBranches[0].data[1]); - ParseRuleList *results = NULL; - uint word = 0; - const uint words_nr = words.size(); - ResultWordList::const_iterator word_iter = words.begin(); - - for (word_iter = words.begin(); word_iter != words.end(); ++word_iter, ++word) { - ParseRuleList *new_work = NULL; - ParseRuleList *reduced_rules = NULL; - ParseRuleList *seeker, *subseeker; - - if (verbose) - con->DebugPrintf("Adding word %d...\n", word); - - seeker = work; - while (seeker) { - if (seeker->rule->_numSpecials <= (words_nr - word)) - reduced_rules = _vocab_add_rule(reduced_rules, _vsatisfy_rule(seeker->rule, *word_iter)); - - seeker = seeker->next; - } - - if (reduced_rules == NULL) { - freeRuleList(work); - if (verbose) - con->DebugPrintf("No results.\n"); - return 1; - } - - freeRuleList(work); - - if (word + 1 < words_nr) { - seeker = reduced_rules; - - while (seeker) { - if (seeker->rule->_numSpecials) { - int my_id = seeker->rule->_data[seeker->rule->_firstSpecial]; - - subseeker = _parserRules; - while (subseeker) { - if (subseeker->rule->_id == my_id) - new_work = _vocab_add_rule(new_work, _vinsert(seeker->rule, subseeker->rule)); - - subseeker = subseeker->next; - } - } - - seeker = seeker->next; - } - freeRuleList(reduced_rules); - } else // last word - new_work = reduced_rules; - - work = new_work; - if (verbose) - con->DebugPrintf("Now at %d candidates\n", _vocab_rule_list_length(work)); - if (work == NULL) { - if (verbose) - con->DebugPrintf("No results.\n"); - return 1; - } - } - - results = work; - - if (verbose) { - con->DebugPrintf("All results (excluding the surrounding '(141 %03x' and ')'):\n", _parserBranches[0].id); - results->print(); - con->DebugPrintf("\n"); - } - - // now use the first result - { - int temp, pos; - - _parserNodes[0].type = kParseTreeBranchNode; - _parserNodes[0].content.branches[0] = 1; - _parserNodes[0].content.branches[1] = 2; - - _parserNodes[1].type = kParseTreeLeafNode; - _parserNodes[1].content.value = 0x141; - - _parserNodes[2].type = kParseTreeBranchNode; - _parserNodes[2].content.branches[0] = 0; - _parserNodes[2].content.branches[1] = 0; - - pos = 2; - - temp = _vbpt_append(_parserNodes, &pos, 2, _parserBranches[0].id); - //_vbpt_write_subexpression(nodes, &pos, results[_vocab_rule_list_length(results)].rule, 0, temp); - _vbpt_write_subexpression(_parserNodes, &pos, results->rule, 0, temp); - } - - freeRuleList(results); - - return 0; -} - -} // End of namespace Sci diff --git a/engines/sci/engine/kernel.h b/engines/sci/engine/kernel.h index abf090eab6..9ba33aff97 100644 --- a/engines/sci/engine/kernel.h +++ b/engines/sci/engine/kernel.h @@ -31,7 +31,6 @@ #include "common/rect.h" #include "sci/sci.h" // for USE_OLD_MUSIC_FUNCTIONS -#include "sci/vocabulary.h" #include "sci/engine/vm_types.h" // for reg_t #include "sci/engine/vm.h" diff --git a/engines/sci/engine/kparse.cpp b/engines/sci/engine/kparse.cpp new file mode 100644 index 0000000000..5cb24f4fac --- /dev/null +++ b/engines/sci/engine/kparse.cpp @@ -0,0 +1,210 @@ +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +/* String and parser handling */ + +#include "sci/resource.h" +#include "sci/engine/state.h" +#include "sci/engine/message.h" +#include "sci/engine/kernel.h" + +namespace Sci { + +/*************************************************************/ +/* Parser */ +/**********/ + + +reg_t kSaid(EngineState *s, int argc, reg_t *argv) { + reg_t heap_said_block = argv[0]; + byte *said_block; + int new_lastmatch; +#ifdef DEBUG_PARSER + const int debug_parser = 1; +#else + const int debug_parser = 0; +#endif + + if (!heap_said_block.segment) + return NULL_REG; + + said_block = (byte *)s->_segMan->derefBulkPtr(heap_said_block, 0); + + if (!said_block) { + warning("Said on non-string, pointer %04x:%04x", PRINT_REG(heap_said_block)); + return NULL_REG; + } + +#ifdef DEBUG_PARSER + debugC(2, kDebugLevelParser, "Said block:", 0); + s->_voc->decipherSaidBlock(said_block); +#endif + + if (s->parser_event.isNull() || (GET_SEL32V(s->_segMan, s->parser_event, claimed))) { + return NULL_REG; + } + + new_lastmatch = said(s, said_block, debug_parser); + if (new_lastmatch != SAID_NO_MATCH) { /* Build and possibly display a parse tree */ + +#ifdef DEBUG_PARSER + printf("kSaid: Match.\n"); +#endif + + s->r_acc = make_reg(0, 1); + + if (new_lastmatch != SAID_PARTIAL_MATCH) + PUT_SEL32V(s->_segMan, s->parser_event, claimed, 1); + + } else { + return NULL_REG; + } + return s->r_acc; +} + + +reg_t kSetSynonyms(EngineState *s, int argc, reg_t *argv) { + SegManager *segMan = s->_segMan; + reg_t object = argv[0]; + List *list; + Node *node; + int script; + int numSynonyms = 0; + + s->_voc->clearSynonyms(); + + list = s->_segMan->lookupList(GET_SEL32(segMan, object, elements)); + node = s->_segMan->lookupNode(list->first); + + while (node) { + reg_t objpos = node->value; + int seg; + + script = GET_SEL32V(segMan, objpos, number); + seg = s->_segMan->getScriptSegment(script); + + if (seg > 0) + numSynonyms = s->_segMan->getScript(seg)->getSynonymsNr(); + + if (numSynonyms) { + byte *synonyms = s->_segMan->getScript(seg)->getSynonyms(); + + if (synonyms) { + debugC(2, kDebugLevelParser, "Setting %d synonyms for script.%d\n", + numSynonyms, script); + + if (numSynonyms > 16384) { + error("Segtable corruption: script.%03d has %d synonyms", + script, numSynonyms); + /* We used to reset the corrupted value here. I really don't think it's appropriate. + * Lars */ + } else + for (int i = 0; i < numSynonyms; i++) { + synonym_t tmp; + tmp.replaceant = (int16)READ_LE_UINT16(synonyms + i * 4); + tmp.replacement = (int16)READ_LE_UINT16(synonyms + i * 4 + 2); + s->_voc->addSynonym(tmp); + } + } else + warning("Synonyms of script.%03d were requested, but script is not available", script); + + } + + node = s->_segMan->lookupNode(node->succ); + } + + debugC(2, kDebugLevelParser, "A total of %d synonyms are active now.\n", numSynonyms); + + return s->r_acc; +} + + + +reg_t kParse(EngineState *s, int argc, reg_t *argv) { + SegManager *segMan = s->_segMan; + reg_t stringpos = argv[0]; + Common::String string = s->_segMan->getString(stringpos); + char *error; + ResultWordList words; + reg_t event = argv[1]; + Vocabulary *voc = s->_voc; + + s->parser_event = event; + + bool res = voc->tokenizeString(words, string.c_str(), &error); + s->parserIsValid = false; /* not valid */ + + if (res && !words.empty()) { + s->_voc->synonymizeTokens(words); + + s->r_acc = make_reg(0, 1); + +#ifdef DEBUG_PARSER + debugC(2, kDebugLevelParser, "Parsed to the following blocks:\n", 0); + + for (ResultWordList::const_iterator i = words.begin(); i != words.end(); ++i) + debugC(2, kDebugLevelParser, " Type[%04x] Group[%04x]\n", i->_class, i->_group); +#endif + + int syntax_fail = voc->parseGNF(words); + + if (syntax_fail) { + s->r_acc = make_reg(0, 1); + PUT_SEL32V(segMan, event, claimed, 1); + + invoke_selector(INV_SEL(s->_gameObj, syntaxFail, kStopOnInvalidSelector), 2, s->parser_base, stringpos); + /* Issue warning */ + + debugC(2, kDebugLevelParser, "Tree building failed\n"); + + } else { + s->parserIsValid = true; + PUT_SEL32V(segMan, event, claimed, 0); + +#ifdef DEBUG_PARSER + s->_voc->dumpParseTree(); +#endif + } + + } else { + + s->r_acc = make_reg(0, 0); + PUT_SEL32V(segMan, event, claimed, 1); + if (error) { + s->_segMan->strcpy(s->parser_base, error); + debugC(2, kDebugLevelParser, "Word unknown: %s\n", error); + /* Issue warning: */ + + invoke_selector(INV_SEL(s->_gameObj, wordFail, kStopOnInvalidSelector), 2, s->parser_base, stringpos); + free(error); + return make_reg(0, 1); /* Tell them that it didn't work */ + } + } + + return s->r_acc; +} + + +} // End of namespace Sci diff --git a/engines/sci/engine/kstring.cpp b/engines/sci/engine/kstring.cpp index 19c3f1540f..2e4ecce802 100644 --- a/engines/sci/engine/kstring.cpp +++ b/engines/sci/engine/kstring.cpp @@ -66,182 +66,6 @@ Common::String kernel_lookup_text(EngineState *s, reg_t address, int index) { } } - -/*************************************************************/ -/* Parser */ -/**********/ - - -reg_t kSaid(EngineState *s, int argc, reg_t *argv) { - reg_t heap_said_block = argv[0]; - byte *said_block; - int new_lastmatch; -#ifdef DEBUG_PARSER - const int debug_parser = 1; -#else - const int debug_parser = 0; -#endif - - if (!heap_said_block.segment) - return NULL_REG; - - said_block = (byte *)s->_segMan->derefBulkPtr(heap_said_block, 0); - - if (!said_block) { - warning("Said on non-string, pointer %04x:%04x", PRINT_REG(heap_said_block)); - return NULL_REG; - } - -#ifdef DEBUG_PARSER - debugC(2, kDebugLevelParser, "Said block:", 0); - s->_voc->decipherSaidBlock(said_block); -#endif - - if (s->parser_event.isNull() || (GET_SEL32V(s->_segMan, s->parser_event, claimed))) { - return NULL_REG; - } - - new_lastmatch = said(s, said_block, debug_parser); - if (new_lastmatch != SAID_NO_MATCH) { /* Build and possibly display a parse tree */ - -#ifdef DEBUG_PARSER - printf("kSaid: Match.\n"); -#endif - - s->r_acc = make_reg(0, 1); - - if (new_lastmatch != SAID_PARTIAL_MATCH) - PUT_SEL32V(s->_segMan, s->parser_event, claimed, 1); - - } else { - return NULL_REG; - } - return s->r_acc; -} - - -reg_t kSetSynonyms(EngineState *s, int argc, reg_t *argv) { - SegManager *segMan = s->_segMan; - reg_t object = argv[0]; - List *list; - Node *node; - int script; - int numSynonyms = 0; - - s->_voc->clearSynonyms(); - - list = s->_segMan->lookupList(GET_SEL32(segMan, object, elements)); - node = s->_segMan->lookupNode(list->first); - - while (node) { - reg_t objpos = node->value; - int seg; - - script = GET_SEL32V(segMan, objpos, number); - seg = s->_segMan->getScriptSegment(script); - - if (seg > 0) - numSynonyms = s->_segMan->getScript(seg)->getSynonymsNr(); - - if (numSynonyms) { - byte *synonyms = s->_segMan->getScript(seg)->getSynonyms(); - - if (synonyms) { - debugC(2, kDebugLevelParser, "Setting %d synonyms for script.%d\n", - numSynonyms, script); - - if (numSynonyms > 16384) { - error("Segtable corruption: script.%03d has %d synonyms", - script, numSynonyms); - /* We used to reset the corrupted value here. I really don't think it's appropriate. - * Lars */ - } else - for (int i = 0; i < numSynonyms; i++) { - synonym_t tmp; - tmp.replaceant = (int16)READ_LE_UINT16(synonyms + i * 4); - tmp.replacement = (int16)READ_LE_UINT16(synonyms + i * 4 + 2); - s->_voc->addSynonym(tmp); - } - } else - warning("Synonyms of script.%03d were requested, but script is not available", script); - - } - - node = s->_segMan->lookupNode(node->succ); - } - - debugC(2, kDebugLevelParser, "A total of %d synonyms are active now.\n", numSynonyms); - - return s->r_acc; -} - - - -reg_t kParse(EngineState *s, int argc, reg_t *argv) { - SegManager *segMan = s->_segMan; - reg_t stringpos = argv[0]; - Common::String string = s->_segMan->getString(stringpos); - char *error; - ResultWordList words; - reg_t event = argv[1]; - Vocabulary *voc = s->_voc; - - s->parser_event = event; - - bool res = voc->tokenizeString(words, string.c_str(), &error); - s->parserIsValid = false; /* not valid */ - - if (res && !words.empty()) { - s->_voc->synonymizeTokens(words); - - s->r_acc = make_reg(0, 1); - -#ifdef DEBUG_PARSER - debugC(2, kDebugLevelParser, "Parsed to the following blocks:\n", 0); - - for (ResultWordList::const_iterator i = words.begin(); i != words.end(); ++i) - debugC(2, kDebugLevelParser, " Type[%04x] Group[%04x]\n", i->_class, i->_group); -#endif - - int syntax_fail = voc->parseGNF(words); - - if (syntax_fail) { - s->r_acc = make_reg(0, 1); - PUT_SEL32V(segMan, event, claimed, 1); - - invoke_selector(INV_SEL(s->_gameObj, syntaxFail, kStopOnInvalidSelector), 2, s->parser_base, stringpos); - /* Issue warning */ - - debugC(2, kDebugLevelParser, "Tree building failed\n"); - - } else { - s->parserIsValid = true; - PUT_SEL32V(segMan, event, claimed, 0); - -#ifdef DEBUG_PARSER - s->_voc->dumpParseTree(); -#endif - } - - } else { - - s->r_acc = make_reg(0, 0); - PUT_SEL32V(segMan, event, claimed, 1); - if (error) { - s->_segMan->strcpy(s->parser_base, error); - debugC(2, kDebugLevelParser, "Word unknown: %s\n", error); - /* Issue warning: */ - - invoke_selector(INV_SEL(s->_gameObj, wordFail, kStopOnInvalidSelector), 2, s->parser_base, stringpos); - free(error); - return make_reg(0, 1); /* Tell them that it dind't work */ - } - } - - return s->r_acc; -} - - reg_t kStrEnd(EngineState *s, int argc, reg_t *argv) { reg_t address = argv[0]; address.offset += s->_segMan->strlen(address); diff --git a/engines/sci/engine/said.cpp b/engines/sci/engine/said.cpp deleted file mode 100644 index 32bd1fcf8d..0000000000 --- a/engines/sci/engine/said.cpp +++ /dev/null @@ -1,2483 +0,0 @@ -/* A Bison parser, made by GNU Bison 2.3. */ - -/* Skeleton implementation for Bison's Yacc-like parsers in C - - Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003, 2004, 2005, 2006 - Free Software Foundation, Inc. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 51 Franklin Street, Fifth Floor, - Boston, MA 02110-1301, USA. */ - -/* As a special exception, you may create a larger work that contains - part or all of the Bison parser skeleton and distribute that work - under terms of your choice, so long as that work isn't itself a - parser generator using the skeleton or a modified version thereof - as a parser skeleton. Alternatively, if you modify or redistribute - the parser skeleton itself, you may (at your option) remove this - special exception, which will cause the skeleton and the resulting - Bison output files to be licensed under the GNU General Public - License without this special exception. - - This special exception was added by the Free Software Foundation in - version 2.2 of Bison. */ - -/* C LALR(1) parser skeleton written by Richard Stallman, by - simplifying the original so-called "semantic" parser. */ - -/* All symbols defined below should begin with yy or YY, to avoid - infringing on user name space. This should be done even for local - variables, as they might otherwise be expanded by user macros. - There are some unavoidable exceptions within include files to - define necessary library symbols; they are noted "INFRINGES ON - USER NAME SPACE" below. */ - -/* Identify Bison output. */ -#define YYBISON 1 - -/* Bison version. */ -#define YYBISON_VERSION "2.3" - -/* Skeleton name. */ -#define YYSKELETON_NAME "yacc.c" - -/* Pure parsers. */ -#define YYPURE 0 - -/* Using locations. */ -#define YYLSP_NEEDED 0 - - - -/* Tokens. */ -#ifndef YYTOKENTYPE -# define YYTOKENTYPE - /* Put the tokens into the symbol table, so that GDB and other debuggers - know about them. */ - enum yytokentype { - WGROUP = 258, - YY_COMMA = 259, - YY_AMP = 260, - YY_SLASH = 261, - YY_PARENO = 262, - YY_PARENC = 263, - YY_BRACKETSO = 264, - YY_BRACKETSC = 265, - YY_HASH = 266, - YY_LT = 267, - YY_GT = 268, - YY_BRACKETSO_LT = 269, - YY_BRACKETSO_SLASH = 270, - YY_LT_BRACKETSO = 271, - YY_LT_PARENO = 272 - }; -#endif -/* Tokens. */ -#define WGROUP 258 -#define YY_COMMA 259 -#define YY_AMP 260 -#define YY_SLASH 261 -#define YY_PARENO 262 -#define YY_PARENC 263 -#define YY_BRACKETSO 264 -#define YY_BRACKETSC 265 -#define YY_HASH 266 -#define YY_LT 267 -#define YY_GT 268 -#define YY_BRACKETSO_LT 269 -#define YY_BRACKETSO_SLASH 270 -#define YY_LT_BRACKETSO 271 -#define YY_LT_PARENO 272 - - - - -/* Copy the first part of user declarations. */ -#line 1 "said.y" - -/* ScummVM - Graphic Adventure Engine - * - * ScummVM is the legal property of its developers, whose names - * are too numerous to list here. Please refer to the COPYRIGHT - * file distributed with this source distribution. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - * - * $URL$ - * $Id$ - * - */ - -#include "sci/engine/state.h" - - -// Bison generates an empty switch statement that gives a warning in MSVC. -// This disables that warning. -#ifdef _MSC_VER -#pragma warning(disable:4065) -#endif - - -namespace Sci { - -#define SAID_BRANCH_NULL 0 - -#define MAX_SAID_TOKENS 128 - -// Maximum number of words to be expected in a parsed sentence -#define AUGMENT_MAX_WORDS 64 - - -#define ANYWORD 0xfff - -#define WORD_TYPE_BASE 0x141 -#define WORD_TYPE_REF 0x144 -#define WORD_TYPE_SYNTACTIC_SUGAR 0x145 - -#define AUGMENT_SENTENCE_PART_BRACKETS 0x152 - -// Minor numbers -#define AUGMENT_SENTENCE_MINOR_MATCH_PHRASE 0x14c -#define AUGMENT_SENTENCE_MINOR_MATCH_WORD 0x153 -#define AUGMENT_SENTENCE_MINOR_RECURSE 0x144 -#define AUGMENT_SENTENCE_MINOR_PARENTHESES 0x14f - - -#undef YYDEBUG /*1*/ -//#define SAID_DEBUG*/ -//#define SCI_DEBUG_PARSE_TREE_AUGMENTATION // uncomment to debug parse tree augmentation - - -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION -#define scidprintf printf -#else -void print_nothing(...) { } -#define scidprintf print_nothing -#endif - - -static char *said_parse_error; - -static int said_token; -static int said_tokens_nr; -static int said_tokens[MAX_SAID_TOKENS]; -static int said_blessed; // increminated by said_top_branch - -static int said_tree_pos; // Set to 0 if we're out of space -#define SAID_TREE_START 4; // Reserve space for the 4 top nodes - -#define VALUE_IGNORE -424242 - -static parse_tree_node_t said_tree[VOCAB_TREE_NODES]; - -typedef int wgroup_t; -typedef int tree_t; -typedef int said_spec_t; - -static tree_t said_aug_branch(int, int, tree_t, tree_t); -static tree_t said_attach_branch(tree_t, tree_t); -/* -static tree_t said_wgroup_branch(wgroup_t); -*/ -static said_spec_t said_top_branch(tree_t); -static tree_t said_paren(tree_t, tree_t); -static tree_t said_value(int, tree_t); -static tree_t said_terminal(int); - -static int yylex(void); - -static int yyerror(const char *s) { - said_parse_error = strdup(s); - return 1; /* Abort */ -} - - - -/* Enabling traces. */ -#ifndef YYDEBUG -# define YYDEBUG 0 -#endif - -/* Enabling verbose error messages. */ -#ifdef YYERROR_VERBOSE -# undef YYERROR_VERBOSE -# define YYERROR_VERBOSE 1 -#else -# define YYERROR_VERBOSE 0 -#endif - -/* Enabling the token table. */ -#ifndef YYTOKEN_TABLE -# define YYTOKEN_TABLE 0 -#endif - -#if ! defined YYSTYPE && ! defined YYSTYPE_IS_DECLARED -typedef int YYSTYPE; -# define yystype YYSTYPE /* obsolescent; will be withdrawn */ -# define YYSTYPE_IS_DECLARED 1 -# define YYSTYPE_IS_TRIVIAL 1 -#endif - - - -/* Copy the second part of user declarations. */ - - -/* Line 216 of yacc.c. */ -#line 250 "said.tab.c" - -#ifdef short -# undef short -#endif - -#ifdef YYTYPE_UINT8 -typedef YYTYPE_UINT8 yytype_uint8; -#else -typedef unsigned char yytype_uint8; -#endif - -#ifdef YYTYPE_INT8 -typedef YYTYPE_INT8 yytype_int8; -#elif (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -typedef signed char yytype_int8; -#else -typedef short int yytype_int8; -#endif - -#ifdef YYTYPE_UINT16 -typedef YYTYPE_UINT16 yytype_uint16; -#else -typedef unsigned short int yytype_uint16; -#endif - -#ifdef YYTYPE_INT16 -typedef YYTYPE_INT16 yytype_int16; -#else -typedef short int yytype_int16; -#endif - -#ifndef YYSIZE_T -# ifdef __SIZE_TYPE__ -# define YYSIZE_T __SIZE_TYPE__ -# elif defined size_t -# define YYSIZE_T size_t -# elif ! defined YYSIZE_T && (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -# include /* INFRINGES ON USER NAME SPACE */ -# define YYSIZE_T size_t -# else -# define YYSIZE_T unsigned int -# endif -#endif - -#define YYSIZE_MAXIMUM ((YYSIZE_T) -1) - -#ifndef YY_ -# if YYENABLE_NLS -# if ENABLE_NLS -# include /* INFRINGES ON USER NAME SPACE */ -# define YY_(msgid) dgettext ("bison-runtime", msgid) -# endif -# endif -# ifndef YY_ -# define YY_(msgid) msgid -# endif -#endif - -/* Suppress unused-variable warnings by "using" E. */ -#if ! defined lint || defined __GNUC__ -# define YYUSE(e) ((void) (e)) -#else -# define YYUSE(e) /* empty */ -#endif - -/* Identity function, used to suppress warnings about constant conditions. */ -#ifndef lint -# define YYID(n) (n) -#else -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static int -YYID (int i) -#else -static int -YYID (i) - int i; -#endif -{ - return i; -} -#endif - -#if ! defined yyoverflow || YYERROR_VERBOSE - -/* The parser invokes alloca or malloc; define the necessary symbols. */ - -# ifdef YYSTACK_USE_ALLOCA -# if YYSTACK_USE_ALLOCA -# ifdef __GNUC__ -# define YYSTACK_ALLOC __builtin_alloca -# elif defined __BUILTIN_VA_ARG_INCR -# include /* INFRINGES ON USER NAME SPACE */ -# elif defined _AIX -# define YYSTACK_ALLOC __alloca -# elif defined _MSC_VER -# include /* INFRINGES ON USER NAME SPACE */ -# define alloca _alloca -# else -# define YYSTACK_ALLOC alloca -# if ! defined _ALLOCA_H && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -# include /* INFRINGES ON USER NAME SPACE */ -# ifndef _STDLIB_H -# define _STDLIB_H 1 -# endif -# endif -# endif -# endif -# endif - -# ifdef YYSTACK_ALLOC - /* Pacify GCC's `empty if-body' warning. */ -# define YYSTACK_FREE(Ptr) do { /* empty */; } while (YYID (0)) -# ifndef YYSTACK_ALLOC_MAXIMUM - /* The OS might guarantee only one guard page at the bottom of the stack, - and a page size can be as small as 4096 bytes. So we cannot safely - invoke alloca (N) if N exceeds 4096. Use a slightly smaller number - to allow for a few compiler-allocated temporary stack slots. */ -# define YYSTACK_ALLOC_MAXIMUM 4032 /* reasonable circa 2006 */ -# endif -# else -# define YYSTACK_ALLOC YYMALLOC -# define YYSTACK_FREE YYFREE -# ifndef YYSTACK_ALLOC_MAXIMUM -# define YYSTACK_ALLOC_MAXIMUM YYSIZE_MAXIMUM -# endif -# if (defined __cplusplus && ! defined _STDLIB_H \ - && ! ((defined YYMALLOC || defined malloc) \ - && (defined YYFREE || defined free))) -# include /* INFRINGES ON USER NAME SPACE */ -# ifndef _STDLIB_H -# define _STDLIB_H 1 -# endif -# endif -# ifndef YYMALLOC -# define YYMALLOC malloc -# if ! defined malloc && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -void *malloc (YYSIZE_T); /* INFRINGES ON USER NAME SPACE */ -# endif -# endif -# ifndef YYFREE -# define YYFREE free -# if ! defined free && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -void free (void *); /* INFRINGES ON USER NAME SPACE */ -# endif -# endif -# endif -#endif /* ! defined yyoverflow || YYERROR_VERBOSE */ - - -#if (! defined yyoverflow \ - && (! defined __cplusplus \ - || (defined YYSTYPE_IS_TRIVIAL && YYSTYPE_IS_TRIVIAL))) - -/* A type that is properly aligned for any stack member. */ -union yyalloc -{ - yytype_int16 yyss; - YYSTYPE yyvs; - }; - -/* The size of the maximum gap between one aligned stack and the next. */ -# define YYSTACK_GAP_MAXIMUM (sizeof (union yyalloc) - 1) - -/* The size of an array large to enough to hold all stacks, each with - N elements. */ -# define YYSTACK_BYTES(N) \ - ((N) * (sizeof (yytype_int16) + sizeof (YYSTYPE)) \ - + YYSTACK_GAP_MAXIMUM) - -/* Copy COUNT objects from FROM to TO. The source and destination do - not overlap. */ -# ifndef YYCOPY -# if defined __GNUC__ && 1 < __GNUC__ -# define YYCOPY(To, From, Count) \ - __builtin_memcpy (To, From, (Count) * sizeof (*(From))) -# else -# define YYCOPY(To, From, Count) \ - do \ - { \ - YYSIZE_T yyi; \ - for (yyi = 0; yyi < (Count); yyi++) \ - (To)[yyi] = (From)[yyi]; \ - } \ - while (YYID (0)) -# endif -# endif - -/* Relocate STACK from its old location to the new one. The - local variables YYSIZE and YYSTACKSIZE give the old and new number of - elements in the stack, and YYPTR gives the new location of the - stack. Advance YYPTR to a properly aligned location for the next - stack. */ -# define YYSTACK_RELOCATE(Stack) \ - do \ - { \ - YYSIZE_T yynewbytes; \ - YYCOPY (&yyptr->Stack, Stack, yysize); \ - Stack = &yyptr->Stack; \ - yynewbytes = yystacksize * sizeof (*Stack) + YYSTACK_GAP_MAXIMUM; \ - yyptr += yynewbytes / sizeof (*yyptr); \ - } \ - while (YYID (0)) - -#endif - -/* YYFINAL -- State number of the termination state. */ -#define YYFINAL 23 -/* YYLAST -- Last index in YYTABLE. */ -#define YYLAST 80 - -/* YYNTOKENS -- Number of terminals. */ -#define YYNTOKENS 18 -/* YYNNTS -- Number of nonterminals. */ -#define YYNNTS 13 -/* YYNRULES -- Number of rules. */ -#define YYNRULES 35 -/* YYNRULES -- Number of states. */ -#define YYNSTATES 69 - -/* YYTRANSLATE(YYLEX) -- Bison symbol number corresponding to YYLEX. */ -#define YYUNDEFTOK 2 -#define YYMAXUTOK 272 - -#define YYTRANSLATE(YYX) \ - ((unsigned int) (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYUNDEFTOK) - -/* YYTRANSLATE[YYLEX] -- Bison symbol number corresponding to YYLEX. */ -static const yytype_uint8 yytranslate[] = -{ - 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 2, 1, 2, 3, 4, - 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, - 15, 16, 17 -}; - -#if YYDEBUG -/* YYPRHS[YYN] -- Index of the first RHS symbol of rule number YYN in - YYRHS. */ -static const yytype_uint8 yyprhs[] = -{ - 0, 0, 3, 6, 10, 15, 16, 18, 19, 21, - 24, 29, 31, 34, 39, 41, 43, 45, 49, 51, - 55, 59, 64, 70, 73, 75, 77, 79, 83, 88, - 92, 97, 100, 105, 109, 112 -}; - -/* YYRHS -- A `-1'-separated list of the rules' RHS. */ -static const yytype_int8 yyrhs[] = -{ - 19, 0, -1, 21, 20, -1, 21, 22, 20, -1, - 21, 22, 23, 20, -1, -1, 13, -1, -1, 27, - -1, 6, 27, -1, 15, 6, 27, 10, -1, 6, - -1, 6, 27, -1, 15, 6, 27, 10, -1, 6, - -1, 3, -1, 26, -1, 9, 26, 10, -1, 24, - -1, 7, 27, 8, -1, 26, 4, 26, -1, 26, - 14, 29, 10, -1, 26, 4, 9, 26, 10, -1, - 25, 28, -1, 25, -1, 28, -1, 29, -1, 14, - 29, 10, -1, 29, 14, 29, 10, -1, 12, 24, - 30, -1, 17, 7, 27, 8, -1, 12, 26, -1, - 16, 9, 26, 10, -1, 12, 26, 30, -1, 12, - 26, -1, 17, 7, 27, 8, -1 -}; - -/* YYRLINE[YYN] -- source line where rule number YYN was defined. */ -static const yytype_uint8 yyrline[] = -{ - 0, 129, 129, 131, 133, 139, 140, 147, 148, 154, - 156, 158, 164, 166, 168, 173, 178, 180, 185, 187, - 189, 191, 193, 198, 200, 202, 207, 209, 211, 216, - 218, 220, 222, 227, 229, 231 -}; -#endif - -#if YYDEBUG || YYERROR_VERBOSE || YYTOKEN_TABLE -/* YYTNAME[SYMBOL-NUM] -- String name of the symbol SYMBOL-NUM. - First, the terminals, then, starting at YYNTOKENS, nonterminals. */ -static const char *const yytname[] = -{ - "$end", "error", "$undefined", "WGROUP", "YY_COMMA", "YY_AMP", - "YY_SLASH", "YY_PARENO", "YY_PARENC", "YY_BRACKETSO", "YY_BRACKETSC", - "YY_HASH", "YY_LT", "YY_GT", "YY_BRACKETSO_LT", "YY_BRACKETSO_SLASH", - "YY_LT_BRACKETSO", "YY_LT_PARENO", "$accept", "saidspec", "optcont", - "leftspec", "midspec", "rightspec", "word", "cwordset", "wordset", - "expr", "cwordrefset", "wordrefset", "recref", 0 -}; -#endif - -# ifdef YYPRINT -/* YYTOKNUM[YYLEX-NUM] -- Internal token number corresponding to - token YYLEX-NUM. */ -static const yytype_uint16 yytoknum[] = -{ - 0, 256, 257, 258, 259, 260, 261, 262, 263, 264, - 265, 266, 267, 268, 269, 270, 271, 272 -}; -# endif - -/* YYR1[YYN] -- Symbol number of symbol that rule YYN derives. */ -static const yytype_uint8 yyr1[] = -{ - 0, 18, 19, 19, 19, 20, 20, 21, 21, 22, - 22, 22, 23, 23, 23, 24, 25, 25, 26, 26, - 26, 26, 26, 27, 27, 27, 28, 28, 28, 29, - 29, 29, 29, 30, 30, 30 -}; - -/* YYR2[YYN] -- Number of symbols composing right hand side of rule YYN. */ -static const yytype_uint8 yyr2[] = -{ - 0, 2, 2, 3, 4, 0, 1, 0, 1, 2, - 4, 1, 2, 4, 1, 1, 1, 3, 1, 3, - 3, 4, 5, 2, 1, 1, 1, 3, 4, 3, - 4, 2, 4, 3, 2, 4 -}; - -/* YYDEFACT[STATE-NAME] -- Default rule to reduce with in state - STATE-NUM when YYTABLE doesn't specify something else to do. Zero - means the default is an error. */ -static const yytype_uint8 yydefact[] = -{ - 7, 15, 0, 0, 0, 0, 0, 0, 0, 5, - 18, 24, 16, 8, 25, 26, 0, 0, 18, 31, - 0, 0, 0, 1, 11, 6, 0, 2, 5, 23, - 0, 0, 0, 19, 17, 0, 0, 29, 27, 0, - 0, 9, 0, 14, 0, 3, 5, 0, 20, 0, - 0, 34, 0, 32, 30, 0, 12, 0, 4, 0, - 21, 28, 33, 0, 10, 0, 22, 35, 13 -}; - -/* YYDEFGOTO[NTERM-NUM]. */ -static const yytype_int8 yydefgoto[] = -{ - -1, 8, 27, 9, 28, 46, 10, 11, 12, 13, - 14, 15, 37 -}; - -/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing - STATE-NUM. */ -#define YYPACT_NINF -24 -static const yytype_int8 yypact[] = -{ - -1, -24, -1, 62, 62, 54, 1, 5, 18, 38, - -24, 47, 3, -24, -24, 12, 23, 15, -3, 3, - 28, 62, -1, -24, -1, -24, 42, -24, 39, -24, - 53, 54, 54, -24, -24, 62, 50, -24, -24, 29, - 41, -24, -1, -1, 52, -24, 55, 62, 3, 57, - 63, 20, -1, -24, -24, 64, -24, -1, -24, 32, - -24, -24, -24, 67, -24, 66, -24, -24, -24 -}; - -/* YYPGOTO[NTERM-NUM]. */ -static const yytype_int8 yypgoto[] = -{ - -24, -24, -23, -24, -24, -24, 68, -24, 0, -2, - 69, -4, 26 -}; - -/* YYTABLE[YYPACT[STATE-NUM]]. What to do in state STATE-NUM. If - positive, shift that token. If negative, reduce the rule which - number is the opposite. If zero, do what YYDEFACT says. - If YYTABLE_NINF, syntax error. */ -#define YYTABLE_NINF -1 -static const yytype_uint8 yytable[] = -{ - 16, 20, 1, 17, 19, 45, 2, 30, 3, 35, - 21, 4, 22, 5, 36, 6, 7, 31, 23, 30, - 40, 39, 41, 58, 30, 34, 32, 49, 50, 31, - 48, 33, 35, 30, 31, 51, 30, 36, 38, 53, - 55, 56, 66, 31, 24, 43, 31, 59, 42, 54, - 63, 25, 25, 26, 44, 65, 1, 52, 57, 4, - 2, 5, 47, 6, 7, 1, 4, 60, 25, 2, - 6, 7, 18, 61, 64, 67, 68, 62, 0, 0, - 29 -}; - -static const yytype_int8 yycheck[] = -{ - 2, 5, 3, 3, 4, 28, 7, 4, 9, 12, - 9, 12, 7, 14, 17, 16, 17, 14, 0, 4, - 22, 21, 24, 46, 4, 10, 14, 31, 32, 14, - 30, 8, 12, 4, 14, 35, 4, 17, 10, 10, - 42, 43, 10, 14, 6, 6, 14, 47, 6, 8, - 52, 13, 13, 15, 15, 57, 3, 7, 6, 12, - 7, 14, 9, 16, 17, 3, 12, 10, 13, 7, - 16, 17, 4, 10, 10, 8, 10, 51, -1, -1, - 11 -}; - -/* YYSTOS[STATE-NUM] -- The (internal number of the) accessing - symbol of state STATE-NUM. */ -static const yytype_uint8 yystos[] = -{ - 0, 3, 7, 9, 12, 14, 16, 17, 19, 21, - 24, 25, 26, 27, 28, 29, 27, 26, 24, 26, - 29, 9, 7, 0, 6, 13, 15, 20, 22, 28, - 4, 14, 14, 8, 10, 12, 17, 30, 10, 26, - 27, 27, 6, 6, 15, 20, 23, 9, 26, 29, - 29, 26, 7, 10, 8, 27, 27, 6, 20, 26, - 10, 10, 30, 27, 10, 27, 10, 8, 10 -}; - -#define yyerrok (yyerrstatus = 0) -#define yyclearin (yychar = YYEMPTY) -#define YYEMPTY (-2) -#define YYEOF 0 - -#define YYACCEPT goto yyacceptlab -#define YYABORT goto yyabortlab -#define YYERROR goto yyerrorlab - - -/* Like YYERROR except do call yyerror. This remains here temporarily - to ease the transition to the new meaning of YYERROR, for GCC. - Once GCC version 2 has supplanted version 1, this can go. */ - -#define YYFAIL goto yyerrlab - -#define YYRECOVERING() (!!yyerrstatus) - -#define YYBACKUP(Token, Value) \ -do \ - if (yychar == YYEMPTY && yylen == 1) \ - { \ - yychar = (Token); \ - yylval = (Value); \ - yytoken = YYTRANSLATE (yychar); \ - YYPOPSTACK (1); \ - goto yybackup; \ - } \ - else \ - { \ - yyerror (YY_("syntax error: cannot back up")); \ - YYERROR; \ - } \ -while (YYID (0)) - - -#define YYTERROR 1 -#define YYERRCODE 256 - - -/* YYLLOC_DEFAULT -- Set CURRENT to span from RHS[1] to RHS[N]. - If N is 0, then set CURRENT to the empty location which ends - the previous symbol: RHS[0] (always defined). */ - -#define YYRHSLOC(Rhs, K) ((Rhs)[K]) -#ifndef YYLLOC_DEFAULT -# define YYLLOC_DEFAULT(Current, Rhs, N) \ - do \ - if (YYID (N)) \ - { \ - (Current).first_line = YYRHSLOC (Rhs, 1).first_line; \ - (Current).first_column = YYRHSLOC (Rhs, 1).first_column; \ - (Current).last_line = YYRHSLOC (Rhs, N).last_line; \ - (Current).last_column = YYRHSLOC (Rhs, N).last_column; \ - } \ - else \ - { \ - (Current).first_line = (Current).last_line = \ - YYRHSLOC (Rhs, 0).last_line; \ - (Current).first_column = (Current).last_column = \ - YYRHSLOC (Rhs, 0).last_column; \ - } \ - while (YYID (0)) -#endif - - -/* YY_LOCATION_PRINT -- Print the location on the stream. - This macro was not mandated originally: define only if we know - we won't break user code: when these are the locations we know. */ - -#ifndef YY_LOCATION_PRINT -# if YYLTYPE_IS_TRIVIAL -# define YY_LOCATION_PRINT(File, Loc) \ - fprintf (File, "%d.%d-%d.%d", \ - (Loc).first_line, (Loc).first_column, \ - (Loc).last_line, (Loc).last_column) -# else -# define YY_LOCATION_PRINT(File, Loc) ((void) 0) -# endif -#endif - - -/* YYLEX -- calling `yylex' with the right arguments. */ - -#ifdef YYLEX_PARAM -# define YYLEX yylex (YYLEX_PARAM) -#else -# define YYLEX yylex () -#endif - -/* Enable debugging if requested. */ -#if YYDEBUG - -# ifndef YYFPRINTF -# include /* INFRINGES ON USER NAME SPACE */ -# define YYFPRINTF fprintf -# endif - -# define YYDPRINTF(Args) \ -do { \ - if (yydebug) \ - YYFPRINTF Args; \ -} while (YYID (0)) - -# define YY_SYMBOL_PRINT(Title, Type, Value, Location) \ -do { \ - if (yydebug) \ - { \ - YYFPRINTF (stderr, "%s ", Title); \ - yy_symbol_print (stderr, \ - Type, Value); \ - YYFPRINTF (stderr, "\n"); \ - } \ -} while (YYID (0)) - - -/*--------------------------------. -| Print this symbol on YYOUTPUT. | -`--------------------------------*/ - -/*ARGSUSED*/ -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static void -yy_symbol_value_print (FILE *yyoutput, int yytype, YYSTYPE const * const yyvaluep) -#else -static void -yy_symbol_value_print (yyoutput, yytype, yyvaluep) - FILE *yyoutput; - int yytype; - YYSTYPE const * const yyvaluep; -#endif -{ - if (!yyvaluep) - return; -# ifdef YYPRINT - if (yytype < YYNTOKENS) - YYPRINT (yyoutput, yytoknum[yytype], *yyvaluep); -# else - YYUSE (yyoutput); -# endif - switch (yytype) - { - default: - break; - } -} - - -/*--------------------------------. -| Print this symbol on YYOUTPUT. | -`--------------------------------*/ - -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static void -yy_symbol_print (FILE *yyoutput, int yytype, YYSTYPE const * const yyvaluep) -#else -static void -yy_symbol_print (yyoutput, yytype, yyvaluep) - FILE *yyoutput; - int yytype; - YYSTYPE const * const yyvaluep; -#endif -{ - if (yytype < YYNTOKENS) - YYFPRINTF (yyoutput, "token %s (", yytname[yytype]); - else - YYFPRINTF (yyoutput, "nterm %s (", yytname[yytype]); - - yy_symbol_value_print (yyoutput, yytype, yyvaluep); - YYFPRINTF (yyoutput, ")"); -} - -/*------------------------------------------------------------------. -| yy_stack_print -- Print the state stack from its BOTTOM up to its | -| TOP (included). | -`------------------------------------------------------------------*/ - -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static void -yy_stack_print (yytype_int16 *bottom, yytype_int16 *top) -#else -static void -yy_stack_print (bottom, top) - yytype_int16 *bottom; - yytype_int16 *top; -#endif -{ - YYFPRINTF (stderr, "Stack now"); - for (; bottom <= top; ++bottom) - YYFPRINTF (stderr, " %d", *bottom); - YYFPRINTF (stderr, "\n"); -} - -# define YY_STACK_PRINT(Bottom, Top) \ -do { \ - if (yydebug) \ - yy_stack_print ((Bottom), (Top)); \ -} while (YYID (0)) - - -/*------------------------------------------------. -| Report that the YYRULE is going to be reduced. | -`------------------------------------------------*/ - -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static void -yy_reduce_print (YYSTYPE *yyvsp, int yyrule) -#else -static void -yy_reduce_print (yyvsp, yyrule) - YYSTYPE *yyvsp; - int yyrule; -#endif -{ - int yynrhs = yyr2[yyrule]; - int yyi; - unsigned long int yylno = yyrline[yyrule]; - YYFPRINTF (stderr, "Reducing stack by rule %d (line %lu):\n", - yyrule - 1, yylno); - /* The symbols being reduced. */ - for (yyi = 0; yyi < yynrhs; yyi++) - { - fprintf (stderr, " $%d = ", yyi + 1); - yy_symbol_print (stderr, yyrhs[yyprhs[yyrule] + yyi], - &(yyvsp[(yyi + 1) - (yynrhs)]) - ); - fprintf (stderr, "\n"); - } -} - -# define YY_REDUCE_PRINT(Rule) \ -do { \ - if (yydebug) \ - yy_reduce_print (yyvsp, Rule); \ -} while (YYID (0)) - -/* Nonzero means print parse trace. It is left uninitialized so that - multiple parsers can coexist. */ -int yydebug; -#else /* !YYDEBUG */ -# define YYDPRINTF(Args) -# define YY_SYMBOL_PRINT(Title, Type, Value, Location) -# define YY_STACK_PRINT(Bottom, Top) -# define YY_REDUCE_PRINT(Rule) -#endif /* !YYDEBUG */ - - -/* YYINITDEPTH -- initial size of the parser's stacks. */ -#ifndef YYINITDEPTH -# define YYINITDEPTH 200 -#endif - -/* YYMAXDEPTH -- maximum size the stacks can grow to (effective only - if the built-in stack extension method is used). - - Do not make this value too large; the results are undefined if - YYSTACK_ALLOC_MAXIMUM < YYSTACK_BYTES (YYMAXDEPTH) - evaluated with infinite-precision integer arithmetic. */ - -#ifndef YYMAXDEPTH -# define YYMAXDEPTH 10000 -#endif - - - -#if YYERROR_VERBOSE - -# ifndef yystrlen -# if defined __GLIBC__ && defined _STRING_H -# define yystrlen strlen -# else -/* Return the length of YYSTR. */ -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static YYSIZE_T -yystrlen (const char *yystr) -#else -static YYSIZE_T -yystrlen (yystr) - const char *yystr; -#endif -{ - YYSIZE_T yylen; - for (yylen = 0; yystr[yylen]; yylen++) - continue; - return yylen; -} -# endif -# endif - -# ifndef yystpcpy -# if defined __GLIBC__ && defined _STRING_H && defined _GNU_SOURCE -# define yystpcpy stpcpy -# else -/* Copy YYSRC to YYDEST, returning the address of the terminating '\0' in - YYDEST. */ -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static char * -yystpcpy (char *yydest, const char *yysrc) -#else -static char * -yystpcpy (yydest, yysrc) - char *yydest; - const char *yysrc; -#endif -{ - char *yyd = yydest; - const char *yys = yysrc; - - while ((*yyd++ = *yys++) != '\0') - continue; - - return yyd - 1; -} -# endif -# endif - -# ifndef yytnamerr -/* Copy to YYRES the contents of YYSTR after stripping away unnecessary - quotes and backslashes, so that it's suitable for yyerror. The - heuristic is that double-quoting is unnecessary unless the string - contains an apostrophe, a comma, or backslash (other than - backslash-backslash). YYSTR is taken from yytname. If YYRES is - null, do not copy; instead, return the length of what the result - would have been. */ -static YYSIZE_T -yytnamerr (char *yyres, const char *yystr) -{ - if (*yystr == '"') - { - YYSIZE_T yyn = 0; - char const *yyp = yystr; - - for (;;) - switch (*++yyp) - { - case '\'': - case ',': - goto do_not_strip_quotes; - - case '\\': - if (*++yyp != '\\') - goto do_not_strip_quotes; - /* Fall through. */ - default: - if (yyres) - yyres[yyn] = *yyp; - yyn++; - break; - - case '"': - if (yyres) - yyres[yyn] = '\0'; - return yyn; - } - do_not_strip_quotes: ; - } - - if (! yyres) - return yystrlen (yystr); - - return yystpcpy (yyres, yystr) - yyres; -} -# endif - -/* Copy into YYRESULT an error message about the unexpected token - YYCHAR while in state YYSTATE. Return the number of bytes copied, - including the terminating null byte. If YYRESULT is null, do not - copy anything; just return the number of bytes that would be - copied. As a special case, return 0 if an ordinary "syntax error" - message will do. Return YYSIZE_MAXIMUM if overflow occurs during - size calculation. */ -static YYSIZE_T -yysyntax_error (char *yyresult, int yystate, int yychar) -{ - int yyn = yypact[yystate]; - - if (! (YYPACT_NINF < yyn && yyn <= YYLAST)) - return 0; - else - { - int yytype = YYTRANSLATE (yychar); - YYSIZE_T yysize0 = yytnamerr (0, yytname[yytype]); - YYSIZE_T yysize = yysize0; - YYSIZE_T yysize1; - int yysize_overflow = 0; - enum { YYERROR_VERBOSE_ARGS_MAXIMUM = 5 }; - char const *yyarg[YYERROR_VERBOSE_ARGS_MAXIMUM]; - int yyx; - -# if 0 - /* This is so xgettext sees the translatable formats that are - constructed on the fly. */ - YY_("syntax error, unexpected %s"); - YY_("syntax error, unexpected %s, expecting %s"); - YY_("syntax error, unexpected %s, expecting %s or %s"); - YY_("syntax error, unexpected %s, expecting %s or %s or %s"); - YY_("syntax error, unexpected %s, expecting %s or %s or %s or %s"); -# endif - char *yyfmt; - char const *yyf; - static char const yyunexpected[] = "syntax error, unexpected %s"; - static char const yyexpecting[] = ", expecting %s"; - static char const yyor[] = " or %s"; - char yyformat[sizeof yyunexpected - + sizeof yyexpecting - 1 - + ((YYERROR_VERBOSE_ARGS_MAXIMUM - 2) - * (sizeof yyor - 1))]; - char const *yyprefix = yyexpecting; - - /* Start YYX at -YYN if negative to avoid negative indexes in - YYCHECK. */ - int yyxbegin = yyn < 0 ? -yyn : 0; - - /* Stay within bounds of both yycheck and yytname. */ - int yychecklim = YYLAST - yyn + 1; - int yyxend = yychecklim < YYNTOKENS ? yychecklim : YYNTOKENS; - int yycount = 1; - - yyarg[0] = yytname[yytype]; - yyfmt = yystpcpy (yyformat, yyunexpected); - - for (yyx = yyxbegin; yyx < yyxend; ++yyx) - if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR) - { - if (yycount == YYERROR_VERBOSE_ARGS_MAXIMUM) - { - yycount = 1; - yysize = yysize0; - yyformat[sizeof yyunexpected - 1] = '\0'; - break; - } - yyarg[yycount++] = yytname[yyx]; - yysize1 = yysize + yytnamerr (0, yytname[yyx]); - yysize_overflow |= (yysize1 < yysize); - yysize = yysize1; - yyfmt = yystpcpy (yyfmt, yyprefix); - yyprefix = yyor; - } - - yyf = YY_(yyformat); - yysize1 = yysize + yystrlen (yyf); - yysize_overflow |= (yysize1 < yysize); - yysize = yysize1; - - if (yysize_overflow) - return YYSIZE_MAXIMUM; - - if (yyresult) - { - /* Avoid sprintf, as that infringes on the user's name space. - Don't have undefined behavior even if the translation - produced a string with the wrong number of "%s"s. */ - char *yyp = yyresult; - int yyi = 0; - while ((*yyp = *yyf) != '\0') - { - if (*yyp == '%' && yyf[1] == 's' && yyi < yycount) - { - yyp += yytnamerr (yyp, yyarg[yyi++]); - yyf += 2; - } - else - { - yyp++; - yyf++; - } - } - } - return yysize; - } -} -#endif /* YYERROR_VERBOSE */ - - -/*-----------------------------------------------. -| Release the memory associated to this symbol. | -`-----------------------------------------------*/ - -/*ARGSUSED*/ -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -static void -yydestruct (const char *yymsg, int yytype, YYSTYPE *yyvaluep) -#else -static void -yydestruct (yymsg, yytype, yyvaluep) - const char *yymsg; - int yytype; - YYSTYPE *yyvaluep; -#endif -{ - YYUSE (yyvaluep); - - if (!yymsg) - yymsg = "Deleting"; - YY_SYMBOL_PRINT (yymsg, yytype, yyvaluep, yylocationp); - - switch (yytype) - { - - default: - break; - } -} - - -/* Prevent warnings from -Wmissing-prototypes. */ - -#ifdef YYPARSE_PARAM -#if defined __STDC__ || defined __cplusplus -int yyparse (void *YYPARSE_PARAM); -#else -int yyparse (); -#endif -#else /* ! YYPARSE_PARAM */ -#if defined __STDC__ || defined __cplusplus -int yyparse (void); -#else -int yyparse (); -#endif -#endif /* ! YYPARSE_PARAM */ - - - -/* The look-ahead symbol. */ -int yychar; - -/* The semantic value of the look-ahead symbol. */ -YYSTYPE yylval; - -/* Number of syntax errors so far. */ -int yynerrs; - - - -/*----------. -| yyparse. | -`----------*/ - -#ifdef YYPARSE_PARAM -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -int -yyparse (void *YYPARSE_PARAM) -#else -int -yyparse (YYPARSE_PARAM) - void *YYPARSE_PARAM; -#endif -#else /* ! YYPARSE_PARAM */ -#if (defined __STDC__ || defined __C99__FUNC__ \ - || defined __cplusplus || defined _MSC_VER) -int -yyparse (void) -#else -int -yyparse () - -#endif -#endif -{ - - int yystate; - int yyn; - int yyresult; - /* Number of tokens to shift before error messages enabled. */ - int yyerrstatus; - /* Look-ahead token as an internal (translated) token number. */ - int yytoken = 0; -#if YYERROR_VERBOSE - /* Buffer for error messages, and its allocated size. */ - char yymsgbuf[128]; - char *yymsg = yymsgbuf; - YYSIZE_T yymsg_alloc = sizeof yymsgbuf; -#endif - - /* Three stacks and their tools: - `yyss': related to states, - `yyvs': related to semantic values, - `yyls': related to locations. - - Refer to the stacks thru separate pointers, to allow yyoverflow - to reallocate them elsewhere. */ - - /* The state stack. */ - yytype_int16 yyssa[YYINITDEPTH]; - yytype_int16 *yyss = yyssa; - yytype_int16 *yyssp; - - /* The semantic value stack. */ - YYSTYPE yyvsa[YYINITDEPTH]; - YYSTYPE *yyvs = yyvsa; - YYSTYPE *yyvsp; - - - -#define YYPOPSTACK(N) (yyvsp -= (N), yyssp -= (N)) - - YYSIZE_T yystacksize = YYINITDEPTH; - - /* The variables used to return semantic value and location from the - action routines. */ - YYSTYPE yyval; - - - /* The number of symbols on the RHS of the reduced rule. - Keep to zero when no symbol should be popped. */ - int yylen = 0; - - YYDPRINTF ((stderr, "Starting parse\n")); - - yystate = 0; - yyerrstatus = 0; - yynerrs = 0; - yychar = YYEMPTY; /* Cause a token to be read. */ - - /* Initialize stack pointers. - Waste one element of value and location stack - so that they stay on the same level as the state stack. - The wasted elements are never initialized. */ - - yyssp = yyss; - yyvsp = yyvs; - - goto yysetstate; - -/*------------------------------------------------------------. -| yynewstate -- Push a new state, which is found in yystate. | -`------------------------------------------------------------*/ - yynewstate: - /* In all cases, when you get here, the value and location stacks - have just been pushed. So pushing a state here evens the stacks. */ - yyssp++; - - yysetstate: - *yyssp = yystate; - - if (yyss + yystacksize - 1 <= yyssp) - { - /* Get the current used size of the three stacks, in elements. */ - YYSIZE_T yysize = yyssp - yyss + 1; - -#ifdef yyoverflow - { - /* Give user a chance to reallocate the stack. Use copies of - these so that the &'s don't force the real ones into - memory. */ - YYSTYPE *yyvs1 = yyvs; - yytype_int16 *yyss1 = yyss; - - - /* Each stack pointer address is followed by the size of the - data in use in that stack, in bytes. This used to be a - conditional around just the two extra args, but that might - be undefined if yyoverflow is a macro. */ - yyoverflow (YY_("memory exhausted"), - &yyss1, yysize * sizeof (*yyssp), - &yyvs1, yysize * sizeof (*yyvsp), - - &yystacksize); - - yyss = yyss1; - yyvs = yyvs1; - } -#else /* no yyoverflow */ -# ifndef YYSTACK_RELOCATE - goto yyexhaustedlab; -# else - /* Extend the stack our own way. */ - if (YYMAXDEPTH <= yystacksize) - goto yyexhaustedlab; - yystacksize *= 2; - if (YYMAXDEPTH < yystacksize) - yystacksize = YYMAXDEPTH; - - { - yytype_int16 *yyss1 = yyss; - union yyalloc *yyptr = - (union yyalloc *) YYSTACK_ALLOC (YYSTACK_BYTES (yystacksize)); - if (! yyptr) - goto yyexhaustedlab; - YYSTACK_RELOCATE (yyss); - YYSTACK_RELOCATE (yyvs); - -# undef YYSTACK_RELOCATE - if (yyss1 != yyssa) - YYSTACK_FREE (yyss1); - } -# endif -#endif /* no yyoverflow */ - - yyssp = yyss + yysize - 1; - yyvsp = yyvs + yysize - 1; - - - YYDPRINTF ((stderr, "Stack size increased to %lu\n", - (unsigned long int) yystacksize)); - - if (yyss + yystacksize - 1 <= yyssp) - YYABORT; - } - - YYDPRINTF ((stderr, "Entering state %d\n", yystate)); - - goto yybackup; - -/*-----------. -| yybackup. | -`-----------*/ -yybackup: - - /* Do appropriate processing given the current state. Read a - look-ahead token if we need one and don't already have one. */ - - /* First try to decide what to do without reference to look-ahead token. */ - yyn = yypact[yystate]; - if (yyn == YYPACT_NINF) - goto yydefault; - - /* Not known => get a look-ahead token if don't already have one. */ - - /* YYCHAR is either YYEMPTY or YYEOF or a valid look-ahead symbol. */ - if (yychar == YYEMPTY) - { - YYDPRINTF ((stderr, "Reading a token: ")); - yychar = YYLEX; - } - - if (yychar <= YYEOF) - { - yychar = yytoken = YYEOF; - YYDPRINTF ((stderr, "Now at end of input.\n")); - } - else - { - yytoken = YYTRANSLATE (yychar); - YY_SYMBOL_PRINT ("Next token is", yytoken, &yylval, &yylloc); - } - - /* If the proper action on seeing token YYTOKEN is to reduce or to - detect an error, take that action. */ - yyn += yytoken; - if (yyn < 0 || YYLAST < yyn || yycheck[yyn] != yytoken) - goto yydefault; - yyn = yytable[yyn]; - if (yyn <= 0) - { - if (yyn == 0 || yyn == YYTABLE_NINF) - goto yyerrlab; - yyn = -yyn; - goto yyreduce; - } - - if (yyn == YYFINAL) - YYACCEPT; - - /* Count tokens shifted since error; after three, turn off error - status. */ - if (yyerrstatus) - yyerrstatus--; - - /* Shift the look-ahead token. */ - YY_SYMBOL_PRINT ("Shifting", yytoken, &yylval, &yylloc); - - /* Discard the shifted token unless it is eof. */ - if (yychar != YYEOF) - yychar = YYEMPTY; - - yystate = yyn; - *++yyvsp = yylval; - - goto yynewstate; - - -/*-----------------------------------------------------------. -| yydefault -- do the default action for the current state. | -`-----------------------------------------------------------*/ -yydefault: - yyn = yydefact[yystate]; - if (yyn == 0) - goto yyerrlab; - goto yyreduce; - - -/*-----------------------------. -| yyreduce -- Do a reduction. | -`-----------------------------*/ -yyreduce: - /* yyn is the number of a rule to reduce with. */ - yylen = yyr2[yyn]; - - /* If YYLEN is nonzero, implement the default value of the action: - `$$ = $1'. - - Otherwise, the following line sets YYVAL to garbage. - This behavior is undocumented and Bison - users should not rely upon it. Assigning to YYVAL - unconditionally makes the parser a bit smaller, and it avoids a - GCC warning that YYVAL may be used uninitialized. */ - yyval = yyvsp[1-yylen]; - - - YY_REDUCE_PRINT (yyn); - switch (yyn) - { - case 2: -#line 130 "said.y" - { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (2)]), (yyvsp[(2) - (2)]))); ;} - break; - - case 3: -#line 132 "said.y" - { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (3)]), said_attach_branch((yyvsp[(2) - (3)]), (yyvsp[(3) - (3)])))); ;} - break; - - case 4: -#line 134 "said.y" - { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (4)]), said_attach_branch((yyvsp[(2) - (4)]), said_attach_branch((yyvsp[(3) - (4)]), (yyvsp[(4) - (4)]))))); ;} - break; - - case 5: -#line 139 "said.y" - { (yyval) = SAID_BRANCH_NULL; ;} - break; - - case 6: -#line 141 "said.y" - { (yyval) = said_paren(said_value(0x14b, said_value(0xf900, said_terminal(0xf900))), SAID_BRANCH_NULL); ;} - break; - - case 7: -#line 147 "said.y" - { (yyval) = SAID_BRANCH_NULL; ;} - break; - - case 8: -#line 149 "said.y" - { (yyval) = said_paren(said_value(0x141, said_value(0x149, (yyvsp[(1) - (1)]))), SAID_BRANCH_NULL); ;} - break; - - case 9: -#line 155 "said.y" - { (yyval) = said_aug_branch(0x142, 0x14a, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} - break; - - case 10: -#line 157 "said.y" - { (yyval) = said_aug_branch(0x152, 0x142, said_aug_branch(0x142, 0x14a, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 11: -#line 159 "said.y" - { (yyval) = SAID_BRANCH_NULL; ;} - break; - - case 12: -#line 165 "said.y" - { (yyval) = said_aug_branch(0x143, 0x14a, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} - break; - - case 13: -#line 167 "said.y" - { (yyval) = said_aug_branch(0x152, 0x143, said_aug_branch(0x143, 0x14a, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 14: -#line 169 "said.y" - { (yyval) = SAID_BRANCH_NULL; ;} - break; - - case 15: -#line 174 "said.y" - { (yyval) = said_paren(said_value(0x141, said_value(0x153, said_terminal((yyvsp[(1) - (1)])))), SAID_BRANCH_NULL); ;} - break; - - case 16: -#line 179 "said.y" - { (yyval) = said_aug_branch(0x141, 0x14f, (yyvsp[(1) - (1)]), SAID_BRANCH_NULL); ;} - break; - - case 17: -#line 181 "said.y" - { (yyval) = said_aug_branch(0x141, 0x14f, said_aug_branch(0x152, 0x14c, said_aug_branch(0x141, 0x14f, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 18: -#line 186 "said.y" - { (yyval) = (yyvsp[(1) - (1)]); ;} - break; - - case 19: -#line 188 "said.y" - { (yyval) = said_aug_branch(0x141, 0x14c, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL); ;} - break; - - case 20: -#line 190 "said.y" - { (yyval) = said_attach_branch((yyvsp[(1) - (3)]), (yyvsp[(3) - (3)])); ;} - break; - - case 21: -#line 192 "said.y" - { (yyval) = said_attach_branch((yyvsp[(1) - (4)]), (yyvsp[(3) - (4)])); ;} - break; - - case 22: -#line 194 "said.y" - { (yyval) = said_attach_branch((yyvsp[(1) - (5)]), (yyvsp[(3) - (5)])); ;} - break; - - case 23: -#line 199 "said.y" - { (yyval) = said_attach_branch((yyvsp[(1) - (2)]), (yyvsp[(2) - (2)])); ;} - break; - - case 24: -#line 201 "said.y" - { (yyval) = (yyvsp[(1) - (1)]); ;} - break; - - case 25: -#line 203 "said.y" - { (yyval) = (yyvsp[(1) - (1)]); ;} - break; - - case 26: -#line 208 "said.y" - { (yyval) = (yyvsp[(1) - (1)]); ;} - break; - - case 27: -#line 210 "said.y" - { (yyval) = said_aug_branch(0x152, 0x144, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL); ;} - break; - - case 28: -#line 212 "said.y" - { (yyval) = said_attach_branch((yyvsp[(1) - (4)]), said_aug_branch(0x152, 0x144, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL)); ;} - break; - - case 29: -#line 217 "said.y" - { (yyval) = said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (3)]), (yyvsp[(3) - (3)])); ;} - break; - - case 30: -#line 219 "said.y" - { (yyval) = said_aug_branch(0x144, 0x14f, said_aug_branch(0x141, 0x144, (yyvsp[(2) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 31: -#line 221 "said.y" - { (yyval) = said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} - break; - - case 32: -#line 223 "said.y" - { (yyval) = said_aug_branch(0x152, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 33: -#line 228 "said.y" - { (yyval) = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL), (yyvsp[(3) - (3)])); ;} - break; - - case 34: -#line 230 "said.y" - { (yyval) = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} - break; - - case 35: -#line 232 "said.y" - { (yyval) = said_aug_branch(0x141, 0x14c, (yyvsp[(2) - (4)]), SAID_BRANCH_NULL); ;} - break; - - -/* Line 1267 of yacc.c. */ -#line 1665 "said.tab.c" - default: break; - } - YY_SYMBOL_PRINT ("-> $$ =", yyr1[yyn], &yyval, &yyloc); - - YYPOPSTACK (yylen); - yylen = 0; - YY_STACK_PRINT (yyss, yyssp); - - *++yyvsp = yyval; - - - /* Now `shift' the result of the reduction. Determine what state - that goes to, based on the state we popped back to and the rule - number reduced by. */ - - yyn = yyr1[yyn]; - - yystate = yypgoto[yyn - YYNTOKENS] + *yyssp; - if (0 <= yystate && yystate <= YYLAST && yycheck[yystate] == *yyssp) - yystate = yytable[yystate]; - else - yystate = yydefgoto[yyn - YYNTOKENS]; - - goto yynewstate; - - -/*------------------------------------. -| yyerrlab -- here on detecting error | -`------------------------------------*/ -yyerrlab: - /* If not already recovering from an error, report this error. */ - if (!yyerrstatus) - { - ++yynerrs; -#if ! YYERROR_VERBOSE - yyerror (YY_("syntax error")); -#else - { - YYSIZE_T yysize = yysyntax_error (0, yystate, yychar); - if (yymsg_alloc < yysize && yymsg_alloc < YYSTACK_ALLOC_MAXIMUM) - { - YYSIZE_T yyalloc = 2 * yysize; - if (! (yysize <= yyalloc && yyalloc <= YYSTACK_ALLOC_MAXIMUM)) - yyalloc = YYSTACK_ALLOC_MAXIMUM; - if (yymsg != yymsgbuf) - YYSTACK_FREE (yymsg); - yymsg = (char *) YYSTACK_ALLOC (yyalloc); - if (yymsg) - yymsg_alloc = yyalloc; - else - { - yymsg = yymsgbuf; - yymsg_alloc = sizeof yymsgbuf; - } - } - - if (0 < yysize && yysize <= yymsg_alloc) - { - (void) yysyntax_error (yymsg, yystate, yychar); - yyerror (yymsg); - } - else - { - yyerror (YY_("syntax error")); - if (yysize != 0) - goto yyexhaustedlab; - } - } -#endif - } - - - - if (yyerrstatus == 3) - { - /* If just tried and failed to reuse look-ahead token after an - error, discard it. */ - - if (yychar <= YYEOF) - { - /* Return failure if at end of input. */ - if (yychar == YYEOF) - YYABORT; - } - else - { - yydestruct ("Error: discarding", - yytoken, &yylval); - yychar = YYEMPTY; - } - } - - /* Else will try to reuse look-ahead token after shifting the error - token. */ - goto yyerrlab1; - - -/*---------------------------------------------------. -| yyerrorlab -- error raised explicitly by YYERROR. | -`---------------------------------------------------*/ -yyerrorlab: - - /* Pacify compilers like GCC when the user code never invokes - YYERROR and the label yyerrorlab therefore never appears in user - code. */ - if (/*CONSTCOND*/ 0) - goto yyerrorlab; - - /* Do not reclaim the symbols of the rule which action triggered - this YYERROR. */ - YYPOPSTACK (yylen); - yylen = 0; - YY_STACK_PRINT (yyss, yyssp); - yystate = *yyssp; - goto yyerrlab1; - - -/*-------------------------------------------------------------. -| yyerrlab1 -- common code for both syntax error and YYERROR. | -`-------------------------------------------------------------*/ -yyerrlab1: - yyerrstatus = 3; /* Each real token shifted decrements this. */ - - for (;;) - { - yyn = yypact[yystate]; - if (yyn != YYPACT_NINF) - { - yyn += YYTERROR; - if (0 <= yyn && yyn <= YYLAST && yycheck[yyn] == YYTERROR) - { - yyn = yytable[yyn]; - if (0 < yyn) - break; - } - } - - /* Pop the current state because it cannot handle the error token. */ - if (yyssp == yyss) - YYABORT; - - - yydestruct ("Error: popping", - yystos[yystate], yyvsp); - YYPOPSTACK (1); - yystate = *yyssp; - YY_STACK_PRINT (yyss, yyssp); - } - - if (yyn == YYFINAL) - YYACCEPT; - - *++yyvsp = yylval; - - - /* Shift the error token. */ - YY_SYMBOL_PRINT ("Shifting", yystos[yyn], yyvsp, yylsp); - - yystate = yyn; - goto yynewstate; - - -/*-------------------------------------. -| yyacceptlab -- YYACCEPT comes here. | -`-------------------------------------*/ -yyacceptlab: - yyresult = 0; - goto yyreturn; - -/*-----------------------------------. -| yyabortlab -- YYABORT comes here. | -`-----------------------------------*/ -yyabortlab: - yyresult = 1; - goto yyreturn; - -#ifndef yyoverflow -/*-------------------------------------------------. -| yyexhaustedlab -- memory exhaustion comes here. | -`-------------------------------------------------*/ -yyexhaustedlab: - yyerror (YY_("memory exhausted")); - yyresult = 2; - /* Fall through. */ -#endif - -yyreturn: - if (yychar != YYEOF && yychar != YYEMPTY) - yydestruct ("Cleanup: discarding lookahead", - yytoken, &yylval); - /* Do not reclaim the symbols of the rule which action triggered - this YYABORT or YYACCEPT. */ - YYPOPSTACK (yylen); - YY_STACK_PRINT (yyss, yyssp); - while (yyssp != yyss) - { - yydestruct ("Cleanup: popping", - yystos[*yyssp], yyvsp); - YYPOPSTACK (1); - } -#ifndef yyoverflow - if (yyss != yyssa) - YYSTACK_FREE (yyss); -#endif -#if YYERROR_VERBOSE - if (yymsg != yymsgbuf) - YYSTACK_FREE (yymsg); -#endif - /* Make sure YYID is used. */ - return YYID (yyresult); -} - - -#line 235 "said.y" - - -int parse_yy_token_lookup[] = {YY_COMMA, YY_AMP, YY_SLASH, YY_PARENO, YY_PARENC, YY_BRACKETSO, YY_BRACKETSC, YY_HASH, YY_LT, YY_GT}; - -static int yylex(void) { - int retval = said_tokens[said_token++]; - - if (retval < SAID_LONG(SAID_FIRST)) { - yylval = retval; - retval = WGROUP; - } else { - retval >>= 8; - - if (retval == SAID_TERM) - retval = 0; - else { - assert(retval >= SAID_FIRST); - retval = parse_yy_token_lookup[retval - SAID_FIRST]; - if (retval == YY_BRACKETSO) { - if ((said_tokens[said_token] >> 8) == SAID_LT) - retval = YY_BRACKETSO_LT; - else - if ((said_tokens[said_token] >> 8) == SAID_SLASH) - retval = YY_BRACKETSO_SLASH; - } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_BRACKO) { - retval = YY_LT_BRACKETSO; - } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_PARENO) { - retval = YY_LT_PARENO; - } - } - } - - return retval; -} - -static int said_next_node() { - return ((said_tree_pos == 0) || (said_tree_pos >= VOCAB_TREE_NODES)) ? said_tree_pos = 0 : said_tree_pos++; -} - -#define SAID_NEXT_NODE said_next_node() - -static int said_leaf_node(tree_t pos, int value) { - said_tree[pos].type = kParseTreeLeafNode; - - if (value != VALUE_IGNORE) - said_tree[pos].content.value = value; - - return pos; -} - -static int said_branch_node(tree_t pos, int left, int right) { - said_tree[pos].type = kParseTreeBranchNode; - - if (left != VALUE_IGNORE) - said_tree[pos].content.branches[0] = left; - - if (right != VALUE_IGNORE) - said_tree[pos].content.branches[1] = right; - - return pos; -} - -static tree_t said_paren(tree_t t1, tree_t t2) { - if (t1) - return said_branch_node(SAID_NEXT_NODE, t1, t2); - else - return t2; -} - -static tree_t said_value(int val, tree_t t) { - return said_branch_node(SAID_NEXT_NODE, said_leaf_node(SAID_NEXT_NODE, val), t); - -} - -static tree_t said_terminal(int val) { - return said_leaf_node(SAID_NEXT_NODE, val); -} - -static tree_t said_aug_branch(int n1, int n2, tree_t t1, tree_t t2) { - int retval; - - retval = said_branch_node(SAID_NEXT_NODE, - said_branch_node(SAID_NEXT_NODE, - said_leaf_node(SAID_NEXT_NODE, n1), - said_branch_node(SAID_NEXT_NODE, - said_leaf_node(SAID_NEXT_NODE, n2), - t1) - ), - t2); - -#ifdef SAID_DEBUG - fprintf(stderr, "AUG(0x%x, 0x%x, [%04x], [%04x]) = [%04x]\n", n1, n2, t1, t2, retval); -#endif - - return retval; -} - -static tree_t said_attach_branch(tree_t base, tree_t attacheant) { -#ifdef SAID_DEBUG - fprintf(stderr, "ATT2([%04x], [%04x]) = [%04x]\n", base, attacheant, base); -#endif - - if (!attacheant) - return base; - if (!base) - return attacheant; - - if (!base) - return 0; // Happens if we're out of space - - said_branch_node(base, VALUE_IGNORE, attacheant); - - return base; -} - -static said_spec_t said_top_branch(tree_t first) { -#ifdef SAID_DEBUG - fprintf(stderr, "TOP([%04x])\n", first); -#endif - said_branch_node(0, 1, 2); - said_leaf_node(1, 0x141); // Magic number #1 - said_branch_node(2, 3, first); - said_leaf_node(3, 0x13f); // Magic number #2 - - ++said_blessed; - - return 0; -} - -static int said_parse_spec(byte *spec) { - int nextitem; - - said_parse_error = NULL; - said_token = 0; - said_tokens_nr = 0; - said_blessed = 0; - - said_tree_pos = SAID_TREE_START; - - do { - nextitem = *spec++; - if (nextitem < SAID_FIRST) - said_tokens[said_tokens_nr++] = nextitem << 8 | *spec++; - else - said_tokens[said_tokens_nr++] = SAID_LONG(nextitem); - - } while ((nextitem != SAID_TERM) && (said_tokens_nr < MAX_SAID_TOKENS)); - - if (nextitem == SAID_TERM) - yyparse(); - else { - warning("SAID spec is too long"); - return 1; - } - - if (said_parse_error) { - warning("Error while parsing SAID spec: %s", said_parse_error); - free(said_parse_error); - return 1; - } - - if (said_tree_pos == 0) { - warning("Out of tree space while parsing SAID spec"); - return 1; - } - - if (said_blessed != 1) { - warning("Found multiple top branches"); - return 1; - } - - return 0; -} - -/**********************/ -/**** Augmentation ****/ -/**********************/ - -// primitive functions - -#define AUG_READ_BRANCH(a, br, p) \ - if (tree[p].type != kParseTreeBranchNode) \ - return 0; \ - a = tree[p].content.branches[br]; - -#define AUG_READ_VALUE(a, p) \ - if (tree[p].type != kParseTreeLeafNode) \ - return 0; \ - a = tree[p].content.value; - -#define AUG_ASSERT(i) \ - if (!i) return 0; - -static int aug_get_next_sibling(parse_tree_node_t *tree, int pos, int *first, int *second) { - // Returns the next sibling relative to the specified position in 'tree', - // sets *first and *second to its augment node values, returns the new position - // or 0 if there was no next sibling - int seek, valpos; - - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(seek, 0, pos); - AUG_ASSERT(seek); - - // Now retrieve first value - AUG_READ_BRANCH(valpos, 0, seek); - AUG_ASSERT(valpos); - AUG_READ_VALUE(*first, valpos); - - // Get second value - AUG_READ_BRANCH(seek, 1, seek); - AUG_ASSERT(seek); - AUG_READ_BRANCH(valpos, 0, seek); - AUG_ASSERT(valpos); - AUG_READ_VALUE(*second, valpos); - - return pos; -} - -static int aug_get_wgroup(parse_tree_node_t *tree, int pos) { - // Returns 0 if pos in tree is not the root of a 3-element list, otherwise - // it returns the last element (which, in practice, is the word group - int val; - - AUG_READ_BRANCH(pos, 0, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_VALUE(val, pos); - - return val; -} - -static int aug_get_base_node(parse_tree_node_t *tree) { - int startpos = 0; - AUG_READ_BRANCH(startpos, 1, startpos); - - return startpos; -} - -// semi-primitive functions - -static int aug_get_first_child(parse_tree_node_t *tree, int pos, int *first, int *second) { - // like aug_get_next_sibling, except that it recurses into the tree and - // finds the first child (usually *not* Ayanami Rei) of the current branch - // rather than its next sibling. - AUG_READ_BRANCH(pos, 0, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - - return aug_get_next_sibling(tree, pos, first, second); -} - -static void aug_find_words_recursively(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, - int *ref_words, int *ref_words_nr, int maxwords, int refbranch) { - // Finds and lists all base (141) and reference (144) words */ - int major, minor; - int word; - int pos = aug_get_first_child(tree, startpos, &major, &minor); - - //if (major == WORD_TYPE_REF) - // refbranch = 1; - - while (pos) { - if ((word = aug_get_wgroup(tree, pos))) { // found a word - if (!refbranch && major == WORD_TYPE_BASE) { - if ((*base_words_nr) == maxwords) { - warning("Out of regular words"); - return; // return gracefully - } - - base_words[*base_words_nr] = word; // register word - ++(*base_words_nr); - - } - if (major == WORD_TYPE_REF || refbranch) { - if ((*ref_words_nr) == maxwords) { - warning("Out of reference words"); - return; // return gracefully - } - - ref_words[*ref_words_nr] = word; // register word - ++(*ref_words_nr); - - } - if (major != WORD_TYPE_SYNTACTIC_SUGAR && major != WORD_TYPE_BASE && major != WORD_TYPE_REF) - warning("aug_find_words_recursively(): Unknown word type %03x", major); - - } else // Did NOT find a word group: Attempt to recurse - aug_find_words_recursively(tree, pos, base_words, base_words_nr, - ref_words, ref_words_nr, maxwords, refbranch || major == WORD_TYPE_REF); - - pos = aug_get_next_sibling(tree, pos, &major, &minor); - } -} - - -static void aug_find_words(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, - int *ref_words, int *ref_words_nr, int maxwords) { - // initializing wrapper for aug_find_words_recursively() - *base_words_nr = 0; - *ref_words_nr = 0; - - aug_find_words_recursively(tree, startpos, base_words, base_words_nr, ref_words, ref_words_nr, maxwords, 0); -} - - -static int aug_contains_word(int *list, int length, int word) { - int i; - - if (word == ANYWORD) - return (length); - - for (i = 0; i < length; i++) - if (list[i] == word) - return 1; - - return 0; -} - - -static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_branch, int major, int minor, int *base_words, int base_words_nr, - int *ref_words, int ref_words_nr); - -static int augment_match_expression_p(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_basepos, int major, int minor, - int *base_words, int base_words_nr, int *ref_words, int ref_words_nr) { - int cmajor, cminor, cpos; - cpos = aug_get_first_child(saidt, augment_pos, &cmajor, &cminor); - if (!cpos) { - warning("augment_match_expression_p(): Empty condition"); - return 1; - } - - scidprintf("Attempting to match (%03x %03x (%03x %03x\n", major, minor, cmajor, cminor); - - if ((major == WORD_TYPE_BASE) && (minor == AUGMENT_SENTENCE_MINOR_RECURSE)) - return augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr); - - switch (major) { - - case WORD_TYPE_BASE: - while (cpos) { - if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { - int word = aug_get_wgroup(saidt, cpos); - scidprintf("Looking for word %03x\n", word); - - if (aug_contains_word(base_words, base_words_nr, word)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { - if (augment_sentence_expression(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { - int gc_major, gc_minor; - int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); - - while (gchild) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, - minor, base_words, base_words_nr, - ref_words, ref_words_nr)) - return 1; - gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); - } - } else - warning("augment_match_expression_p(): Unknown type 141 minor number %3x", cminor); - - cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); - - } - break; - - case WORD_TYPE_REF: - while (cpos) { - if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { - int word = aug_get_wgroup(saidt, cpos); - scidprintf("Looking for refword %03x\n", word); - - if (aug_contains_word(ref_words, ref_words_nr, word)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { - int gc_major, gc_minor; - int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); - - while (gchild) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, - minor, base_words, base_words_nr, - ref_words, ref_words_nr)) - return 1; - gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); - } - } else - warning("augment_match_expression_p(): Unknown type 144 minor number %3x", cminor); - - cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); - - } - break; - - case AUGMENT_SENTENCE_PART_BRACKETS: - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - - scidprintf("Didn't match subexpression; checking sub-bracked predicate %03x\n", cmajor); - - switch (cmajor) { - case WORD_TYPE_BASE: - if (!base_words_nr) - return 1; - break; - - case WORD_TYPE_REF: - if (!ref_words_nr) - return 1; - break; - - default: - warning("augment_match_expression_p(): (subp1) Unkonwn sub-bracket predicate %03x", cmajor); - } - - break; - - default: - warning("augment_match_expression_p(): Unknown predicate %03x", major); - - } - - scidprintf("augment_match_expression_p(): Generic failure\n"); - - return 0; -} - -static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_branch, int major, int minor, int *base_words, int base_words_nr, - int *ref_words, int ref_words_nr) { - int check_major, check_minor; - int check_pos = aug_get_first_child(saidt, augment_pos, &check_major, &check_minor); - do { - if (!(augment_match_expression_p(saidt, check_pos, parset, parse_branch, check_major, check_minor, - base_words, base_words_nr, ref_words, ref_words_nr))) - return 0; - } while ((check_pos = aug_get_next_sibling(saidt, check_pos, &check_major, &check_minor))); - - return 1; -} - -static int augment_sentence_part(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, int parse_basepos, int major, int minor) { - int pmajor, pminor; - int parse_branch = parse_basepos; - int optional = 0; - int foundwords = 0; - - scidprintf("Augmenting (%03x %03x\n", major, minor); - - if (major == AUGMENT_SENTENCE_PART_BRACKETS) { // '[/ foo]' is true if '/foo' or if there - // exists no x for which '/x' is true - if ((augment_pos = aug_get_first_child(saidt, augment_pos, &major, &minor))) { - scidprintf("Optional part: Now augmenting (%03x %03x\n", major, minor); - optional = 1; - } else { - scidprintf("Matched empty optional expression\n"); - return 1; - } - } - - if ((major < 0x141) || (major > 0x143)) { - scidprintf("augment_sentence_part(): Unexpected sentence part major number %03x\n", major); - return 0; - } - - while ((parse_branch = aug_get_next_sibling(parset, parse_branch, &pmajor, &pminor))) { - if (pmajor == major) { // found matching sentence part - int success; - int base_words_nr; - int ref_words_nr; - int base_words[AUGMENT_MAX_WORDS]; - int ref_words[AUGMENT_MAX_WORDS]; -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION - int i; -#endif - - scidprintf("Found match with pminor = %03x\n", pminor); - aug_find_words(parset, parse_branch, base_words, &base_words_nr, ref_words, &ref_words_nr, AUGMENT_MAX_WORDS); - foundwords |= (ref_words_nr | base_words_nr); -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION - printf("%d base words:", base_words_nr); - for (i = 0; i < base_words_nr; i++) - printf(" %03x", base_words[i]); - printf("\n%d reference words:", ref_words_nr); - for (i = 0; i < ref_words_nr; i++) - printf(" %03x", ref_words[i]); - printf("\n"); -#endif - - success = augment_sentence_expression(saidt, augment_pos, parset, parse_basepos, major, minor, - base_words, base_words_nr, ref_words, ref_words_nr); - - if (success) { - scidprintf("SUCCESS on augmenting (%03x %03x\n", major, minor); - return 1; - } - } - } - - if (optional && (foundwords == 0)) { - scidprintf("Found no words and optional branch => SUCCESS on augmenting (%03x %03x\n", major, minor); - return 1; - } - scidprintf("FAILURE on augmenting (%03x %03x\n", major, minor); - - return 0; -} - -static int augment_parse_nodes(parse_tree_node_t *parset, parse_tree_node_t *saidt) { - int augment_basepos = 0; - int parse_basepos; - int major, minor; - int dontclaim = 0; - - parse_basepos = aug_get_base_node(parset); - if (!parse_basepos) { - warning("augment_parse_nodes(): Parse tree is corrupt"); - return 0; - } - - augment_basepos = aug_get_base_node(saidt); - if (!augment_basepos) { - warning("augment_parse_nodes(): Said tree is corrupt"); - return 0; - } - - while ((augment_basepos = aug_get_next_sibling(saidt, augment_basepos, &major, &minor))) { - if ((major == 0x14b) && (minor == SAID_LONG(SAID_GT))) - dontclaim = 1; // special case - else // normal sentence part - if (!(augment_sentence_part(saidt, augment_basepos, parset, parse_basepos, major, minor))) { - scidprintf("Returning failure\n"); - return 0; // fail - } - } - - scidprintf("Returning success with dontclaim=%d\n", dontclaim); - - if (dontclaim) - return SAID_PARTIAL_MATCH; - else - return 1; // full match -} - - -/*******************/ -/**** Main code ****/ -/*******************/ - -int said(EngineState *s, byte *spec, bool verbose) { - int retval; - - parse_tree_node_t *parse_tree_ptr = s->_voc->_parserNodes; - - if (s->parserIsValid) { - if (said_parse_spec(spec)) { - printf("Offending spec was: "); - s->_voc->decipherSaidBlock(spec); - return SAID_NO_MATCH; - } - - if (verbose) - vocab_dump_parse_tree("Said-tree", said_tree); // Nothing better to do yet - retval = augment_parse_nodes(parse_tree_ptr, &(said_tree[0])); - - if (!retval) - return SAID_NO_MATCH; - else if (retval != SAID_PARTIAL_MATCH) - return SAID_FULL_MATCH; - else - return SAID_PARTIAL_MATCH; - } - - return SAID_NO_MATCH; -} - - -#ifdef SAID_DEBUG_PROGRAM -int main (int argc, char *argv) { - byte block[] = {0x01, 0x00, 0xf8, 0xf5, 0x02, 0x01, 0xf6, 0xf2, 0x02, 0x01, 0xf2, 0x01, 0x03, 0xff}; - EngineState s; - - s.parser_valid = 1; - said(&s, block); -} -#endif - -} // End of namespace Sci - diff --git a/engines/sci/engine/said.y b/engines/sci/engine/said.y deleted file mode 100644 index 1a3b2f52c8..0000000000 --- a/engines/sci/engine/said.y +++ /dev/null @@ -1,838 +0,0 @@ -%{ -/* ScummVM - Graphic Adventure Engine - * - * ScummVM is the legal property of its developers, whose names - * are too numerous to list here. Please refer to the COPYRIGHT - * file distributed with this source distribution. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - * - * $URL$ - * $Id$ - * - */ - -#include "sci/engine/state.h" - - -// Bison generates an empty switch statement that gives a warning in MSVC. -// This disables that warning. -#ifdef _MSC_VER -#pragma warning(disable:4065) -#endif - - -namespace Sci { - -#define SAID_BRANCH_NULL 0 - -#define MAX_SAID_TOKENS 128 - -// Maximum number of words to be expected in a parsed sentence -#define AUGMENT_MAX_WORDS 64 - - -#define ANYWORD 0xfff - -#define WORD_TYPE_BASE 0x141 -#define WORD_TYPE_REF 0x144 -#define WORD_TYPE_SYNTACTIC_SUGAR 0x145 - -#define AUGMENT_SENTENCE_PART_BRACKETS 0x152 - -// Minor numbers -#define AUGMENT_SENTENCE_MINOR_MATCH_PHRASE 0x14c -#define AUGMENT_SENTENCE_MINOR_MATCH_WORD 0x153 -#define AUGMENT_SENTENCE_MINOR_RECURSE 0x144 -#define AUGMENT_SENTENCE_MINOR_PARENTHESES 0x14f - - -#undef YYDEBUG /*1*/ -//#define SAID_DEBUG*/ -//#define SCI_DEBUG_PARSE_TREE_AUGMENTATION // uncomment to debug parse tree augmentation - - -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION -#define scidprintf printf -#else -void print_nothing(...) { } -#define scidprintf print_nothing -#endif - - -static char *said_parse_error; - -static int said_token; -static int said_tokens_nr; -static int said_tokens[MAX_SAID_TOKENS]; -static int said_blessed; // increminated by said_top_branch - -static int said_tree_pos; // Set to 0 if we're out of space -#define SAID_TREE_START 4; // Reserve space for the 4 top nodes - -#define VALUE_IGNORE -424242 - -static parse_tree_node_t said_tree[VOCAB_TREE_NODES]; - -typedef int wgroup_t; -typedef int tree_t; -typedef int said_spec_t; - -static tree_t said_aug_branch(int, int, tree_t, tree_t); -static tree_t said_attach_branch(tree_t, tree_t); -/* -static tree_t said_wgroup_branch(wgroup_t); -*/ -static said_spec_t said_top_branch(tree_t); -static tree_t said_paren(tree_t, tree_t); -static tree_t said_value(int, tree_t); -static tree_t said_terminal(int); - -static int yylex(); - -static int yyerror(const char *s) { - said_parse_error = strdup(s); - return 1; /* Abort */ -} - -%} - -%token WGROUP /* Word group */ -%token YY_COMMA /* 0xf0 */ -%token YY_AMP /* 0xf1 */ -%token YY_SLASH /* 0xf2 */ -%token YY_PARENO /* 0xf3 */ -%token YY_PARENC /* 0xf4 */ -%token YY_BRACKETSO /* 0xf5 */ -%token YY_BRACKETSC /* 0xf6 */ -%token YY_HASH /* 0xf7 */ -%token YY_LT /* 0xf8 */ -%token YY_GT /* 0xf9 */ -%token YY_BRACKETSO_LT /* special token used to imitate LR(2) behaviour */ -%token YY_BRACKETSO_SLASH /* special token used to imitate LR(2) behaviour */ -%token YY_LT_BRACKETSO /* special token used to imitate LR(2) behaviour */ -%token YY_LT_PARENO /* special token used to imitate LR(2) behaviour */ - -%% - -saidspec : leftspec optcont - { $$ = said_top_branch(said_attach_branch($1, $2)); } - | leftspec midspec optcont - { $$ = said_top_branch(said_attach_branch($1, said_attach_branch($2, $3))); } - | leftspec midspec rightspec optcont - { $$ = said_top_branch(said_attach_branch($1, said_attach_branch($2, said_attach_branch($3, $4)))); } - ; - - -optcont : /* empty */ - { $$ = SAID_BRANCH_NULL; } - | YY_GT - { $$ = said_paren(said_value(0x14b, said_value(0xf900, said_terminal(0xf900))), SAID_BRANCH_NULL); } - ; - - - -leftspec : /* empty */ - { $$ = SAID_BRANCH_NULL; } - | expr - { $$ = said_paren(said_value(0x141, said_value(0x149, $1)), SAID_BRANCH_NULL); } - ; - - - -midspec : YY_SLASH expr - { $$ = said_aug_branch(0x142, 0x14a, $2, SAID_BRANCH_NULL); } - | YY_BRACKETSO_SLASH YY_SLASH expr YY_BRACKETSC - { $$ = said_aug_branch(0x152, 0x142, said_aug_branch(0x142, 0x14a, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - | YY_SLASH - { $$ = SAID_BRANCH_NULL; } - ; - - - -rightspec : YY_SLASH expr - { $$ = said_aug_branch(0x143, 0x14a, $2, SAID_BRANCH_NULL); } - | YY_BRACKETSO_SLASH YY_SLASH expr YY_BRACKETSC - { $$ = said_aug_branch(0x152, 0x143, said_aug_branch(0x143, 0x14a, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - | YY_SLASH - { $$ = SAID_BRANCH_NULL; } - ; - - -word : WGROUP - { $$ = said_paren(said_value(0x141, said_value(0x153, said_terminal($1))), SAID_BRANCH_NULL); } - ; - - -cwordset : wordset - { $$ = said_aug_branch(0x141, 0x14f, $1, SAID_BRANCH_NULL); } - | YY_BRACKETSO wordset YY_BRACKETSC - { $$ = said_aug_branch(0x141, 0x14f, said_aug_branch(0x152, 0x14c, said_aug_branch(0x141, 0x14f, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - ; - - -wordset : word - { $$ = $1; } - | YY_PARENO expr YY_PARENC - { $$ = said_aug_branch(0x141, 0x14c, $2, SAID_BRANCH_NULL); } - | wordset YY_COMMA wordset - { $$ = said_attach_branch($1, $3); } - | wordset YY_BRACKETSO_LT wordrefset YY_BRACKETSC - { $$ = said_attach_branch($1, $3); } - | wordset YY_COMMA YY_BRACKETSO wordset YY_BRACKETSC - { $$ = said_attach_branch($1, $3); } - ; - - -expr : cwordset cwordrefset - { $$ = said_attach_branch($1, $2); } - | cwordset - { $$ = $1; } - | cwordrefset - { $$ = $1; } - ; - - -cwordrefset : wordrefset - { $$ = $1; } - | YY_BRACKETSO_LT wordrefset YY_BRACKETSC - { $$ = said_aug_branch(0x152, 0x144, $2, SAID_BRANCH_NULL); } - | wordrefset YY_BRACKETSO_LT wordrefset YY_BRACKETSC - { $$ = said_attach_branch($1, said_aug_branch(0x152, 0x144, $3, SAID_BRANCH_NULL)); } - ; - - -wordrefset : YY_LT word recref - { $$ = said_aug_branch(0x144, 0x14f, $2, $3); } - | YY_LT_PARENO YY_PARENO expr YY_PARENC - { $$ = said_aug_branch(0x144, 0x14f, said_aug_branch(0x141, 0x144, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - | YY_LT wordset - { $$ = said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL); } - | YY_LT_BRACKETSO YY_BRACKETSO wordset YY_BRACKETSC - { $$ = said_aug_branch(0x152, 0x144, said_aug_branch(0x144, 0x14f, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - ; - - -recref : YY_LT wordset recref - { $$ = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL), $3); } - | YY_LT wordset - { $$ = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } - | YY_LT_PARENO YY_PARENO expr YY_PARENC - { $$ = said_aug_branch(0x141, 0x14c, $2, SAID_BRANCH_NULL); } - ; - -%% - -int parse_yy_token_lookup[] = {YY_COMMA, YY_AMP, YY_SLASH, YY_PARENO, YY_PARENC, YY_BRACKETSO, YY_BRACKETSC, YY_HASH, YY_LT, YY_GT}; - -static int yylex() { - int retval = said_tokens[said_token++]; - - if (retval < SAID_LONG(SAID_FIRST)) { - yylval = retval; - retval = WGROUP; - } else { - retval >>= 8; - - if (retval == SAID_TERM) - retval = 0; - else { - assert(retval >= SAID_FIRST); - retval = parse_yy_token_lookup[retval - SAID_FIRST]; - if (retval == YY_BRACKETSO) { - if ((said_tokens[said_token] >> 8) == SAID_LT) - retval = YY_BRACKETSO_LT; - else - if ((said_tokens[said_token] >> 8) == SAID_SLASH) - retval = YY_BRACKETSO_SLASH; - } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_BRACKO) { - retval = YY_LT_BRACKETSO; - } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_PARENO) { - retval = YY_LT_PARENO; - } - } - } - - return retval; -} - -static int said_next_node() { - return ((said_tree_pos == 0) || (said_tree_pos >= VOCAB_TREE_NODES)) ? said_tree_pos = 0 : said_tree_pos++; -} - -#define SAID_NEXT_NODE said_next_node() - -static int said_leaf_node(tree_t pos, int value) { - said_tree[pos].type = PARSE_TREE_NODE_LEAF; - - if (value != VALUE_IGNORE) - said_tree[pos].content.value = value; - - return pos; -} - -static int said_branch_node(tree_t pos, int left, int right) { - said_tree[pos].type = PARSE_TREE_NODE_BRANCH; - - if (left != VALUE_IGNORE) - said_tree[pos].content.branches[0] = left; - - if (right != VALUE_IGNORE) - said_tree[pos].content.branches[1] = right; - - return pos; -} - -static tree_t said_paren(tree_t t1, tree_t t2) { - if (t1) - return said_branch_node(SAID_NEXT_NODE, t1, t2); - else - return t2; -} - -static tree_t said_value(int val, tree_t t) { - return said_branch_node(SAID_NEXT_NODE, said_leaf_node(SAID_NEXT_NODE, val), t); - -} - -static tree_t said_terminal(int val) { - return said_leaf_node(SAID_NEXT_NODE, val); -} - -static tree_t said_aug_branch(int n1, int n2, tree_t t1, tree_t t2) { - int retval; - - retval = said_branch_node(SAID_NEXT_NODE, - said_branch_node(SAID_NEXT_NODE, - said_leaf_node(SAID_NEXT_NODE, n1), - said_branch_node(SAID_NEXT_NODE, - said_leaf_node(SAID_NEXT_NODE, n2), - t1) - ), - t2); - -#ifdef SAID_DEBUG - fprintf(stderr, "AUG(0x%x, 0x%x, [%04x], [%04x]) = [%04x]\n", n1, n2, t1, t2, retval); -#endif - - return retval; -} - -static tree_t said_attach_branch(tree_t base, tree_t attacheant) { -#ifdef SAID_DEBUG - fprintf(stderr, "ATT2([%04x], [%04x]) = [%04x]\n", base, attacheant, base); -#endif - - if (!attacheant) - return base; - if (!base) - return attacheant; - - if (!base) - return 0; // Happens if we're out of space - - said_branch_node(base, VALUE_IGNORE, attacheant); - - return base; -} - -static said_spec_t said_top_branch(tree_t first) { -#ifdef SAID_DEBUG - fprintf(stderr, "TOP([%04x])\n", first); -#endif - said_branch_node(0, 1, 2); - said_leaf_node(1, 0x141); // Magic number #1 - said_branch_node(2, 3, first); - said_leaf_node(3, 0x13f); // Magic number #2 - - ++said_blessed; - - return 0; -} - -static int said_parse_spec(byte *spec) { - int nextitem; - - said_parse_error = NULL; - said_token = 0; - said_tokens_nr = 0; - said_blessed = 0; - - said_tree_pos = SAID_TREE_START; - - do { - nextitem = *spec++; - if (nextitem < SAID_FIRST) - said_tokens[said_tokens_nr++] = nextitem << 8 | *spec++; - else - said_tokens[said_tokens_nr++] = SAID_LONG(nextitem); - - } while ((nextitem != SAID_TERM) && (said_tokens_nr < MAX_SAID_TOKENS)); - - if (nextitem == SAID_TERM) - yyparse(); - else { - warning("SAID spec is too long"); - return 1; - } - - if (said_parse_error) { - warning("Error while parsing SAID spec: %s", said_parse_error); - free(said_parse_error); - return 1; - } - - if (said_tree_pos == 0) { - warning("Out of tree space while parsing SAID spec"); - return 1; - } - - if (said_blessed != 1) { - warning("Found multiple top branches"); - return 1; - } - - return 0; -} - -/**********************/ -/**** Augmentation ****/ -/**********************/ - -// primitive functions - -#define AUG_READ_BRANCH(a, br, p) \ - if (tree[p].type != PARSE_TREE_NODE_BRANCH) \ - return 0; \ - a = tree[p].content.branches[br]; - -#define AUG_READ_VALUE(a, p) \ - if (tree[p].type != PARSE_TREE_NODE_LEAF) \ - return 0; \ - a = tree[p].content.value; - -#define AUG_ASSERT(i) \ - if (!i) return 0; - -static int aug_get_next_sibling(parse_tree_node_t *tree, int pos, int *first, int *second) { - // Returns the next sibling relative to the specified position in 'tree', - // sets *first and *second to its augment node values, returns the new position - // or 0 if there was no next sibling - int seek, valpos; - - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(seek, 0, pos); - AUG_ASSERT(seek); - - // Now retrieve first value - AUG_READ_BRANCH(valpos, 0, seek); - AUG_ASSERT(valpos); - AUG_READ_VALUE(*first, valpos); - - // Get second value - AUG_READ_BRANCH(seek, 1, seek); - AUG_ASSERT(seek); - AUG_READ_BRANCH(valpos, 0, seek); - AUG_ASSERT(valpos); - AUG_READ_VALUE(*second, valpos); - - return pos; -} - -static int aug_get_wgroup(parse_tree_node_t *tree, int pos) { - // Returns 0 if pos in tree is not the root of a 3-element list, otherwise - // it returns the last element (which, in practice, is the word group - int val; - - AUG_READ_BRANCH(pos, 0, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - AUG_READ_VALUE(val, pos); - - return val; -} - -static int aug_get_base_node(parse_tree_node_t *tree) { - int startpos = 0; - AUG_READ_BRANCH(startpos, 1, startpos); - - return startpos; -} - -// semi-primitive functions - -static int aug_get_first_child(parse_tree_node_t *tree, int pos, int *first, int *second) { - // like aug_get_next_sibling, except that it recurses into the tree and - // finds the first child (usually *not* Ayanami Rei) of the current branch - // rather than its next sibling. - AUG_READ_BRANCH(pos, 0, pos); - AUG_ASSERT(pos); - AUG_READ_BRANCH(pos, 1, pos); - AUG_ASSERT(pos); - - return aug_get_next_sibling(tree, pos, first, second); -} - -static void aug_find_words_recursively(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, - int *ref_words, int *ref_words_nr, int maxwords, int refbranch) { - // Finds and lists all base (141) and reference (144) words */ - int major, minor; - int word; - int pos = aug_get_first_child(tree, startpos, &major, &minor); - - //if (major == WORD_TYPE_REF) - // refbranch = 1; - - while (pos) { - if ((word = aug_get_wgroup(tree, pos))) { // found a word - if (!refbranch && major == WORD_TYPE_BASE) { - if ((*base_words_nr) == maxwords) { - warning("Out of regular words"); - return; // return gracefully - } - - base_words[*base_words_nr] = word; // register word - ++(*base_words_nr); - - } - if (major == WORD_TYPE_REF || refbranch) { - if ((*ref_words_nr) == maxwords) { - warning("Out of reference words"); - return; // return gracefully - } - - ref_words[*ref_words_nr] = word; // register word - ++(*ref_words_nr); - - } - if (major != WORD_TYPE_SYNTACTIC_SUGAR && major != WORD_TYPE_BASE && major != WORD_TYPE_REF) - warning("aug_find_words_recursively(): Unknown word type %03x", major); - - } else // Did NOT find a word group: Attempt to recurse - aug_find_words_recursively(tree, pos, base_words, base_words_nr, - ref_words, ref_words_nr, maxwords, refbranch || major == WORD_TYPE_REF); - - pos = aug_get_next_sibling(tree, pos, &major, &minor); - } -} - - -static void aug_find_words(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, - int *ref_words, int *ref_words_nr, int maxwords) { - // initializing wrapper for aug_find_words_recursively() - *base_words_nr = 0; - *ref_words_nr = 0; - - aug_find_words_recursively(tree, startpos, base_words, base_words_nr, ref_words, ref_words_nr, maxwords, 0); -} - - -static int aug_contains_word(int *list, int length, int word) { - int i; - - if (word == ANYWORD) - return (length); - - for (i = 0; i < length; i++) - if (list[i] == word) - return 1; - - return 0; -} - - -static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_branch, int major, int minor, int *base_words, int base_words_nr, - int *ref_words, int ref_words_nr); - -static int augment_match_expression_p(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_basepos, int major, int minor, - int *base_words, int base_words_nr, int *ref_words, int ref_words_nr) { - int cmajor, cminor, cpos; - cpos = aug_get_first_child(saidt, augment_pos, &cmajor, &cminor); - if (!cpos) { - warning("augment_match_expression_p(): Empty condition"); - return 1; - } - - scidprintf("Attempting to match (%03x %03x (%03x %03x\n", major, minor, cmajor, cminor); - - if ((major == WORD_TYPE_BASE) && (minor == AUGMENT_SENTENCE_MINOR_RECURSE)) - return augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr); - - switch (major) { - - case WORD_TYPE_BASE: - while (cpos) { - if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { - int word = aug_get_wgroup(saidt, cpos); - scidprintf("Looking for word %03x\n", word); - - if (aug_contains_word(base_words, base_words_nr, word)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { - if (augment_sentence_expression(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { - int gc_major, gc_minor; - int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); - - while (gchild) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, - minor, base_words, base_words_nr, - ref_words, ref_words_nr)) - return 1; - gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); - } - } else - warning("augment_match_expression_p(): Unknown type 141 minor number %3x", cminor); - - cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); - - } - break; - - case WORD_TYPE_REF: - while (cpos) { - if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { - int word = aug_get_wgroup(saidt, cpos); - scidprintf("Looking for refword %03x\n", word); - - if (aug_contains_word(ref_words, ref_words_nr, word)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { - int gc_major, gc_minor; - int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); - - while (gchild) { - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, - minor, base_words, base_words_nr, - ref_words, ref_words_nr)) - return 1; - gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); - } - } else - warning("augment_match_expression_p(): Unknown type 144 minor number %3x", cminor); - - cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); - - } - break; - - case AUGMENT_SENTENCE_PART_BRACKETS: - if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, - base_words, base_words_nr, ref_words, ref_words_nr)) - return 1; - - scidprintf("Didn't match subexpression; checking sub-bracked predicate %03x\n", cmajor); - - switch (cmajor) { - case WORD_TYPE_BASE: - if (!base_words_nr) - return 1; - break; - - case WORD_TYPE_REF: - if (!ref_words_nr) - return 1; - break; - - default: - warning("augment_match_expression_p(): (subp1) Unkonwn sub-bracket predicate %03x", cmajor); - } - - break; - - default: - warning("augment_match_expression_p(): Unknown predicate %03x", major); - - } - - scidprintf("augment_match_expression_p(): Generic failure\n"); - - return 0; -} - -static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, - int parse_branch, int major, int minor, int *base_words, int base_words_nr, - int *ref_words, int ref_words_nr) { - int check_major, check_minor; - int check_pos = aug_get_first_child(saidt, augment_pos, &check_major, &check_minor); - do { - if (!(augment_match_expression_p(saidt, check_pos, parset, parse_branch, check_major, check_minor, - base_words, base_words_nr, ref_words, ref_words_nr))) - return 0; - } while ((check_pos = aug_get_next_sibling(saidt, check_pos, &check_major, &check_minor))); - - return 1; -} - -static int augment_sentence_part(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, int parse_basepos, int major, int minor) { - int pmajor, pminor; - int parse_branch = parse_basepos; - int optional = 0; - int foundwords = 0; - - scidprintf("Augmenting (%03x %03x\n", major, minor); - - if (major == AUGMENT_SENTENCE_PART_BRACKETS) { // '[/ foo]' is true if '/foo' or if there - // exists no x for which '/x' is true - if ((augment_pos = aug_get_first_child(saidt, augment_pos, &major, &minor))) { - scidprintf("Optional part: Now augmenting (%03x %03x\n", major, minor); - optional = 1; - } else { - scidprintf("Matched empty optional expression\n"); - return 1; - } - } - - if ((major < 0x141) || (major > 0x143)) { - scidprintf("augment_sentence_part(): Unexpected sentence part major number %03x\n", major); - return 0; - } - - while ((parse_branch = aug_get_next_sibling(parset, parse_branch, &pmajor, &pminor))) { - if (pmajor == major) { // found matching sentence part - int success; - int base_words_nr; - int ref_words_nr; - int base_words[AUGMENT_MAX_WORDS]; - int ref_words[AUGMENT_MAX_WORDS]; -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION - int i; -#endif - - scidprintf("Found match with pminor = %03x\n", pminor); - aug_find_words(parset, parse_branch, base_words, &base_words_nr, ref_words, &ref_words_nr, AUGMENT_MAX_WORDS); - foundwords |= (ref_words_nr | base_words_nr); -#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION - printf("%d base words:", base_words_nr); - for (i = 0; i < base_words_nr; i++) - printf(" %03x", base_words[i]); - printf("\n%d reference words:", ref_words_nr); - for (i = 0; i < ref_words_nr; i++) - printf(" %03x", ref_words[i]); - printf("\n"); -#endif - - success = augment_sentence_expression(saidt, augment_pos, parset, parse_basepos, major, minor, - base_words, base_words_nr, ref_words, ref_words_nr); - - if (success) { - scidprintf("SUCCESS on augmenting (%03x %03x\n", major, minor); - return 1; - } - } - } - - if (optional && (foundwords == 0)) { - scidprintf("Found no words and optional branch => SUCCESS on augmenting (%03x %03x\n", major, minor); - return 1; - } - scidprintf("FAILURE on augmenting (%03x %03x\n", major, minor); - - return 0; -} - -static int augment_parse_nodes(parse_tree_node_t *parset, parse_tree_node_t *saidt) { - int augment_basepos = 0; - int parse_basepos; - int major, minor; - int dontclaim = 0; - - parse_basepos = aug_get_base_node(parset); - if (!parse_basepos) { - warning("augment_parse_nodes(): Parse tree is corrupt"); - return 0; - } - - augment_basepos = aug_get_base_node(saidt); - if (!augment_basepos) { - warning("augment_parse_nodes(): Said tree is corrupt"); - return 0; - } - - while ((augment_basepos = aug_get_next_sibling(saidt, augment_basepos, &major, &minor))) { - if ((major == 0x14b) && (minor == SAID_LONG(SAID_GT))) - dontclaim = 1; // special case - else // normal sentence part - if (!(augment_sentence_part(saidt, augment_basepos, parset, parse_basepos, major, minor))) { - scidprintf("Returning failure\n"); - return 0; // fail - } - } - - scidprintf("Returning success with dontclaim=%d\n", dontclaim); - - if (dontclaim) - return SAID_PARTIAL_MATCH; - else - return 1; // full match -} - - -/*******************/ -/**** Main code ****/ -/*******************/ - -int said(EngineState *s, byte *spec, bool verbose) { - int retval; - - parse_tree_node_t *parse_tree_ptr = s->_voc->_parser_nodes; - - if (s->parserIsValid) { - if (said_parse_spec(s, spec)) { - warning("Offending spec was: "); - s->_voc->decipherSaidBlock(spec); - return SAID_NO_MATCH; - } - - if (verbose) - vocab_dump_parse_tree("Said-tree", said_tree); // Nothing better to do yet - retval = augment_parse_nodes(parse_tree_ptr, &(said_tree[0])); - - if (!retval) - return SAID_NO_MATCH; - else if (retval != SAID_PARTIAL_MATCH) - return SAID_FULL_MATCH; - else - return SAID_PARTIAL_MATCH; - } - - return SAID_NO_MATCH; -} - - -#ifdef SAID_DEBUG_PROGRAM -int main (int argc, char *argv) { - byte block[] = {0x01, 0x00, 0xf8, 0xf5, 0x02, 0x01, 0xf6, 0xf2, 0x02, 0x01, 0xf2, 0x01, 0x03, 0xff}; - EngineState s; - - s.parser_valid = 1; - said(&s, block); -} -#endif - -} // End of namespace Sci diff --git a/engines/sci/engine/state.h b/engines/sci/engine/state.h index ace1f019aa..17b635f8b9 100644 --- a/engines/sci/engine/state.h +++ b/engines/sci/engine/state.h @@ -36,11 +36,13 @@ namespace Common { } #include "sci/sci.h" -#include "sci/vocabulary.h" #include "sci/resource.h" #include "sci/engine/kernel.h" // for kfunct_sig_pair_t #include "sci/engine/script.h" #include "sci/engine/seg_manager.h" + +#include "sci/parser/vocabulary.h" + #include "sci/sound/audio.h" #ifdef USE_OLD_MUSIC_FUNCTIONS #include "sci/sound/iterator/core.h" diff --git a/engines/sci/module.mk b/engines/sci/module.mk index 10dc73818d..599f86d425 100644 --- a/engines/sci/module.mk +++ b/engines/sci/module.mk @@ -7,10 +7,8 @@ MODULE_OBJS := \ event.o \ resource.o \ sci.o \ - vocabulary.o \ engine/game.o \ engine/gc.o \ - engine/grammar.o \ engine/kernel.o \ engine/kevent.o \ engine/kfile.o \ @@ -20,6 +18,7 @@ MODULE_OBJS := \ engine/kmenu.o \ engine/kmisc.o \ engine/kmovement.o \ + engine/kparse.o \ engine/kpathing.o \ engine/kscripts.o \ engine/ksound.o \ @@ -51,6 +50,9 @@ MODULE_OBJS := \ graphics/transitions.o \ graphics/view.o \ graphics/windowmgr.o \ + parser/grammar.o \ + parser/said.o \ + parser/vocabulary.o \ sound/audio.o \ sound/midiparser_sci.o \ sound/music.o \ diff --git a/engines/sci/parser/grammar.cpp b/engines/sci/parser/grammar.cpp new file mode 100644 index 0000000000..9ee3b8aeaa --- /dev/null +++ b/engines/sci/parser/grammar.cpp @@ -0,0 +1,583 @@ +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +/* Functionality to transform the context-free SCI grammar rules into + * strict Greibach normal form (strict GNF), and to test SCI input against + * that grammar, writing an appropriate node tree if successful. + */ + +#include "sci/parser/vocabulary.h" +#include "sci/console.h" +#include "common/array.h" + +namespace Sci { + +#define TOKEN_OPAREN 0xff000000 +#define TOKEN_CPAREN 0xfe000000 +#define TOKEN_TERMINAL_CLASS 0x10000 +#define TOKEN_TERMINAL_GROUP 0x20000 +#define TOKEN_STUFFING_WORD 0x40000 +#define TOKEN_NON_NT (TOKEN_OPAREN | TOKEN_TERMINAL_CLASS | TOKEN_TERMINAL_GROUP | TOKEN_STUFFING_WORD) +#define TOKEN_TERMINAL (TOKEN_TERMINAL_CLASS | TOKEN_TERMINAL_GROUP) + +static int _allocd_rules = 0; // FIXME: Avoid non-const global vars + +struct ParseRule { + int _id; /**< non-terminal ID */ + uint _firstSpecial; /**< first terminal or non-terminal */ + uint _numSpecials; /**< number of terminals and non-terminals */ + Common::Array _data; /**< actual data */ + + ~ParseRule() { + assert(_allocd_rules > 0); + --_allocd_rules; + } + + // FIXME remove this one again? + bool operator==(const ParseRule &other) const { + return _id == other._id && + _firstSpecial == other._firstSpecial && + _numSpecials == other._numSpecials && + _data == other._data; + } +}; + + +struct ParseRuleList { + int terminal; /**< Terminal character this rule matches against or 0 for a non-terminal rule */ + ParseRule *rule; + ParseRuleList *next; + + void print() const; + + ParseRuleList(ParseRule *r) : rule(r), next(0) { + int term = rule->_data[rule->_firstSpecial]; + terminal = ((term & TOKEN_TERMINAL) ? term : 0); + } + + ~ParseRuleList() { + delete rule; + delete next; + } +}; + + +static void vocab_print_rule(ParseRule *rule) { + int wspace = 0; + + if (!rule) { + warning("NULL rule"); + return; + } + + printf("[%03x] -> ", rule->_id); + + if (rule->_data.empty()) + printf("e"); + + for (uint i = 0; i < rule->_data.size(); i++) { + uint token = rule->_data[i]; + + if (token == TOKEN_OPAREN) { + if (i == rule->_firstSpecial) + printf("_"); + + printf("("); + wspace = 0; + } else if (token == TOKEN_CPAREN) { + if (i == rule->_firstSpecial) + printf("_"); + + printf(")"); + wspace = 0; + } else { + if (wspace) + printf(" "); + + if (i == rule->_firstSpecial) + printf("_"); + if (token & TOKEN_TERMINAL_CLASS) + printf("C(%04x)", token & 0xffff); + else if (token & TOKEN_TERMINAL_GROUP) + printf("G(%04x)", token & 0xffff); + else if (token & TOKEN_STUFFING_WORD) + printf("%03x", token & 0xffff); + else + printf("[%03x]", token); /* non-terminal */ + wspace = 1; + } + + if (i == rule->_firstSpecial) + printf("_"); + } + printf(" [%d specials]", rule->_numSpecials); +} + +static ParseRule *_vdup(ParseRule *a) { + ++_allocd_rules; + return new ParseRule(*a); +} + +static ParseRule *_vinsert(ParseRule *turkey, ParseRule *stuffing) { + uint firstnt = turkey->_firstSpecial; + + // Search for first TOKEN_NON_NT in 'turkey' + while ((firstnt < turkey->_data.size()) && (turkey->_data[firstnt] & TOKEN_NON_NT)) + firstnt++; + + // If no TOKEN_NON_NT found, or if it doesn't match the id of 'stuffing', abort. + if ((firstnt == turkey->_data.size()) || (turkey->_data[firstnt] != stuffing->_id)) + return NULL; + + // Create a new rule as a copy of 'turkey', where the token firstnt has been substituted + // by the rule 'stuffing'. + ++_allocd_rules; + + ParseRule *rule = new ParseRule(*turkey); + rule->_numSpecials += stuffing->_numSpecials - 1; + rule->_firstSpecial = firstnt + stuffing->_firstSpecial; + rule->_data.resize(turkey->_data.size() - 1 + stuffing->_data.size()); + + // Replace rule->_data[firstnt] by all of stuffing->_data + Common::copy(stuffing->_data.begin(), stuffing->_data.end(), rule->_data.begin() + firstnt); + + if (firstnt < turkey->_data.size() - 1) + Common::copy(turkey->_data.begin() + firstnt + 1, turkey->_data.end(), + rule->_data.begin() + firstnt + stuffing->_data.size()); + + return rule; +} + +static ParseRule *_vbuild_rule(const parse_tree_branch_t *branch) { + int tokens = 0, tokenpos = 0, i; + + while (tokenpos < 10 && branch->data[tokenpos]) { + int type = branch->data[tokenpos]; + tokenpos += 2; + + if ((type == VOCAB_TREE_NODE_COMPARE_TYPE) || (type == VOCAB_TREE_NODE_COMPARE_GROUP) || (type == VOCAB_TREE_NODE_FORCE_STORAGE)) + ++tokens; + else if (type > VOCAB_TREE_NODE_LAST_WORD_STORAGE) + tokens += 5; + else + return NULL; // invalid + } + + ParseRule *rule = new ParseRule(); + + ++_allocd_rules; + rule->_id = branch->id; + rule->_numSpecials = tokenpos >> 1; + rule->_data.resize(tokens); + rule->_firstSpecial = 0; + + tokens = 0; + for (i = 0; i < tokenpos; i += 2) { + int type = branch->data[i]; + int value = branch->data[i + 1]; + + if (type == VOCAB_TREE_NODE_COMPARE_TYPE) + rule->_data[tokens++] = value | TOKEN_TERMINAL_CLASS; + else if (type == VOCAB_TREE_NODE_COMPARE_GROUP) + rule->_data[tokens++] = value | TOKEN_TERMINAL_GROUP; + else if (type == VOCAB_TREE_NODE_FORCE_STORAGE) + rule->_data[tokens++] = value | TOKEN_STUFFING_WORD; + else { // normal inductive rule + rule->_data[tokens++] = TOKEN_OPAREN; + rule->_data[tokens++] = type | TOKEN_STUFFING_WORD; + rule->_data[tokens++] = value | TOKEN_STUFFING_WORD; + + if (i == 0) + rule->_firstSpecial = tokens; + + rule->_data[tokens++] = value; // The non-terminal + rule->_data[tokens++] = TOKEN_CPAREN; + } + } + + return rule; +} + +static ParseRule *_vsatisfy_rule(ParseRule *rule, const ResultWord &input) { + int dep; + + if (!rule->_numSpecials) + return NULL; + + dep = rule->_data[rule->_firstSpecial]; + + if (((dep & TOKEN_TERMINAL_CLASS) && ((dep & 0xffff) & input._class)) || + ((dep & TOKEN_TERMINAL_GROUP) && ((dep & 0xffff) & input._group))) { + ParseRule *retval = new ParseRule(*rule); + ++_allocd_rules; + retval->_data[rule->_firstSpecial] = TOKEN_STUFFING_WORD | input._group; + retval->_numSpecials--; + retval->_firstSpecial = 0; + + if (retval->_numSpecials) { // find first special, if it exists + for (uint i = rule->_firstSpecial; i < retval->_data.size(); ++i) { + int tmp = retval->_data[i]; + if (!(tmp & TOKEN_NON_NT) || (tmp & TOKEN_TERMINAL)) { + retval->_firstSpecial = i; + break; + } + } + } + + return retval; + } else + return NULL; +} + +void Vocabulary::freeRuleList(ParseRuleList *list) { + delete list; +} + +static ParseRuleList *_vocab_add_rule(ParseRuleList *list, ParseRule *rule) { + if (!rule) + return list; + + ParseRuleList *new_elem = new ParseRuleList(rule); + + if (list) { + const int term = new_elem->terminal; +/* if (term < list->terminal) { + new_elem->next = list; + return new_elem; + } else {*/ + ParseRuleList *seeker = list; + + while (seeker->next/* && seeker->next->terminal <= term*/) { + if (seeker->next->terminal == term) { + if (*(seeker->next->rule) == *rule) { + delete rule; + delete new_elem; + return list; // No duplicate rules + } + } + seeker = seeker->next; + } + + new_elem->next = seeker->next; + seeker->next = new_elem; + return list; + } else { + return new_elem; + } +} + +void ParseRuleList::print() const { + const ParseRuleList *list = this; + int pos = 0; + while (list) { + printf("R%03d: ", pos); + vocab_print_rule(list->rule); + printf("\n"); + list = list->next; + pos++; + } + printf("%d rules total.\n", pos); +} + +static ParseRuleList *_vocab_split_rule_list(ParseRuleList *list) { + assert(list); + if (!list->next || (list->next->terminal)) { + ParseRuleList *tmp = list->next; + list->next = NULL; + return tmp; + } else + return _vocab_split_rule_list(list->next); +} + +static void _vocab_free_empty_rule_list(ParseRuleList *list) { + assert(list); + if (list->next) + _vocab_free_empty_rule_list(list->next); + list->next = 0; + list->rule = 0; + delete list; +} + +static ParseRuleList *_vocab_merge_rule_lists(ParseRuleList *l1, ParseRuleList *l2) { + ParseRuleList *retval = l1, *seeker = l2; + while (seeker) { + retval = _vocab_add_rule(retval, seeker->rule); + seeker = seeker->next; + } + _vocab_free_empty_rule_list(l2); + + return retval; +} + +static int _vocab_rule_list_length(ParseRuleList *list) { + return ((list) ? _vocab_rule_list_length(list->next) + 1 : 0); +} + +static ParseRuleList *_vocab_clone_rule_list_by_id(ParseRuleList *list, int id) { + ParseRuleList *result = NULL; + ParseRuleList *seeker = list; + + while (seeker) { + if (seeker->rule->_id == id) { + result = _vocab_add_rule(result, _vdup(seeker->rule)); + } + seeker = seeker->next; + } + + return result; +} + +ParseRuleList *Vocabulary::buildGNF(bool verbose) { + int iterations = 0; + int last_termrules, termrules = 0; + int ntrules_nr; + ParseRuleList *ntlist = NULL; + ParseRuleList *tlist, *new_tlist; + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + + for (uint i = 1; i < _parserBranches.size(); i++) { // branch rule 0 is treated specially + ParseRule *rule = _vbuild_rule(&_parserBranches[i]); + if (!rule) + return NULL; + ntlist = _vocab_add_rule(ntlist, rule); + } + + tlist = _vocab_split_rule_list(ntlist); + ntrules_nr = _vocab_rule_list_length(ntlist); + + if (verbose) + con->DebugPrintf("Starting with %d rules\n", ntrules_nr); + + new_tlist = tlist; + tlist = NULL; + + do { + ParseRuleList *new_new_tlist = NULL; + ParseRuleList *ntseeker, *tseeker; + last_termrules = termrules; + + ntseeker = ntlist; + while (ntseeker) { + tseeker = new_tlist; + + while (tseeker) { + ParseRule *newrule = _vinsert(ntseeker->rule, tseeker->rule); + if (newrule) + new_new_tlist = _vocab_add_rule(new_new_tlist, newrule); + tseeker = tseeker->next; + } + + ntseeker = ntseeker->next; + } + + tlist = _vocab_merge_rule_lists(tlist, new_tlist); + new_tlist = new_new_tlist; + termrules = _vocab_rule_list_length(new_new_tlist); + + if (verbose) + con->DebugPrintf("After iteration #%d: %d new term rules\n", ++iterations, termrules); + + } while (termrules && (iterations < 30)); + + freeRuleList(ntlist); + + if (verbose) { + con->DebugPrintf("\nGNF rules:\n"); + tlist->print(); + con->DebugPrintf("%d allocd rules\n", _allocd_rules); + con->DebugPrintf("Freeing rule list...\n"); + freeRuleList(tlist); + return NULL; + } + + return tlist; +} + +static int _vbpt_pareno(parse_tree_node_t *nodes, int *pos, int base) { + // Opens parentheses + nodes[base].content.branches[0] = (*pos) + 1; + nodes[++(*pos)].type = kParseTreeBranchNode; + nodes[*pos].content.branches[0] = 0; + nodes[*pos].content.branches[1] = 0; + return *pos; +} + +static int _vbpt_parenc(parse_tree_node_t *nodes, int *pos, int paren) { + // Closes parentheses for appending + nodes[paren].content.branches[1] = ++(*pos); + nodes[*pos].type = kParseTreeBranchNode; + nodes[*pos].content.branches[0] = 0; + nodes[*pos].content.branches[1] = 0; + return *pos; +} + +static int _vbpt_append(parse_tree_node_t *nodes, int *pos, int base, int value) { + // writes one value to an existing base node and creates a successor node for writing + nodes[base].content.branches[0] = ++(*pos); + nodes[*pos].type = kParseTreeLeafNode; + nodes[*pos].content.value = value; + nodes[base].content.branches[1] = ++(*pos); + nodes[*pos].type = kParseTreeBranchNode; + nodes[*pos].content.branches[0] = 0; + nodes[*pos].content.branches[1] = 0; + return *pos; +} + +static int _vbpt_terminate(parse_tree_node_t *nodes, int *pos, int base, int value) { + // Terminates, overwriting a nextwrite forknode + nodes[base].type = kParseTreeLeafNode; + nodes[base].content.value = value; + return *pos; +} + +static int _vbpt_write_subexpression(parse_tree_node_t *nodes, int *pos, ParseRule *rule, uint rulepos, int writepos) { + uint token; + + while ((token = ((rulepos < rule->_data.size()) ? rule->_data[rulepos++] : TOKEN_CPAREN)) != TOKEN_CPAREN) { + uint nexttoken = (rulepos < rule->_data.size()) ? rule->_data[rulepos] : TOKEN_CPAREN; + if (token == TOKEN_OPAREN) { + int writepos2 = _vbpt_pareno(nodes, pos, writepos); + rulepos = _vbpt_write_subexpression(nodes, pos, rule, rulepos, writepos2); + nexttoken = (rulepos < rule->_data.size()) ? rule->_data[rulepos] : TOKEN_CPAREN; + if (nexttoken != TOKEN_CPAREN) + writepos = _vbpt_parenc(nodes, pos, writepos); + } else if (token & TOKEN_STUFFING_WORD) { + if (nexttoken == TOKEN_CPAREN) + writepos = _vbpt_terminate(nodes, pos, writepos, token & 0xffff); + else + writepos = _vbpt_append(nodes, pos, writepos, token & 0xffff); + } else { + printf("\nError in parser (grammar.cpp, _vbpt_write_subexpression()): Rule data broken in rule "); + vocab_print_rule(rule); + printf(", at token position %d\n", *pos); + return rulepos; + } + } + + return rulepos; +} + +int Vocabulary::parseGNF(const ResultWordList &words, bool verbose) { + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + // Get the start rules: + ParseRuleList *work = _vocab_clone_rule_list_by_id(_parserRules, _parserBranches[0].data[1]); + ParseRuleList *results = NULL; + uint word = 0; + const uint words_nr = words.size(); + ResultWordList::const_iterator word_iter = words.begin(); + + for (word_iter = words.begin(); word_iter != words.end(); ++word_iter, ++word) { + ParseRuleList *new_work = NULL; + ParseRuleList *reduced_rules = NULL; + ParseRuleList *seeker, *subseeker; + + if (verbose) + con->DebugPrintf("Adding word %d...\n", word); + + seeker = work; + while (seeker) { + if (seeker->rule->_numSpecials <= (words_nr - word)) + reduced_rules = _vocab_add_rule(reduced_rules, _vsatisfy_rule(seeker->rule, *word_iter)); + + seeker = seeker->next; + } + + if (reduced_rules == NULL) { + freeRuleList(work); + if (verbose) + con->DebugPrintf("No results.\n"); + return 1; + } + + freeRuleList(work); + + if (word + 1 < words_nr) { + seeker = reduced_rules; + + while (seeker) { + if (seeker->rule->_numSpecials) { + int my_id = seeker->rule->_data[seeker->rule->_firstSpecial]; + + subseeker = _parserRules; + while (subseeker) { + if (subseeker->rule->_id == my_id) + new_work = _vocab_add_rule(new_work, _vinsert(seeker->rule, subseeker->rule)); + + subseeker = subseeker->next; + } + } + + seeker = seeker->next; + } + freeRuleList(reduced_rules); + } else // last word + new_work = reduced_rules; + + work = new_work; + if (verbose) + con->DebugPrintf("Now at %d candidates\n", _vocab_rule_list_length(work)); + if (work == NULL) { + if (verbose) + con->DebugPrintf("No results.\n"); + return 1; + } + } + + results = work; + + if (verbose) { + con->DebugPrintf("All results (excluding the surrounding '(141 %03x' and ')'):\n", _parserBranches[0].id); + results->print(); + con->DebugPrintf("\n"); + } + + // now use the first result + { + int temp, pos; + + _parserNodes[0].type = kParseTreeBranchNode; + _parserNodes[0].content.branches[0] = 1; + _parserNodes[0].content.branches[1] = 2; + + _parserNodes[1].type = kParseTreeLeafNode; + _parserNodes[1].content.value = 0x141; + + _parserNodes[2].type = kParseTreeBranchNode; + _parserNodes[2].content.branches[0] = 0; + _parserNodes[2].content.branches[1] = 0; + + pos = 2; + + temp = _vbpt_append(_parserNodes, &pos, 2, _parserBranches[0].id); + //_vbpt_write_subexpression(nodes, &pos, results[_vocab_rule_list_length(results)].rule, 0, temp); + _vbpt_write_subexpression(_parserNodes, &pos, results->rule, 0, temp); + } + + freeRuleList(results); + + return 0; +} + +} // End of namespace Sci diff --git a/engines/sci/parser/said.cpp b/engines/sci/parser/said.cpp new file mode 100644 index 0000000000..32bd1fcf8d --- /dev/null +++ b/engines/sci/parser/said.cpp @@ -0,0 +1,2483 @@ +/* A Bison parser, made by GNU Bison 2.3. */ + +/* Skeleton implementation for Bison's Yacc-like parsers in C + + Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003, 2004, 2005, 2006 + Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, + Boston, MA 02110-1301, USA. */ + +/* As a special exception, you may create a larger work that contains + part or all of the Bison parser skeleton and distribute that work + under terms of your choice, so long as that work isn't itself a + parser generator using the skeleton or a modified version thereof + as a parser skeleton. Alternatively, if you modify or redistribute + the parser skeleton itself, you may (at your option) remove this + special exception, which will cause the skeleton and the resulting + Bison output files to be licensed under the GNU General Public + License without this special exception. + + This special exception was added by the Free Software Foundation in + version 2.2 of Bison. */ + +/* C LALR(1) parser skeleton written by Richard Stallman, by + simplifying the original so-called "semantic" parser. */ + +/* All symbols defined below should begin with yy or YY, to avoid + infringing on user name space. This should be done even for local + variables, as they might otherwise be expanded by user macros. + There are some unavoidable exceptions within include files to + define necessary library symbols; they are noted "INFRINGES ON + USER NAME SPACE" below. */ + +/* Identify Bison output. */ +#define YYBISON 1 + +/* Bison version. */ +#define YYBISON_VERSION "2.3" + +/* Skeleton name. */ +#define YYSKELETON_NAME "yacc.c" + +/* Pure parsers. */ +#define YYPURE 0 + +/* Using locations. */ +#define YYLSP_NEEDED 0 + + + +/* Tokens. */ +#ifndef YYTOKENTYPE +# define YYTOKENTYPE + /* Put the tokens into the symbol table, so that GDB and other debuggers + know about them. */ + enum yytokentype { + WGROUP = 258, + YY_COMMA = 259, + YY_AMP = 260, + YY_SLASH = 261, + YY_PARENO = 262, + YY_PARENC = 263, + YY_BRACKETSO = 264, + YY_BRACKETSC = 265, + YY_HASH = 266, + YY_LT = 267, + YY_GT = 268, + YY_BRACKETSO_LT = 269, + YY_BRACKETSO_SLASH = 270, + YY_LT_BRACKETSO = 271, + YY_LT_PARENO = 272 + }; +#endif +/* Tokens. */ +#define WGROUP 258 +#define YY_COMMA 259 +#define YY_AMP 260 +#define YY_SLASH 261 +#define YY_PARENO 262 +#define YY_PARENC 263 +#define YY_BRACKETSO 264 +#define YY_BRACKETSC 265 +#define YY_HASH 266 +#define YY_LT 267 +#define YY_GT 268 +#define YY_BRACKETSO_LT 269 +#define YY_BRACKETSO_SLASH 270 +#define YY_LT_BRACKETSO 271 +#define YY_LT_PARENO 272 + + + + +/* Copy the first part of user declarations. */ +#line 1 "said.y" + +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +#include "sci/engine/state.h" + + +// Bison generates an empty switch statement that gives a warning in MSVC. +// This disables that warning. +#ifdef _MSC_VER +#pragma warning(disable:4065) +#endif + + +namespace Sci { + +#define SAID_BRANCH_NULL 0 + +#define MAX_SAID_TOKENS 128 + +// Maximum number of words to be expected in a parsed sentence +#define AUGMENT_MAX_WORDS 64 + + +#define ANYWORD 0xfff + +#define WORD_TYPE_BASE 0x141 +#define WORD_TYPE_REF 0x144 +#define WORD_TYPE_SYNTACTIC_SUGAR 0x145 + +#define AUGMENT_SENTENCE_PART_BRACKETS 0x152 + +// Minor numbers +#define AUGMENT_SENTENCE_MINOR_MATCH_PHRASE 0x14c +#define AUGMENT_SENTENCE_MINOR_MATCH_WORD 0x153 +#define AUGMENT_SENTENCE_MINOR_RECURSE 0x144 +#define AUGMENT_SENTENCE_MINOR_PARENTHESES 0x14f + + +#undef YYDEBUG /*1*/ +//#define SAID_DEBUG*/ +//#define SCI_DEBUG_PARSE_TREE_AUGMENTATION // uncomment to debug parse tree augmentation + + +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION +#define scidprintf printf +#else +void print_nothing(...) { } +#define scidprintf print_nothing +#endif + + +static char *said_parse_error; + +static int said_token; +static int said_tokens_nr; +static int said_tokens[MAX_SAID_TOKENS]; +static int said_blessed; // increminated by said_top_branch + +static int said_tree_pos; // Set to 0 if we're out of space +#define SAID_TREE_START 4; // Reserve space for the 4 top nodes + +#define VALUE_IGNORE -424242 + +static parse_tree_node_t said_tree[VOCAB_TREE_NODES]; + +typedef int wgroup_t; +typedef int tree_t; +typedef int said_spec_t; + +static tree_t said_aug_branch(int, int, tree_t, tree_t); +static tree_t said_attach_branch(tree_t, tree_t); +/* +static tree_t said_wgroup_branch(wgroup_t); +*/ +static said_spec_t said_top_branch(tree_t); +static tree_t said_paren(tree_t, tree_t); +static tree_t said_value(int, tree_t); +static tree_t said_terminal(int); + +static int yylex(void); + +static int yyerror(const char *s) { + said_parse_error = strdup(s); + return 1; /* Abort */ +} + + + +/* Enabling traces. */ +#ifndef YYDEBUG +# define YYDEBUG 0 +#endif + +/* Enabling verbose error messages. */ +#ifdef YYERROR_VERBOSE +# undef YYERROR_VERBOSE +# define YYERROR_VERBOSE 1 +#else +# define YYERROR_VERBOSE 0 +#endif + +/* Enabling the token table. */ +#ifndef YYTOKEN_TABLE +# define YYTOKEN_TABLE 0 +#endif + +#if ! defined YYSTYPE && ! defined YYSTYPE_IS_DECLARED +typedef int YYSTYPE; +# define yystype YYSTYPE /* obsolescent; will be withdrawn */ +# define YYSTYPE_IS_DECLARED 1 +# define YYSTYPE_IS_TRIVIAL 1 +#endif + + + +/* Copy the second part of user declarations. */ + + +/* Line 216 of yacc.c. */ +#line 250 "said.tab.c" + +#ifdef short +# undef short +#endif + +#ifdef YYTYPE_UINT8 +typedef YYTYPE_UINT8 yytype_uint8; +#else +typedef unsigned char yytype_uint8; +#endif + +#ifdef YYTYPE_INT8 +typedef YYTYPE_INT8 yytype_int8; +#elif (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +typedef signed char yytype_int8; +#else +typedef short int yytype_int8; +#endif + +#ifdef YYTYPE_UINT16 +typedef YYTYPE_UINT16 yytype_uint16; +#else +typedef unsigned short int yytype_uint16; +#endif + +#ifdef YYTYPE_INT16 +typedef YYTYPE_INT16 yytype_int16; +#else +typedef short int yytype_int16; +#endif + +#ifndef YYSIZE_T +# ifdef __SIZE_TYPE__ +# define YYSIZE_T __SIZE_TYPE__ +# elif defined size_t +# define YYSIZE_T size_t +# elif ! defined YYSIZE_T && (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +# include /* INFRINGES ON USER NAME SPACE */ +# define YYSIZE_T size_t +# else +# define YYSIZE_T unsigned int +# endif +#endif + +#define YYSIZE_MAXIMUM ((YYSIZE_T) -1) + +#ifndef YY_ +# if YYENABLE_NLS +# if ENABLE_NLS +# include /* INFRINGES ON USER NAME SPACE */ +# define YY_(msgid) dgettext ("bison-runtime", msgid) +# endif +# endif +# ifndef YY_ +# define YY_(msgid) msgid +# endif +#endif + +/* Suppress unused-variable warnings by "using" E. */ +#if ! defined lint || defined __GNUC__ +# define YYUSE(e) ((void) (e)) +#else +# define YYUSE(e) /* empty */ +#endif + +/* Identity function, used to suppress warnings about constant conditions. */ +#ifndef lint +# define YYID(n) (n) +#else +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static int +YYID (int i) +#else +static int +YYID (i) + int i; +#endif +{ + return i; +} +#endif + +#if ! defined yyoverflow || YYERROR_VERBOSE + +/* The parser invokes alloca or malloc; define the necessary symbols. */ + +# ifdef YYSTACK_USE_ALLOCA +# if YYSTACK_USE_ALLOCA +# ifdef __GNUC__ +# define YYSTACK_ALLOC __builtin_alloca +# elif defined __BUILTIN_VA_ARG_INCR +# include /* INFRINGES ON USER NAME SPACE */ +# elif defined _AIX +# define YYSTACK_ALLOC __alloca +# elif defined _MSC_VER +# include /* INFRINGES ON USER NAME SPACE */ +# define alloca _alloca +# else +# define YYSTACK_ALLOC alloca +# if ! defined _ALLOCA_H && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +# include /* INFRINGES ON USER NAME SPACE */ +# ifndef _STDLIB_H +# define _STDLIB_H 1 +# endif +# endif +# endif +# endif +# endif + +# ifdef YYSTACK_ALLOC + /* Pacify GCC's `empty if-body' warning. */ +# define YYSTACK_FREE(Ptr) do { /* empty */; } while (YYID (0)) +# ifndef YYSTACK_ALLOC_MAXIMUM + /* The OS might guarantee only one guard page at the bottom of the stack, + and a page size can be as small as 4096 bytes. So we cannot safely + invoke alloca (N) if N exceeds 4096. Use a slightly smaller number + to allow for a few compiler-allocated temporary stack slots. */ +# define YYSTACK_ALLOC_MAXIMUM 4032 /* reasonable circa 2006 */ +# endif +# else +# define YYSTACK_ALLOC YYMALLOC +# define YYSTACK_FREE YYFREE +# ifndef YYSTACK_ALLOC_MAXIMUM +# define YYSTACK_ALLOC_MAXIMUM YYSIZE_MAXIMUM +# endif +# if (defined __cplusplus && ! defined _STDLIB_H \ + && ! ((defined YYMALLOC || defined malloc) \ + && (defined YYFREE || defined free))) +# include /* INFRINGES ON USER NAME SPACE */ +# ifndef _STDLIB_H +# define _STDLIB_H 1 +# endif +# endif +# ifndef YYMALLOC +# define YYMALLOC malloc +# if ! defined malloc && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +void *malloc (YYSIZE_T); /* INFRINGES ON USER NAME SPACE */ +# endif +# endif +# ifndef YYFREE +# define YYFREE free +# if ! defined free && ! defined _STDLIB_H && (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +void free (void *); /* INFRINGES ON USER NAME SPACE */ +# endif +# endif +# endif +#endif /* ! defined yyoverflow || YYERROR_VERBOSE */ + + +#if (! defined yyoverflow \ + && (! defined __cplusplus \ + || (defined YYSTYPE_IS_TRIVIAL && YYSTYPE_IS_TRIVIAL))) + +/* A type that is properly aligned for any stack member. */ +union yyalloc +{ + yytype_int16 yyss; + YYSTYPE yyvs; + }; + +/* The size of the maximum gap between one aligned stack and the next. */ +# define YYSTACK_GAP_MAXIMUM (sizeof (union yyalloc) - 1) + +/* The size of an array large to enough to hold all stacks, each with + N elements. */ +# define YYSTACK_BYTES(N) \ + ((N) * (sizeof (yytype_int16) + sizeof (YYSTYPE)) \ + + YYSTACK_GAP_MAXIMUM) + +/* Copy COUNT objects from FROM to TO. The source and destination do + not overlap. */ +# ifndef YYCOPY +# if defined __GNUC__ && 1 < __GNUC__ +# define YYCOPY(To, From, Count) \ + __builtin_memcpy (To, From, (Count) * sizeof (*(From))) +# else +# define YYCOPY(To, From, Count) \ + do \ + { \ + YYSIZE_T yyi; \ + for (yyi = 0; yyi < (Count); yyi++) \ + (To)[yyi] = (From)[yyi]; \ + } \ + while (YYID (0)) +# endif +# endif + +/* Relocate STACK from its old location to the new one. The + local variables YYSIZE and YYSTACKSIZE give the old and new number of + elements in the stack, and YYPTR gives the new location of the + stack. Advance YYPTR to a properly aligned location for the next + stack. */ +# define YYSTACK_RELOCATE(Stack) \ + do \ + { \ + YYSIZE_T yynewbytes; \ + YYCOPY (&yyptr->Stack, Stack, yysize); \ + Stack = &yyptr->Stack; \ + yynewbytes = yystacksize * sizeof (*Stack) + YYSTACK_GAP_MAXIMUM; \ + yyptr += yynewbytes / sizeof (*yyptr); \ + } \ + while (YYID (0)) + +#endif + +/* YYFINAL -- State number of the termination state. */ +#define YYFINAL 23 +/* YYLAST -- Last index in YYTABLE. */ +#define YYLAST 80 + +/* YYNTOKENS -- Number of terminals. */ +#define YYNTOKENS 18 +/* YYNNTS -- Number of nonterminals. */ +#define YYNNTS 13 +/* YYNRULES -- Number of rules. */ +#define YYNRULES 35 +/* YYNRULES -- Number of states. */ +#define YYNSTATES 69 + +/* YYTRANSLATE(YYLEX) -- Bison symbol number corresponding to YYLEX. */ +#define YYUNDEFTOK 2 +#define YYMAXUTOK 272 + +#define YYTRANSLATE(YYX) \ + ((unsigned int) (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYUNDEFTOK) + +/* YYTRANSLATE[YYLEX] -- Bison symbol number corresponding to YYLEX. */ +static const yytype_uint8 yytranslate[] = +{ + 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 1, 2, 3, 4, + 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, + 15, 16, 17 +}; + +#if YYDEBUG +/* YYPRHS[YYN] -- Index of the first RHS symbol of rule number YYN in + YYRHS. */ +static const yytype_uint8 yyprhs[] = +{ + 0, 0, 3, 6, 10, 15, 16, 18, 19, 21, + 24, 29, 31, 34, 39, 41, 43, 45, 49, 51, + 55, 59, 64, 70, 73, 75, 77, 79, 83, 88, + 92, 97, 100, 105, 109, 112 +}; + +/* YYRHS -- A `-1'-separated list of the rules' RHS. */ +static const yytype_int8 yyrhs[] = +{ + 19, 0, -1, 21, 20, -1, 21, 22, 20, -1, + 21, 22, 23, 20, -1, -1, 13, -1, -1, 27, + -1, 6, 27, -1, 15, 6, 27, 10, -1, 6, + -1, 6, 27, -1, 15, 6, 27, 10, -1, 6, + -1, 3, -1, 26, -1, 9, 26, 10, -1, 24, + -1, 7, 27, 8, -1, 26, 4, 26, -1, 26, + 14, 29, 10, -1, 26, 4, 9, 26, 10, -1, + 25, 28, -1, 25, -1, 28, -1, 29, -1, 14, + 29, 10, -1, 29, 14, 29, 10, -1, 12, 24, + 30, -1, 17, 7, 27, 8, -1, 12, 26, -1, + 16, 9, 26, 10, -1, 12, 26, 30, -1, 12, + 26, -1, 17, 7, 27, 8, -1 +}; + +/* YYRLINE[YYN] -- source line where rule number YYN was defined. */ +static const yytype_uint8 yyrline[] = +{ + 0, 129, 129, 131, 133, 139, 140, 147, 148, 154, + 156, 158, 164, 166, 168, 173, 178, 180, 185, 187, + 189, 191, 193, 198, 200, 202, 207, 209, 211, 216, + 218, 220, 222, 227, 229, 231 +}; +#endif + +#if YYDEBUG || YYERROR_VERBOSE || YYTOKEN_TABLE +/* YYTNAME[SYMBOL-NUM] -- String name of the symbol SYMBOL-NUM. + First, the terminals, then, starting at YYNTOKENS, nonterminals. */ +static const char *const yytname[] = +{ + "$end", "error", "$undefined", "WGROUP", "YY_COMMA", "YY_AMP", + "YY_SLASH", "YY_PARENO", "YY_PARENC", "YY_BRACKETSO", "YY_BRACKETSC", + "YY_HASH", "YY_LT", "YY_GT", "YY_BRACKETSO_LT", "YY_BRACKETSO_SLASH", + "YY_LT_BRACKETSO", "YY_LT_PARENO", "$accept", "saidspec", "optcont", + "leftspec", "midspec", "rightspec", "word", "cwordset", "wordset", + "expr", "cwordrefset", "wordrefset", "recref", 0 +}; +#endif + +# ifdef YYPRINT +/* YYTOKNUM[YYLEX-NUM] -- Internal token number corresponding to + token YYLEX-NUM. */ +static const yytype_uint16 yytoknum[] = +{ + 0, 256, 257, 258, 259, 260, 261, 262, 263, 264, + 265, 266, 267, 268, 269, 270, 271, 272 +}; +# endif + +/* YYR1[YYN] -- Symbol number of symbol that rule YYN derives. */ +static const yytype_uint8 yyr1[] = +{ + 0, 18, 19, 19, 19, 20, 20, 21, 21, 22, + 22, 22, 23, 23, 23, 24, 25, 25, 26, 26, + 26, 26, 26, 27, 27, 27, 28, 28, 28, 29, + 29, 29, 29, 30, 30, 30 +}; + +/* YYR2[YYN] -- Number of symbols composing right hand side of rule YYN. */ +static const yytype_uint8 yyr2[] = +{ + 0, 2, 2, 3, 4, 0, 1, 0, 1, 2, + 4, 1, 2, 4, 1, 1, 1, 3, 1, 3, + 3, 4, 5, 2, 1, 1, 1, 3, 4, 3, + 4, 2, 4, 3, 2, 4 +}; + +/* YYDEFACT[STATE-NAME] -- Default rule to reduce with in state + STATE-NUM when YYTABLE doesn't specify something else to do. Zero + means the default is an error. */ +static const yytype_uint8 yydefact[] = +{ + 7, 15, 0, 0, 0, 0, 0, 0, 0, 5, + 18, 24, 16, 8, 25, 26, 0, 0, 18, 31, + 0, 0, 0, 1, 11, 6, 0, 2, 5, 23, + 0, 0, 0, 19, 17, 0, 0, 29, 27, 0, + 0, 9, 0, 14, 0, 3, 5, 0, 20, 0, + 0, 34, 0, 32, 30, 0, 12, 0, 4, 0, + 21, 28, 33, 0, 10, 0, 22, 35, 13 +}; + +/* YYDEFGOTO[NTERM-NUM]. */ +static const yytype_int8 yydefgoto[] = +{ + -1, 8, 27, 9, 28, 46, 10, 11, 12, 13, + 14, 15, 37 +}; + +/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing + STATE-NUM. */ +#define YYPACT_NINF -24 +static const yytype_int8 yypact[] = +{ + -1, -24, -1, 62, 62, 54, 1, 5, 18, 38, + -24, 47, 3, -24, -24, 12, 23, 15, -3, 3, + 28, 62, -1, -24, -1, -24, 42, -24, 39, -24, + 53, 54, 54, -24, -24, 62, 50, -24, -24, 29, + 41, -24, -1, -1, 52, -24, 55, 62, 3, 57, + 63, 20, -1, -24, -24, 64, -24, -1, -24, 32, + -24, -24, -24, 67, -24, 66, -24, -24, -24 +}; + +/* YYPGOTO[NTERM-NUM]. */ +static const yytype_int8 yypgoto[] = +{ + -24, -24, -23, -24, -24, -24, 68, -24, 0, -2, + 69, -4, 26 +}; + +/* YYTABLE[YYPACT[STATE-NUM]]. What to do in state STATE-NUM. If + positive, shift that token. If negative, reduce the rule which + number is the opposite. If zero, do what YYDEFACT says. + If YYTABLE_NINF, syntax error. */ +#define YYTABLE_NINF -1 +static const yytype_uint8 yytable[] = +{ + 16, 20, 1, 17, 19, 45, 2, 30, 3, 35, + 21, 4, 22, 5, 36, 6, 7, 31, 23, 30, + 40, 39, 41, 58, 30, 34, 32, 49, 50, 31, + 48, 33, 35, 30, 31, 51, 30, 36, 38, 53, + 55, 56, 66, 31, 24, 43, 31, 59, 42, 54, + 63, 25, 25, 26, 44, 65, 1, 52, 57, 4, + 2, 5, 47, 6, 7, 1, 4, 60, 25, 2, + 6, 7, 18, 61, 64, 67, 68, 62, 0, 0, + 29 +}; + +static const yytype_int8 yycheck[] = +{ + 2, 5, 3, 3, 4, 28, 7, 4, 9, 12, + 9, 12, 7, 14, 17, 16, 17, 14, 0, 4, + 22, 21, 24, 46, 4, 10, 14, 31, 32, 14, + 30, 8, 12, 4, 14, 35, 4, 17, 10, 10, + 42, 43, 10, 14, 6, 6, 14, 47, 6, 8, + 52, 13, 13, 15, 15, 57, 3, 7, 6, 12, + 7, 14, 9, 16, 17, 3, 12, 10, 13, 7, + 16, 17, 4, 10, 10, 8, 10, 51, -1, -1, + 11 +}; + +/* YYSTOS[STATE-NUM] -- The (internal number of the) accessing + symbol of state STATE-NUM. */ +static const yytype_uint8 yystos[] = +{ + 0, 3, 7, 9, 12, 14, 16, 17, 19, 21, + 24, 25, 26, 27, 28, 29, 27, 26, 24, 26, + 29, 9, 7, 0, 6, 13, 15, 20, 22, 28, + 4, 14, 14, 8, 10, 12, 17, 30, 10, 26, + 27, 27, 6, 6, 15, 20, 23, 9, 26, 29, + 29, 26, 7, 10, 8, 27, 27, 6, 20, 26, + 10, 10, 30, 27, 10, 27, 10, 8, 10 +}; + +#define yyerrok (yyerrstatus = 0) +#define yyclearin (yychar = YYEMPTY) +#define YYEMPTY (-2) +#define YYEOF 0 + +#define YYACCEPT goto yyacceptlab +#define YYABORT goto yyabortlab +#define YYERROR goto yyerrorlab + + +/* Like YYERROR except do call yyerror. This remains here temporarily + to ease the transition to the new meaning of YYERROR, for GCC. + Once GCC version 2 has supplanted version 1, this can go. */ + +#define YYFAIL goto yyerrlab + +#define YYRECOVERING() (!!yyerrstatus) + +#define YYBACKUP(Token, Value) \ +do \ + if (yychar == YYEMPTY && yylen == 1) \ + { \ + yychar = (Token); \ + yylval = (Value); \ + yytoken = YYTRANSLATE (yychar); \ + YYPOPSTACK (1); \ + goto yybackup; \ + } \ + else \ + { \ + yyerror (YY_("syntax error: cannot back up")); \ + YYERROR; \ + } \ +while (YYID (0)) + + +#define YYTERROR 1 +#define YYERRCODE 256 + + +/* YYLLOC_DEFAULT -- Set CURRENT to span from RHS[1] to RHS[N]. + If N is 0, then set CURRENT to the empty location which ends + the previous symbol: RHS[0] (always defined). */ + +#define YYRHSLOC(Rhs, K) ((Rhs)[K]) +#ifndef YYLLOC_DEFAULT +# define YYLLOC_DEFAULT(Current, Rhs, N) \ + do \ + if (YYID (N)) \ + { \ + (Current).first_line = YYRHSLOC (Rhs, 1).first_line; \ + (Current).first_column = YYRHSLOC (Rhs, 1).first_column; \ + (Current).last_line = YYRHSLOC (Rhs, N).last_line; \ + (Current).last_column = YYRHSLOC (Rhs, N).last_column; \ + } \ + else \ + { \ + (Current).first_line = (Current).last_line = \ + YYRHSLOC (Rhs, 0).last_line; \ + (Current).first_column = (Current).last_column = \ + YYRHSLOC (Rhs, 0).last_column; \ + } \ + while (YYID (0)) +#endif + + +/* YY_LOCATION_PRINT -- Print the location on the stream. + This macro was not mandated originally: define only if we know + we won't break user code: when these are the locations we know. */ + +#ifndef YY_LOCATION_PRINT +# if YYLTYPE_IS_TRIVIAL +# define YY_LOCATION_PRINT(File, Loc) \ + fprintf (File, "%d.%d-%d.%d", \ + (Loc).first_line, (Loc).first_column, \ + (Loc).last_line, (Loc).last_column) +# else +# define YY_LOCATION_PRINT(File, Loc) ((void) 0) +# endif +#endif + + +/* YYLEX -- calling `yylex' with the right arguments. */ + +#ifdef YYLEX_PARAM +# define YYLEX yylex (YYLEX_PARAM) +#else +# define YYLEX yylex () +#endif + +/* Enable debugging if requested. */ +#if YYDEBUG + +# ifndef YYFPRINTF +# include /* INFRINGES ON USER NAME SPACE */ +# define YYFPRINTF fprintf +# endif + +# define YYDPRINTF(Args) \ +do { \ + if (yydebug) \ + YYFPRINTF Args; \ +} while (YYID (0)) + +# define YY_SYMBOL_PRINT(Title, Type, Value, Location) \ +do { \ + if (yydebug) \ + { \ + YYFPRINTF (stderr, "%s ", Title); \ + yy_symbol_print (stderr, \ + Type, Value); \ + YYFPRINTF (stderr, "\n"); \ + } \ +} while (YYID (0)) + + +/*--------------------------------. +| Print this symbol on YYOUTPUT. | +`--------------------------------*/ + +/*ARGSUSED*/ +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static void +yy_symbol_value_print (FILE *yyoutput, int yytype, YYSTYPE const * const yyvaluep) +#else +static void +yy_symbol_value_print (yyoutput, yytype, yyvaluep) + FILE *yyoutput; + int yytype; + YYSTYPE const * const yyvaluep; +#endif +{ + if (!yyvaluep) + return; +# ifdef YYPRINT + if (yytype < YYNTOKENS) + YYPRINT (yyoutput, yytoknum[yytype], *yyvaluep); +# else + YYUSE (yyoutput); +# endif + switch (yytype) + { + default: + break; + } +} + + +/*--------------------------------. +| Print this symbol on YYOUTPUT. | +`--------------------------------*/ + +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static void +yy_symbol_print (FILE *yyoutput, int yytype, YYSTYPE const * const yyvaluep) +#else +static void +yy_symbol_print (yyoutput, yytype, yyvaluep) + FILE *yyoutput; + int yytype; + YYSTYPE const * const yyvaluep; +#endif +{ + if (yytype < YYNTOKENS) + YYFPRINTF (yyoutput, "token %s (", yytname[yytype]); + else + YYFPRINTF (yyoutput, "nterm %s (", yytname[yytype]); + + yy_symbol_value_print (yyoutput, yytype, yyvaluep); + YYFPRINTF (yyoutput, ")"); +} + +/*------------------------------------------------------------------. +| yy_stack_print -- Print the state stack from its BOTTOM up to its | +| TOP (included). | +`------------------------------------------------------------------*/ + +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static void +yy_stack_print (yytype_int16 *bottom, yytype_int16 *top) +#else +static void +yy_stack_print (bottom, top) + yytype_int16 *bottom; + yytype_int16 *top; +#endif +{ + YYFPRINTF (stderr, "Stack now"); + for (; bottom <= top; ++bottom) + YYFPRINTF (stderr, " %d", *bottom); + YYFPRINTF (stderr, "\n"); +} + +# define YY_STACK_PRINT(Bottom, Top) \ +do { \ + if (yydebug) \ + yy_stack_print ((Bottom), (Top)); \ +} while (YYID (0)) + + +/*------------------------------------------------. +| Report that the YYRULE is going to be reduced. | +`------------------------------------------------*/ + +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static void +yy_reduce_print (YYSTYPE *yyvsp, int yyrule) +#else +static void +yy_reduce_print (yyvsp, yyrule) + YYSTYPE *yyvsp; + int yyrule; +#endif +{ + int yynrhs = yyr2[yyrule]; + int yyi; + unsigned long int yylno = yyrline[yyrule]; + YYFPRINTF (stderr, "Reducing stack by rule %d (line %lu):\n", + yyrule - 1, yylno); + /* The symbols being reduced. */ + for (yyi = 0; yyi < yynrhs; yyi++) + { + fprintf (stderr, " $%d = ", yyi + 1); + yy_symbol_print (stderr, yyrhs[yyprhs[yyrule] + yyi], + &(yyvsp[(yyi + 1) - (yynrhs)]) + ); + fprintf (stderr, "\n"); + } +} + +# define YY_REDUCE_PRINT(Rule) \ +do { \ + if (yydebug) \ + yy_reduce_print (yyvsp, Rule); \ +} while (YYID (0)) + +/* Nonzero means print parse trace. It is left uninitialized so that + multiple parsers can coexist. */ +int yydebug; +#else /* !YYDEBUG */ +# define YYDPRINTF(Args) +# define YY_SYMBOL_PRINT(Title, Type, Value, Location) +# define YY_STACK_PRINT(Bottom, Top) +# define YY_REDUCE_PRINT(Rule) +#endif /* !YYDEBUG */ + + +/* YYINITDEPTH -- initial size of the parser's stacks. */ +#ifndef YYINITDEPTH +# define YYINITDEPTH 200 +#endif + +/* YYMAXDEPTH -- maximum size the stacks can grow to (effective only + if the built-in stack extension method is used). + + Do not make this value too large; the results are undefined if + YYSTACK_ALLOC_MAXIMUM < YYSTACK_BYTES (YYMAXDEPTH) + evaluated with infinite-precision integer arithmetic. */ + +#ifndef YYMAXDEPTH +# define YYMAXDEPTH 10000 +#endif + + + +#if YYERROR_VERBOSE + +# ifndef yystrlen +# if defined __GLIBC__ && defined _STRING_H +# define yystrlen strlen +# else +/* Return the length of YYSTR. */ +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static YYSIZE_T +yystrlen (const char *yystr) +#else +static YYSIZE_T +yystrlen (yystr) + const char *yystr; +#endif +{ + YYSIZE_T yylen; + for (yylen = 0; yystr[yylen]; yylen++) + continue; + return yylen; +} +# endif +# endif + +# ifndef yystpcpy +# if defined __GLIBC__ && defined _STRING_H && defined _GNU_SOURCE +# define yystpcpy stpcpy +# else +/* Copy YYSRC to YYDEST, returning the address of the terminating '\0' in + YYDEST. */ +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static char * +yystpcpy (char *yydest, const char *yysrc) +#else +static char * +yystpcpy (yydest, yysrc) + char *yydest; + const char *yysrc; +#endif +{ + char *yyd = yydest; + const char *yys = yysrc; + + while ((*yyd++ = *yys++) != '\0') + continue; + + return yyd - 1; +} +# endif +# endif + +# ifndef yytnamerr +/* Copy to YYRES the contents of YYSTR after stripping away unnecessary + quotes and backslashes, so that it's suitable for yyerror. The + heuristic is that double-quoting is unnecessary unless the string + contains an apostrophe, a comma, or backslash (other than + backslash-backslash). YYSTR is taken from yytname. If YYRES is + null, do not copy; instead, return the length of what the result + would have been. */ +static YYSIZE_T +yytnamerr (char *yyres, const char *yystr) +{ + if (*yystr == '"') + { + YYSIZE_T yyn = 0; + char const *yyp = yystr; + + for (;;) + switch (*++yyp) + { + case '\'': + case ',': + goto do_not_strip_quotes; + + case '\\': + if (*++yyp != '\\') + goto do_not_strip_quotes; + /* Fall through. */ + default: + if (yyres) + yyres[yyn] = *yyp; + yyn++; + break; + + case '"': + if (yyres) + yyres[yyn] = '\0'; + return yyn; + } + do_not_strip_quotes: ; + } + + if (! yyres) + return yystrlen (yystr); + + return yystpcpy (yyres, yystr) - yyres; +} +# endif + +/* Copy into YYRESULT an error message about the unexpected token + YYCHAR while in state YYSTATE. Return the number of bytes copied, + including the terminating null byte. If YYRESULT is null, do not + copy anything; just return the number of bytes that would be + copied. As a special case, return 0 if an ordinary "syntax error" + message will do. Return YYSIZE_MAXIMUM if overflow occurs during + size calculation. */ +static YYSIZE_T +yysyntax_error (char *yyresult, int yystate, int yychar) +{ + int yyn = yypact[yystate]; + + if (! (YYPACT_NINF < yyn && yyn <= YYLAST)) + return 0; + else + { + int yytype = YYTRANSLATE (yychar); + YYSIZE_T yysize0 = yytnamerr (0, yytname[yytype]); + YYSIZE_T yysize = yysize0; + YYSIZE_T yysize1; + int yysize_overflow = 0; + enum { YYERROR_VERBOSE_ARGS_MAXIMUM = 5 }; + char const *yyarg[YYERROR_VERBOSE_ARGS_MAXIMUM]; + int yyx; + +# if 0 + /* This is so xgettext sees the translatable formats that are + constructed on the fly. */ + YY_("syntax error, unexpected %s"); + YY_("syntax error, unexpected %s, expecting %s"); + YY_("syntax error, unexpected %s, expecting %s or %s"); + YY_("syntax error, unexpected %s, expecting %s or %s or %s"); + YY_("syntax error, unexpected %s, expecting %s or %s or %s or %s"); +# endif + char *yyfmt; + char const *yyf; + static char const yyunexpected[] = "syntax error, unexpected %s"; + static char const yyexpecting[] = ", expecting %s"; + static char const yyor[] = " or %s"; + char yyformat[sizeof yyunexpected + + sizeof yyexpecting - 1 + + ((YYERROR_VERBOSE_ARGS_MAXIMUM - 2) + * (sizeof yyor - 1))]; + char const *yyprefix = yyexpecting; + + /* Start YYX at -YYN if negative to avoid negative indexes in + YYCHECK. */ + int yyxbegin = yyn < 0 ? -yyn : 0; + + /* Stay within bounds of both yycheck and yytname. */ + int yychecklim = YYLAST - yyn + 1; + int yyxend = yychecklim < YYNTOKENS ? yychecklim : YYNTOKENS; + int yycount = 1; + + yyarg[0] = yytname[yytype]; + yyfmt = yystpcpy (yyformat, yyunexpected); + + for (yyx = yyxbegin; yyx < yyxend; ++yyx) + if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR) + { + if (yycount == YYERROR_VERBOSE_ARGS_MAXIMUM) + { + yycount = 1; + yysize = yysize0; + yyformat[sizeof yyunexpected - 1] = '\0'; + break; + } + yyarg[yycount++] = yytname[yyx]; + yysize1 = yysize + yytnamerr (0, yytname[yyx]); + yysize_overflow |= (yysize1 < yysize); + yysize = yysize1; + yyfmt = yystpcpy (yyfmt, yyprefix); + yyprefix = yyor; + } + + yyf = YY_(yyformat); + yysize1 = yysize + yystrlen (yyf); + yysize_overflow |= (yysize1 < yysize); + yysize = yysize1; + + if (yysize_overflow) + return YYSIZE_MAXIMUM; + + if (yyresult) + { + /* Avoid sprintf, as that infringes on the user's name space. + Don't have undefined behavior even if the translation + produced a string with the wrong number of "%s"s. */ + char *yyp = yyresult; + int yyi = 0; + while ((*yyp = *yyf) != '\0') + { + if (*yyp == '%' && yyf[1] == 's' && yyi < yycount) + { + yyp += yytnamerr (yyp, yyarg[yyi++]); + yyf += 2; + } + else + { + yyp++; + yyf++; + } + } + } + return yysize; + } +} +#endif /* YYERROR_VERBOSE */ + + +/*-----------------------------------------------. +| Release the memory associated to this symbol. | +`-----------------------------------------------*/ + +/*ARGSUSED*/ +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +static void +yydestruct (const char *yymsg, int yytype, YYSTYPE *yyvaluep) +#else +static void +yydestruct (yymsg, yytype, yyvaluep) + const char *yymsg; + int yytype; + YYSTYPE *yyvaluep; +#endif +{ + YYUSE (yyvaluep); + + if (!yymsg) + yymsg = "Deleting"; + YY_SYMBOL_PRINT (yymsg, yytype, yyvaluep, yylocationp); + + switch (yytype) + { + + default: + break; + } +} + + +/* Prevent warnings from -Wmissing-prototypes. */ + +#ifdef YYPARSE_PARAM +#if defined __STDC__ || defined __cplusplus +int yyparse (void *YYPARSE_PARAM); +#else +int yyparse (); +#endif +#else /* ! YYPARSE_PARAM */ +#if defined __STDC__ || defined __cplusplus +int yyparse (void); +#else +int yyparse (); +#endif +#endif /* ! YYPARSE_PARAM */ + + + +/* The look-ahead symbol. */ +int yychar; + +/* The semantic value of the look-ahead symbol. */ +YYSTYPE yylval; + +/* Number of syntax errors so far. */ +int yynerrs; + + + +/*----------. +| yyparse. | +`----------*/ + +#ifdef YYPARSE_PARAM +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +int +yyparse (void *YYPARSE_PARAM) +#else +int +yyparse (YYPARSE_PARAM) + void *YYPARSE_PARAM; +#endif +#else /* ! YYPARSE_PARAM */ +#if (defined __STDC__ || defined __C99__FUNC__ \ + || defined __cplusplus || defined _MSC_VER) +int +yyparse (void) +#else +int +yyparse () + +#endif +#endif +{ + + int yystate; + int yyn; + int yyresult; + /* Number of tokens to shift before error messages enabled. */ + int yyerrstatus; + /* Look-ahead token as an internal (translated) token number. */ + int yytoken = 0; +#if YYERROR_VERBOSE + /* Buffer for error messages, and its allocated size. */ + char yymsgbuf[128]; + char *yymsg = yymsgbuf; + YYSIZE_T yymsg_alloc = sizeof yymsgbuf; +#endif + + /* Three stacks and their tools: + `yyss': related to states, + `yyvs': related to semantic values, + `yyls': related to locations. + + Refer to the stacks thru separate pointers, to allow yyoverflow + to reallocate them elsewhere. */ + + /* The state stack. */ + yytype_int16 yyssa[YYINITDEPTH]; + yytype_int16 *yyss = yyssa; + yytype_int16 *yyssp; + + /* The semantic value stack. */ + YYSTYPE yyvsa[YYINITDEPTH]; + YYSTYPE *yyvs = yyvsa; + YYSTYPE *yyvsp; + + + +#define YYPOPSTACK(N) (yyvsp -= (N), yyssp -= (N)) + + YYSIZE_T yystacksize = YYINITDEPTH; + + /* The variables used to return semantic value and location from the + action routines. */ + YYSTYPE yyval; + + + /* The number of symbols on the RHS of the reduced rule. + Keep to zero when no symbol should be popped. */ + int yylen = 0; + + YYDPRINTF ((stderr, "Starting parse\n")); + + yystate = 0; + yyerrstatus = 0; + yynerrs = 0; + yychar = YYEMPTY; /* Cause a token to be read. */ + + /* Initialize stack pointers. + Waste one element of value and location stack + so that they stay on the same level as the state stack. + The wasted elements are never initialized. */ + + yyssp = yyss; + yyvsp = yyvs; + + goto yysetstate; + +/*------------------------------------------------------------. +| yynewstate -- Push a new state, which is found in yystate. | +`------------------------------------------------------------*/ + yynewstate: + /* In all cases, when you get here, the value and location stacks + have just been pushed. So pushing a state here evens the stacks. */ + yyssp++; + + yysetstate: + *yyssp = yystate; + + if (yyss + yystacksize - 1 <= yyssp) + { + /* Get the current used size of the three stacks, in elements. */ + YYSIZE_T yysize = yyssp - yyss + 1; + +#ifdef yyoverflow + { + /* Give user a chance to reallocate the stack. Use copies of + these so that the &'s don't force the real ones into + memory. */ + YYSTYPE *yyvs1 = yyvs; + yytype_int16 *yyss1 = yyss; + + + /* Each stack pointer address is followed by the size of the + data in use in that stack, in bytes. This used to be a + conditional around just the two extra args, but that might + be undefined if yyoverflow is a macro. */ + yyoverflow (YY_("memory exhausted"), + &yyss1, yysize * sizeof (*yyssp), + &yyvs1, yysize * sizeof (*yyvsp), + + &yystacksize); + + yyss = yyss1; + yyvs = yyvs1; + } +#else /* no yyoverflow */ +# ifndef YYSTACK_RELOCATE + goto yyexhaustedlab; +# else + /* Extend the stack our own way. */ + if (YYMAXDEPTH <= yystacksize) + goto yyexhaustedlab; + yystacksize *= 2; + if (YYMAXDEPTH < yystacksize) + yystacksize = YYMAXDEPTH; + + { + yytype_int16 *yyss1 = yyss; + union yyalloc *yyptr = + (union yyalloc *) YYSTACK_ALLOC (YYSTACK_BYTES (yystacksize)); + if (! yyptr) + goto yyexhaustedlab; + YYSTACK_RELOCATE (yyss); + YYSTACK_RELOCATE (yyvs); + +# undef YYSTACK_RELOCATE + if (yyss1 != yyssa) + YYSTACK_FREE (yyss1); + } +# endif +#endif /* no yyoverflow */ + + yyssp = yyss + yysize - 1; + yyvsp = yyvs + yysize - 1; + + + YYDPRINTF ((stderr, "Stack size increased to %lu\n", + (unsigned long int) yystacksize)); + + if (yyss + yystacksize - 1 <= yyssp) + YYABORT; + } + + YYDPRINTF ((stderr, "Entering state %d\n", yystate)); + + goto yybackup; + +/*-----------. +| yybackup. | +`-----------*/ +yybackup: + + /* Do appropriate processing given the current state. Read a + look-ahead token if we need one and don't already have one. */ + + /* First try to decide what to do without reference to look-ahead token. */ + yyn = yypact[yystate]; + if (yyn == YYPACT_NINF) + goto yydefault; + + /* Not known => get a look-ahead token if don't already have one. */ + + /* YYCHAR is either YYEMPTY or YYEOF or a valid look-ahead symbol. */ + if (yychar == YYEMPTY) + { + YYDPRINTF ((stderr, "Reading a token: ")); + yychar = YYLEX; + } + + if (yychar <= YYEOF) + { + yychar = yytoken = YYEOF; + YYDPRINTF ((stderr, "Now at end of input.\n")); + } + else + { + yytoken = YYTRANSLATE (yychar); + YY_SYMBOL_PRINT ("Next token is", yytoken, &yylval, &yylloc); + } + + /* If the proper action on seeing token YYTOKEN is to reduce or to + detect an error, take that action. */ + yyn += yytoken; + if (yyn < 0 || YYLAST < yyn || yycheck[yyn] != yytoken) + goto yydefault; + yyn = yytable[yyn]; + if (yyn <= 0) + { + if (yyn == 0 || yyn == YYTABLE_NINF) + goto yyerrlab; + yyn = -yyn; + goto yyreduce; + } + + if (yyn == YYFINAL) + YYACCEPT; + + /* Count tokens shifted since error; after three, turn off error + status. */ + if (yyerrstatus) + yyerrstatus--; + + /* Shift the look-ahead token. */ + YY_SYMBOL_PRINT ("Shifting", yytoken, &yylval, &yylloc); + + /* Discard the shifted token unless it is eof. */ + if (yychar != YYEOF) + yychar = YYEMPTY; + + yystate = yyn; + *++yyvsp = yylval; + + goto yynewstate; + + +/*-----------------------------------------------------------. +| yydefault -- do the default action for the current state. | +`-----------------------------------------------------------*/ +yydefault: + yyn = yydefact[yystate]; + if (yyn == 0) + goto yyerrlab; + goto yyreduce; + + +/*-----------------------------. +| yyreduce -- Do a reduction. | +`-----------------------------*/ +yyreduce: + /* yyn is the number of a rule to reduce with. */ + yylen = yyr2[yyn]; + + /* If YYLEN is nonzero, implement the default value of the action: + `$$ = $1'. + + Otherwise, the following line sets YYVAL to garbage. + This behavior is undocumented and Bison + users should not rely upon it. Assigning to YYVAL + unconditionally makes the parser a bit smaller, and it avoids a + GCC warning that YYVAL may be used uninitialized. */ + yyval = yyvsp[1-yylen]; + + + YY_REDUCE_PRINT (yyn); + switch (yyn) + { + case 2: +#line 130 "said.y" + { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (2)]), (yyvsp[(2) - (2)]))); ;} + break; + + case 3: +#line 132 "said.y" + { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (3)]), said_attach_branch((yyvsp[(2) - (3)]), (yyvsp[(3) - (3)])))); ;} + break; + + case 4: +#line 134 "said.y" + { (yyval) = said_top_branch(said_attach_branch((yyvsp[(1) - (4)]), said_attach_branch((yyvsp[(2) - (4)]), said_attach_branch((yyvsp[(3) - (4)]), (yyvsp[(4) - (4)]))))); ;} + break; + + case 5: +#line 139 "said.y" + { (yyval) = SAID_BRANCH_NULL; ;} + break; + + case 6: +#line 141 "said.y" + { (yyval) = said_paren(said_value(0x14b, said_value(0xf900, said_terminal(0xf900))), SAID_BRANCH_NULL); ;} + break; + + case 7: +#line 147 "said.y" + { (yyval) = SAID_BRANCH_NULL; ;} + break; + + case 8: +#line 149 "said.y" + { (yyval) = said_paren(said_value(0x141, said_value(0x149, (yyvsp[(1) - (1)]))), SAID_BRANCH_NULL); ;} + break; + + case 9: +#line 155 "said.y" + { (yyval) = said_aug_branch(0x142, 0x14a, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} + break; + + case 10: +#line 157 "said.y" + { (yyval) = said_aug_branch(0x152, 0x142, said_aug_branch(0x142, 0x14a, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 11: +#line 159 "said.y" + { (yyval) = SAID_BRANCH_NULL; ;} + break; + + case 12: +#line 165 "said.y" + { (yyval) = said_aug_branch(0x143, 0x14a, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} + break; + + case 13: +#line 167 "said.y" + { (yyval) = said_aug_branch(0x152, 0x143, said_aug_branch(0x143, 0x14a, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 14: +#line 169 "said.y" + { (yyval) = SAID_BRANCH_NULL; ;} + break; + + case 15: +#line 174 "said.y" + { (yyval) = said_paren(said_value(0x141, said_value(0x153, said_terminal((yyvsp[(1) - (1)])))), SAID_BRANCH_NULL); ;} + break; + + case 16: +#line 179 "said.y" + { (yyval) = said_aug_branch(0x141, 0x14f, (yyvsp[(1) - (1)]), SAID_BRANCH_NULL); ;} + break; + + case 17: +#line 181 "said.y" + { (yyval) = said_aug_branch(0x141, 0x14f, said_aug_branch(0x152, 0x14c, said_aug_branch(0x141, 0x14f, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 18: +#line 186 "said.y" + { (yyval) = (yyvsp[(1) - (1)]); ;} + break; + + case 19: +#line 188 "said.y" + { (yyval) = said_aug_branch(0x141, 0x14c, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL); ;} + break; + + case 20: +#line 190 "said.y" + { (yyval) = said_attach_branch((yyvsp[(1) - (3)]), (yyvsp[(3) - (3)])); ;} + break; + + case 21: +#line 192 "said.y" + { (yyval) = said_attach_branch((yyvsp[(1) - (4)]), (yyvsp[(3) - (4)])); ;} + break; + + case 22: +#line 194 "said.y" + { (yyval) = said_attach_branch((yyvsp[(1) - (5)]), (yyvsp[(3) - (5)])); ;} + break; + + case 23: +#line 199 "said.y" + { (yyval) = said_attach_branch((yyvsp[(1) - (2)]), (yyvsp[(2) - (2)])); ;} + break; + + case 24: +#line 201 "said.y" + { (yyval) = (yyvsp[(1) - (1)]); ;} + break; + + case 25: +#line 203 "said.y" + { (yyval) = (yyvsp[(1) - (1)]); ;} + break; + + case 26: +#line 208 "said.y" + { (yyval) = (yyvsp[(1) - (1)]); ;} + break; + + case 27: +#line 210 "said.y" + { (yyval) = said_aug_branch(0x152, 0x144, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL); ;} + break; + + case 28: +#line 212 "said.y" + { (yyval) = said_attach_branch((yyvsp[(1) - (4)]), said_aug_branch(0x152, 0x144, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL)); ;} + break; + + case 29: +#line 217 "said.y" + { (yyval) = said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (3)]), (yyvsp[(3) - (3)])); ;} + break; + + case 30: +#line 219 "said.y" + { (yyval) = said_aug_branch(0x144, 0x14f, said_aug_branch(0x141, 0x144, (yyvsp[(2) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 31: +#line 221 "said.y" + { (yyval) = said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL); ;} + break; + + case 32: +#line 223 "said.y" + { (yyval) = said_aug_branch(0x152, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(3) - (4)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 33: +#line 228 "said.y" + { (yyval) = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (3)]), SAID_BRANCH_NULL), (yyvsp[(3) - (3)])); ;} + break; + + case 34: +#line 230 "said.y" + { (yyval) = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, (yyvsp[(2) - (2)]), SAID_BRANCH_NULL), SAID_BRANCH_NULL); ;} + break; + + case 35: +#line 232 "said.y" + { (yyval) = said_aug_branch(0x141, 0x14c, (yyvsp[(2) - (4)]), SAID_BRANCH_NULL); ;} + break; + + +/* Line 1267 of yacc.c. */ +#line 1665 "said.tab.c" + default: break; + } + YY_SYMBOL_PRINT ("-> $$ =", yyr1[yyn], &yyval, &yyloc); + + YYPOPSTACK (yylen); + yylen = 0; + YY_STACK_PRINT (yyss, yyssp); + + *++yyvsp = yyval; + + + /* Now `shift' the result of the reduction. Determine what state + that goes to, based on the state we popped back to and the rule + number reduced by. */ + + yyn = yyr1[yyn]; + + yystate = yypgoto[yyn - YYNTOKENS] + *yyssp; + if (0 <= yystate && yystate <= YYLAST && yycheck[yystate] == *yyssp) + yystate = yytable[yystate]; + else + yystate = yydefgoto[yyn - YYNTOKENS]; + + goto yynewstate; + + +/*------------------------------------. +| yyerrlab -- here on detecting error | +`------------------------------------*/ +yyerrlab: + /* If not already recovering from an error, report this error. */ + if (!yyerrstatus) + { + ++yynerrs; +#if ! YYERROR_VERBOSE + yyerror (YY_("syntax error")); +#else + { + YYSIZE_T yysize = yysyntax_error (0, yystate, yychar); + if (yymsg_alloc < yysize && yymsg_alloc < YYSTACK_ALLOC_MAXIMUM) + { + YYSIZE_T yyalloc = 2 * yysize; + if (! (yysize <= yyalloc && yyalloc <= YYSTACK_ALLOC_MAXIMUM)) + yyalloc = YYSTACK_ALLOC_MAXIMUM; + if (yymsg != yymsgbuf) + YYSTACK_FREE (yymsg); + yymsg = (char *) YYSTACK_ALLOC (yyalloc); + if (yymsg) + yymsg_alloc = yyalloc; + else + { + yymsg = yymsgbuf; + yymsg_alloc = sizeof yymsgbuf; + } + } + + if (0 < yysize && yysize <= yymsg_alloc) + { + (void) yysyntax_error (yymsg, yystate, yychar); + yyerror (yymsg); + } + else + { + yyerror (YY_("syntax error")); + if (yysize != 0) + goto yyexhaustedlab; + } + } +#endif + } + + + + if (yyerrstatus == 3) + { + /* If just tried and failed to reuse look-ahead token after an + error, discard it. */ + + if (yychar <= YYEOF) + { + /* Return failure if at end of input. */ + if (yychar == YYEOF) + YYABORT; + } + else + { + yydestruct ("Error: discarding", + yytoken, &yylval); + yychar = YYEMPTY; + } + } + + /* Else will try to reuse look-ahead token after shifting the error + token. */ + goto yyerrlab1; + + +/*---------------------------------------------------. +| yyerrorlab -- error raised explicitly by YYERROR. | +`---------------------------------------------------*/ +yyerrorlab: + + /* Pacify compilers like GCC when the user code never invokes + YYERROR and the label yyerrorlab therefore never appears in user + code. */ + if (/*CONSTCOND*/ 0) + goto yyerrorlab; + + /* Do not reclaim the symbols of the rule which action triggered + this YYERROR. */ + YYPOPSTACK (yylen); + yylen = 0; + YY_STACK_PRINT (yyss, yyssp); + yystate = *yyssp; + goto yyerrlab1; + + +/*-------------------------------------------------------------. +| yyerrlab1 -- common code for both syntax error and YYERROR. | +`-------------------------------------------------------------*/ +yyerrlab1: + yyerrstatus = 3; /* Each real token shifted decrements this. */ + + for (;;) + { + yyn = yypact[yystate]; + if (yyn != YYPACT_NINF) + { + yyn += YYTERROR; + if (0 <= yyn && yyn <= YYLAST && yycheck[yyn] == YYTERROR) + { + yyn = yytable[yyn]; + if (0 < yyn) + break; + } + } + + /* Pop the current state because it cannot handle the error token. */ + if (yyssp == yyss) + YYABORT; + + + yydestruct ("Error: popping", + yystos[yystate], yyvsp); + YYPOPSTACK (1); + yystate = *yyssp; + YY_STACK_PRINT (yyss, yyssp); + } + + if (yyn == YYFINAL) + YYACCEPT; + + *++yyvsp = yylval; + + + /* Shift the error token. */ + YY_SYMBOL_PRINT ("Shifting", yystos[yyn], yyvsp, yylsp); + + yystate = yyn; + goto yynewstate; + + +/*-------------------------------------. +| yyacceptlab -- YYACCEPT comes here. | +`-------------------------------------*/ +yyacceptlab: + yyresult = 0; + goto yyreturn; + +/*-----------------------------------. +| yyabortlab -- YYABORT comes here. | +`-----------------------------------*/ +yyabortlab: + yyresult = 1; + goto yyreturn; + +#ifndef yyoverflow +/*-------------------------------------------------. +| yyexhaustedlab -- memory exhaustion comes here. | +`-------------------------------------------------*/ +yyexhaustedlab: + yyerror (YY_("memory exhausted")); + yyresult = 2; + /* Fall through. */ +#endif + +yyreturn: + if (yychar != YYEOF && yychar != YYEMPTY) + yydestruct ("Cleanup: discarding lookahead", + yytoken, &yylval); + /* Do not reclaim the symbols of the rule which action triggered + this YYABORT or YYACCEPT. */ + YYPOPSTACK (yylen); + YY_STACK_PRINT (yyss, yyssp); + while (yyssp != yyss) + { + yydestruct ("Cleanup: popping", + yystos[*yyssp], yyvsp); + YYPOPSTACK (1); + } +#ifndef yyoverflow + if (yyss != yyssa) + YYSTACK_FREE (yyss); +#endif +#if YYERROR_VERBOSE + if (yymsg != yymsgbuf) + YYSTACK_FREE (yymsg); +#endif + /* Make sure YYID is used. */ + return YYID (yyresult); +} + + +#line 235 "said.y" + + +int parse_yy_token_lookup[] = {YY_COMMA, YY_AMP, YY_SLASH, YY_PARENO, YY_PARENC, YY_BRACKETSO, YY_BRACKETSC, YY_HASH, YY_LT, YY_GT}; + +static int yylex(void) { + int retval = said_tokens[said_token++]; + + if (retval < SAID_LONG(SAID_FIRST)) { + yylval = retval; + retval = WGROUP; + } else { + retval >>= 8; + + if (retval == SAID_TERM) + retval = 0; + else { + assert(retval >= SAID_FIRST); + retval = parse_yy_token_lookup[retval - SAID_FIRST]; + if (retval == YY_BRACKETSO) { + if ((said_tokens[said_token] >> 8) == SAID_LT) + retval = YY_BRACKETSO_LT; + else + if ((said_tokens[said_token] >> 8) == SAID_SLASH) + retval = YY_BRACKETSO_SLASH; + } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_BRACKO) { + retval = YY_LT_BRACKETSO; + } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_PARENO) { + retval = YY_LT_PARENO; + } + } + } + + return retval; +} + +static int said_next_node() { + return ((said_tree_pos == 0) || (said_tree_pos >= VOCAB_TREE_NODES)) ? said_tree_pos = 0 : said_tree_pos++; +} + +#define SAID_NEXT_NODE said_next_node() + +static int said_leaf_node(tree_t pos, int value) { + said_tree[pos].type = kParseTreeLeafNode; + + if (value != VALUE_IGNORE) + said_tree[pos].content.value = value; + + return pos; +} + +static int said_branch_node(tree_t pos, int left, int right) { + said_tree[pos].type = kParseTreeBranchNode; + + if (left != VALUE_IGNORE) + said_tree[pos].content.branches[0] = left; + + if (right != VALUE_IGNORE) + said_tree[pos].content.branches[1] = right; + + return pos; +} + +static tree_t said_paren(tree_t t1, tree_t t2) { + if (t1) + return said_branch_node(SAID_NEXT_NODE, t1, t2); + else + return t2; +} + +static tree_t said_value(int val, tree_t t) { + return said_branch_node(SAID_NEXT_NODE, said_leaf_node(SAID_NEXT_NODE, val), t); + +} + +static tree_t said_terminal(int val) { + return said_leaf_node(SAID_NEXT_NODE, val); +} + +static tree_t said_aug_branch(int n1, int n2, tree_t t1, tree_t t2) { + int retval; + + retval = said_branch_node(SAID_NEXT_NODE, + said_branch_node(SAID_NEXT_NODE, + said_leaf_node(SAID_NEXT_NODE, n1), + said_branch_node(SAID_NEXT_NODE, + said_leaf_node(SAID_NEXT_NODE, n2), + t1) + ), + t2); + +#ifdef SAID_DEBUG + fprintf(stderr, "AUG(0x%x, 0x%x, [%04x], [%04x]) = [%04x]\n", n1, n2, t1, t2, retval); +#endif + + return retval; +} + +static tree_t said_attach_branch(tree_t base, tree_t attacheant) { +#ifdef SAID_DEBUG + fprintf(stderr, "ATT2([%04x], [%04x]) = [%04x]\n", base, attacheant, base); +#endif + + if (!attacheant) + return base; + if (!base) + return attacheant; + + if (!base) + return 0; // Happens if we're out of space + + said_branch_node(base, VALUE_IGNORE, attacheant); + + return base; +} + +static said_spec_t said_top_branch(tree_t first) { +#ifdef SAID_DEBUG + fprintf(stderr, "TOP([%04x])\n", first); +#endif + said_branch_node(0, 1, 2); + said_leaf_node(1, 0x141); // Magic number #1 + said_branch_node(2, 3, first); + said_leaf_node(3, 0x13f); // Magic number #2 + + ++said_blessed; + + return 0; +} + +static int said_parse_spec(byte *spec) { + int nextitem; + + said_parse_error = NULL; + said_token = 0; + said_tokens_nr = 0; + said_blessed = 0; + + said_tree_pos = SAID_TREE_START; + + do { + nextitem = *spec++; + if (nextitem < SAID_FIRST) + said_tokens[said_tokens_nr++] = nextitem << 8 | *spec++; + else + said_tokens[said_tokens_nr++] = SAID_LONG(nextitem); + + } while ((nextitem != SAID_TERM) && (said_tokens_nr < MAX_SAID_TOKENS)); + + if (nextitem == SAID_TERM) + yyparse(); + else { + warning("SAID spec is too long"); + return 1; + } + + if (said_parse_error) { + warning("Error while parsing SAID spec: %s", said_parse_error); + free(said_parse_error); + return 1; + } + + if (said_tree_pos == 0) { + warning("Out of tree space while parsing SAID spec"); + return 1; + } + + if (said_blessed != 1) { + warning("Found multiple top branches"); + return 1; + } + + return 0; +} + +/**********************/ +/**** Augmentation ****/ +/**********************/ + +// primitive functions + +#define AUG_READ_BRANCH(a, br, p) \ + if (tree[p].type != kParseTreeBranchNode) \ + return 0; \ + a = tree[p].content.branches[br]; + +#define AUG_READ_VALUE(a, p) \ + if (tree[p].type != kParseTreeLeafNode) \ + return 0; \ + a = tree[p].content.value; + +#define AUG_ASSERT(i) \ + if (!i) return 0; + +static int aug_get_next_sibling(parse_tree_node_t *tree, int pos, int *first, int *second) { + // Returns the next sibling relative to the specified position in 'tree', + // sets *first and *second to its augment node values, returns the new position + // or 0 if there was no next sibling + int seek, valpos; + + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(seek, 0, pos); + AUG_ASSERT(seek); + + // Now retrieve first value + AUG_READ_BRANCH(valpos, 0, seek); + AUG_ASSERT(valpos); + AUG_READ_VALUE(*first, valpos); + + // Get second value + AUG_READ_BRANCH(seek, 1, seek); + AUG_ASSERT(seek); + AUG_READ_BRANCH(valpos, 0, seek); + AUG_ASSERT(valpos); + AUG_READ_VALUE(*second, valpos); + + return pos; +} + +static int aug_get_wgroup(parse_tree_node_t *tree, int pos) { + // Returns 0 if pos in tree is not the root of a 3-element list, otherwise + // it returns the last element (which, in practice, is the word group + int val; + + AUG_READ_BRANCH(pos, 0, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_VALUE(val, pos); + + return val; +} + +static int aug_get_base_node(parse_tree_node_t *tree) { + int startpos = 0; + AUG_READ_BRANCH(startpos, 1, startpos); + + return startpos; +} + +// semi-primitive functions + +static int aug_get_first_child(parse_tree_node_t *tree, int pos, int *first, int *second) { + // like aug_get_next_sibling, except that it recurses into the tree and + // finds the first child (usually *not* Ayanami Rei) of the current branch + // rather than its next sibling. + AUG_READ_BRANCH(pos, 0, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + + return aug_get_next_sibling(tree, pos, first, second); +} + +static void aug_find_words_recursively(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, + int *ref_words, int *ref_words_nr, int maxwords, int refbranch) { + // Finds and lists all base (141) and reference (144) words */ + int major, minor; + int word; + int pos = aug_get_first_child(tree, startpos, &major, &minor); + + //if (major == WORD_TYPE_REF) + // refbranch = 1; + + while (pos) { + if ((word = aug_get_wgroup(tree, pos))) { // found a word + if (!refbranch && major == WORD_TYPE_BASE) { + if ((*base_words_nr) == maxwords) { + warning("Out of regular words"); + return; // return gracefully + } + + base_words[*base_words_nr] = word; // register word + ++(*base_words_nr); + + } + if (major == WORD_TYPE_REF || refbranch) { + if ((*ref_words_nr) == maxwords) { + warning("Out of reference words"); + return; // return gracefully + } + + ref_words[*ref_words_nr] = word; // register word + ++(*ref_words_nr); + + } + if (major != WORD_TYPE_SYNTACTIC_SUGAR && major != WORD_TYPE_BASE && major != WORD_TYPE_REF) + warning("aug_find_words_recursively(): Unknown word type %03x", major); + + } else // Did NOT find a word group: Attempt to recurse + aug_find_words_recursively(tree, pos, base_words, base_words_nr, + ref_words, ref_words_nr, maxwords, refbranch || major == WORD_TYPE_REF); + + pos = aug_get_next_sibling(tree, pos, &major, &minor); + } +} + + +static void aug_find_words(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, + int *ref_words, int *ref_words_nr, int maxwords) { + // initializing wrapper for aug_find_words_recursively() + *base_words_nr = 0; + *ref_words_nr = 0; + + aug_find_words_recursively(tree, startpos, base_words, base_words_nr, ref_words, ref_words_nr, maxwords, 0); +} + + +static int aug_contains_word(int *list, int length, int word) { + int i; + + if (word == ANYWORD) + return (length); + + for (i = 0; i < length; i++) + if (list[i] == word) + return 1; + + return 0; +} + + +static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_branch, int major, int minor, int *base_words, int base_words_nr, + int *ref_words, int ref_words_nr); + +static int augment_match_expression_p(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_basepos, int major, int minor, + int *base_words, int base_words_nr, int *ref_words, int ref_words_nr) { + int cmajor, cminor, cpos; + cpos = aug_get_first_child(saidt, augment_pos, &cmajor, &cminor); + if (!cpos) { + warning("augment_match_expression_p(): Empty condition"); + return 1; + } + + scidprintf("Attempting to match (%03x %03x (%03x %03x\n", major, minor, cmajor, cminor); + + if ((major == WORD_TYPE_BASE) && (minor == AUGMENT_SENTENCE_MINOR_RECURSE)) + return augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr); + + switch (major) { + + case WORD_TYPE_BASE: + while (cpos) { + if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { + int word = aug_get_wgroup(saidt, cpos); + scidprintf("Looking for word %03x\n", word); + + if (aug_contains_word(base_words, base_words_nr, word)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { + if (augment_sentence_expression(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { + int gc_major, gc_minor; + int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); + + while (gchild) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, + minor, base_words, base_words_nr, + ref_words, ref_words_nr)) + return 1; + gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); + } + } else + warning("augment_match_expression_p(): Unknown type 141 minor number %3x", cminor); + + cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); + + } + break; + + case WORD_TYPE_REF: + while (cpos) { + if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { + int word = aug_get_wgroup(saidt, cpos); + scidprintf("Looking for refword %03x\n", word); + + if (aug_contains_word(ref_words, ref_words_nr, word)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { + int gc_major, gc_minor; + int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); + + while (gchild) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, + minor, base_words, base_words_nr, + ref_words, ref_words_nr)) + return 1; + gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); + } + } else + warning("augment_match_expression_p(): Unknown type 144 minor number %3x", cminor); + + cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); + + } + break; + + case AUGMENT_SENTENCE_PART_BRACKETS: + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + + scidprintf("Didn't match subexpression; checking sub-bracked predicate %03x\n", cmajor); + + switch (cmajor) { + case WORD_TYPE_BASE: + if (!base_words_nr) + return 1; + break; + + case WORD_TYPE_REF: + if (!ref_words_nr) + return 1; + break; + + default: + warning("augment_match_expression_p(): (subp1) Unkonwn sub-bracket predicate %03x", cmajor); + } + + break; + + default: + warning("augment_match_expression_p(): Unknown predicate %03x", major); + + } + + scidprintf("augment_match_expression_p(): Generic failure\n"); + + return 0; +} + +static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_branch, int major, int minor, int *base_words, int base_words_nr, + int *ref_words, int ref_words_nr) { + int check_major, check_minor; + int check_pos = aug_get_first_child(saidt, augment_pos, &check_major, &check_minor); + do { + if (!(augment_match_expression_p(saidt, check_pos, parset, parse_branch, check_major, check_minor, + base_words, base_words_nr, ref_words, ref_words_nr))) + return 0; + } while ((check_pos = aug_get_next_sibling(saidt, check_pos, &check_major, &check_minor))); + + return 1; +} + +static int augment_sentence_part(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, int parse_basepos, int major, int minor) { + int pmajor, pminor; + int parse_branch = parse_basepos; + int optional = 0; + int foundwords = 0; + + scidprintf("Augmenting (%03x %03x\n", major, minor); + + if (major == AUGMENT_SENTENCE_PART_BRACKETS) { // '[/ foo]' is true if '/foo' or if there + // exists no x for which '/x' is true + if ((augment_pos = aug_get_first_child(saidt, augment_pos, &major, &minor))) { + scidprintf("Optional part: Now augmenting (%03x %03x\n", major, minor); + optional = 1; + } else { + scidprintf("Matched empty optional expression\n"); + return 1; + } + } + + if ((major < 0x141) || (major > 0x143)) { + scidprintf("augment_sentence_part(): Unexpected sentence part major number %03x\n", major); + return 0; + } + + while ((parse_branch = aug_get_next_sibling(parset, parse_branch, &pmajor, &pminor))) { + if (pmajor == major) { // found matching sentence part + int success; + int base_words_nr; + int ref_words_nr; + int base_words[AUGMENT_MAX_WORDS]; + int ref_words[AUGMENT_MAX_WORDS]; +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION + int i; +#endif + + scidprintf("Found match with pminor = %03x\n", pminor); + aug_find_words(parset, parse_branch, base_words, &base_words_nr, ref_words, &ref_words_nr, AUGMENT_MAX_WORDS); + foundwords |= (ref_words_nr | base_words_nr); +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION + printf("%d base words:", base_words_nr); + for (i = 0; i < base_words_nr; i++) + printf(" %03x", base_words[i]); + printf("\n%d reference words:", ref_words_nr); + for (i = 0; i < ref_words_nr; i++) + printf(" %03x", ref_words[i]); + printf("\n"); +#endif + + success = augment_sentence_expression(saidt, augment_pos, parset, parse_basepos, major, minor, + base_words, base_words_nr, ref_words, ref_words_nr); + + if (success) { + scidprintf("SUCCESS on augmenting (%03x %03x\n", major, minor); + return 1; + } + } + } + + if (optional && (foundwords == 0)) { + scidprintf("Found no words and optional branch => SUCCESS on augmenting (%03x %03x\n", major, minor); + return 1; + } + scidprintf("FAILURE on augmenting (%03x %03x\n", major, minor); + + return 0; +} + +static int augment_parse_nodes(parse_tree_node_t *parset, parse_tree_node_t *saidt) { + int augment_basepos = 0; + int parse_basepos; + int major, minor; + int dontclaim = 0; + + parse_basepos = aug_get_base_node(parset); + if (!parse_basepos) { + warning("augment_parse_nodes(): Parse tree is corrupt"); + return 0; + } + + augment_basepos = aug_get_base_node(saidt); + if (!augment_basepos) { + warning("augment_parse_nodes(): Said tree is corrupt"); + return 0; + } + + while ((augment_basepos = aug_get_next_sibling(saidt, augment_basepos, &major, &minor))) { + if ((major == 0x14b) && (minor == SAID_LONG(SAID_GT))) + dontclaim = 1; // special case + else // normal sentence part + if (!(augment_sentence_part(saidt, augment_basepos, parset, parse_basepos, major, minor))) { + scidprintf("Returning failure\n"); + return 0; // fail + } + } + + scidprintf("Returning success with dontclaim=%d\n", dontclaim); + + if (dontclaim) + return SAID_PARTIAL_MATCH; + else + return 1; // full match +} + + +/*******************/ +/**** Main code ****/ +/*******************/ + +int said(EngineState *s, byte *spec, bool verbose) { + int retval; + + parse_tree_node_t *parse_tree_ptr = s->_voc->_parserNodes; + + if (s->parserIsValid) { + if (said_parse_spec(spec)) { + printf("Offending spec was: "); + s->_voc->decipherSaidBlock(spec); + return SAID_NO_MATCH; + } + + if (verbose) + vocab_dump_parse_tree("Said-tree", said_tree); // Nothing better to do yet + retval = augment_parse_nodes(parse_tree_ptr, &(said_tree[0])); + + if (!retval) + return SAID_NO_MATCH; + else if (retval != SAID_PARTIAL_MATCH) + return SAID_FULL_MATCH; + else + return SAID_PARTIAL_MATCH; + } + + return SAID_NO_MATCH; +} + + +#ifdef SAID_DEBUG_PROGRAM +int main (int argc, char *argv) { + byte block[] = {0x01, 0x00, 0xf8, 0xf5, 0x02, 0x01, 0xf6, 0xf2, 0x02, 0x01, 0xf2, 0x01, 0x03, 0xff}; + EngineState s; + + s.parser_valid = 1; + said(&s, block); +} +#endif + +} // End of namespace Sci + diff --git a/engines/sci/parser/said.y b/engines/sci/parser/said.y new file mode 100644 index 0000000000..1a3b2f52c8 --- /dev/null +++ b/engines/sci/parser/said.y @@ -0,0 +1,838 @@ +%{ +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +#include "sci/engine/state.h" + + +// Bison generates an empty switch statement that gives a warning in MSVC. +// This disables that warning. +#ifdef _MSC_VER +#pragma warning(disable:4065) +#endif + + +namespace Sci { + +#define SAID_BRANCH_NULL 0 + +#define MAX_SAID_TOKENS 128 + +// Maximum number of words to be expected in a parsed sentence +#define AUGMENT_MAX_WORDS 64 + + +#define ANYWORD 0xfff + +#define WORD_TYPE_BASE 0x141 +#define WORD_TYPE_REF 0x144 +#define WORD_TYPE_SYNTACTIC_SUGAR 0x145 + +#define AUGMENT_SENTENCE_PART_BRACKETS 0x152 + +// Minor numbers +#define AUGMENT_SENTENCE_MINOR_MATCH_PHRASE 0x14c +#define AUGMENT_SENTENCE_MINOR_MATCH_WORD 0x153 +#define AUGMENT_SENTENCE_MINOR_RECURSE 0x144 +#define AUGMENT_SENTENCE_MINOR_PARENTHESES 0x14f + + +#undef YYDEBUG /*1*/ +//#define SAID_DEBUG*/ +//#define SCI_DEBUG_PARSE_TREE_AUGMENTATION // uncomment to debug parse tree augmentation + + +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION +#define scidprintf printf +#else +void print_nothing(...) { } +#define scidprintf print_nothing +#endif + + +static char *said_parse_error; + +static int said_token; +static int said_tokens_nr; +static int said_tokens[MAX_SAID_TOKENS]; +static int said_blessed; // increminated by said_top_branch + +static int said_tree_pos; // Set to 0 if we're out of space +#define SAID_TREE_START 4; // Reserve space for the 4 top nodes + +#define VALUE_IGNORE -424242 + +static parse_tree_node_t said_tree[VOCAB_TREE_NODES]; + +typedef int wgroup_t; +typedef int tree_t; +typedef int said_spec_t; + +static tree_t said_aug_branch(int, int, tree_t, tree_t); +static tree_t said_attach_branch(tree_t, tree_t); +/* +static tree_t said_wgroup_branch(wgroup_t); +*/ +static said_spec_t said_top_branch(tree_t); +static tree_t said_paren(tree_t, tree_t); +static tree_t said_value(int, tree_t); +static tree_t said_terminal(int); + +static int yylex(); + +static int yyerror(const char *s) { + said_parse_error = strdup(s); + return 1; /* Abort */ +} + +%} + +%token WGROUP /* Word group */ +%token YY_COMMA /* 0xf0 */ +%token YY_AMP /* 0xf1 */ +%token YY_SLASH /* 0xf2 */ +%token YY_PARENO /* 0xf3 */ +%token YY_PARENC /* 0xf4 */ +%token YY_BRACKETSO /* 0xf5 */ +%token YY_BRACKETSC /* 0xf6 */ +%token YY_HASH /* 0xf7 */ +%token YY_LT /* 0xf8 */ +%token YY_GT /* 0xf9 */ +%token YY_BRACKETSO_LT /* special token used to imitate LR(2) behaviour */ +%token YY_BRACKETSO_SLASH /* special token used to imitate LR(2) behaviour */ +%token YY_LT_BRACKETSO /* special token used to imitate LR(2) behaviour */ +%token YY_LT_PARENO /* special token used to imitate LR(2) behaviour */ + +%% + +saidspec : leftspec optcont + { $$ = said_top_branch(said_attach_branch($1, $2)); } + | leftspec midspec optcont + { $$ = said_top_branch(said_attach_branch($1, said_attach_branch($2, $3))); } + | leftspec midspec rightspec optcont + { $$ = said_top_branch(said_attach_branch($1, said_attach_branch($2, said_attach_branch($3, $4)))); } + ; + + +optcont : /* empty */ + { $$ = SAID_BRANCH_NULL; } + | YY_GT + { $$ = said_paren(said_value(0x14b, said_value(0xf900, said_terminal(0xf900))), SAID_BRANCH_NULL); } + ; + + + +leftspec : /* empty */ + { $$ = SAID_BRANCH_NULL; } + | expr + { $$ = said_paren(said_value(0x141, said_value(0x149, $1)), SAID_BRANCH_NULL); } + ; + + + +midspec : YY_SLASH expr + { $$ = said_aug_branch(0x142, 0x14a, $2, SAID_BRANCH_NULL); } + | YY_BRACKETSO_SLASH YY_SLASH expr YY_BRACKETSC + { $$ = said_aug_branch(0x152, 0x142, said_aug_branch(0x142, 0x14a, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + | YY_SLASH + { $$ = SAID_BRANCH_NULL; } + ; + + + +rightspec : YY_SLASH expr + { $$ = said_aug_branch(0x143, 0x14a, $2, SAID_BRANCH_NULL); } + | YY_BRACKETSO_SLASH YY_SLASH expr YY_BRACKETSC + { $$ = said_aug_branch(0x152, 0x143, said_aug_branch(0x143, 0x14a, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + | YY_SLASH + { $$ = SAID_BRANCH_NULL; } + ; + + +word : WGROUP + { $$ = said_paren(said_value(0x141, said_value(0x153, said_terminal($1))), SAID_BRANCH_NULL); } + ; + + +cwordset : wordset + { $$ = said_aug_branch(0x141, 0x14f, $1, SAID_BRANCH_NULL); } + | YY_BRACKETSO wordset YY_BRACKETSC + { $$ = said_aug_branch(0x141, 0x14f, said_aug_branch(0x152, 0x14c, said_aug_branch(0x141, 0x14f, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + ; + + +wordset : word + { $$ = $1; } + | YY_PARENO expr YY_PARENC + { $$ = said_aug_branch(0x141, 0x14c, $2, SAID_BRANCH_NULL); } + | wordset YY_COMMA wordset + { $$ = said_attach_branch($1, $3); } + | wordset YY_BRACKETSO_LT wordrefset YY_BRACKETSC + { $$ = said_attach_branch($1, $3); } + | wordset YY_COMMA YY_BRACKETSO wordset YY_BRACKETSC + { $$ = said_attach_branch($1, $3); } + ; + + +expr : cwordset cwordrefset + { $$ = said_attach_branch($1, $2); } + | cwordset + { $$ = $1; } + | cwordrefset + { $$ = $1; } + ; + + +cwordrefset : wordrefset + { $$ = $1; } + | YY_BRACKETSO_LT wordrefset YY_BRACKETSC + { $$ = said_aug_branch(0x152, 0x144, $2, SAID_BRANCH_NULL); } + | wordrefset YY_BRACKETSO_LT wordrefset YY_BRACKETSC + { $$ = said_attach_branch($1, said_aug_branch(0x152, 0x144, $3, SAID_BRANCH_NULL)); } + ; + + +wordrefset : YY_LT word recref + { $$ = said_aug_branch(0x144, 0x14f, $2, $3); } + | YY_LT_PARENO YY_PARENO expr YY_PARENC + { $$ = said_aug_branch(0x144, 0x14f, said_aug_branch(0x141, 0x144, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + | YY_LT wordset + { $$ = said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL); } + | YY_LT_BRACKETSO YY_BRACKETSO wordset YY_BRACKETSC + { $$ = said_aug_branch(0x152, 0x144, said_aug_branch(0x144, 0x14f, $3, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + ; + + +recref : YY_LT wordset recref + { $$ = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL), $3); } + | YY_LT wordset + { $$ = said_aug_branch(0x141, 0x144, said_aug_branch(0x144, 0x14f, $2, SAID_BRANCH_NULL), SAID_BRANCH_NULL); } + | YY_LT_PARENO YY_PARENO expr YY_PARENC + { $$ = said_aug_branch(0x141, 0x14c, $2, SAID_BRANCH_NULL); } + ; + +%% + +int parse_yy_token_lookup[] = {YY_COMMA, YY_AMP, YY_SLASH, YY_PARENO, YY_PARENC, YY_BRACKETSO, YY_BRACKETSC, YY_HASH, YY_LT, YY_GT}; + +static int yylex() { + int retval = said_tokens[said_token++]; + + if (retval < SAID_LONG(SAID_FIRST)) { + yylval = retval; + retval = WGROUP; + } else { + retval >>= 8; + + if (retval == SAID_TERM) + retval = 0; + else { + assert(retval >= SAID_FIRST); + retval = parse_yy_token_lookup[retval - SAID_FIRST]; + if (retval == YY_BRACKETSO) { + if ((said_tokens[said_token] >> 8) == SAID_LT) + retval = YY_BRACKETSO_LT; + else + if ((said_tokens[said_token] >> 8) == SAID_SLASH) + retval = YY_BRACKETSO_SLASH; + } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_BRACKO) { + retval = YY_LT_BRACKETSO; + } else if (retval == YY_LT && (said_tokens[said_token] >> 8) == SAID_PARENO) { + retval = YY_LT_PARENO; + } + } + } + + return retval; +} + +static int said_next_node() { + return ((said_tree_pos == 0) || (said_tree_pos >= VOCAB_TREE_NODES)) ? said_tree_pos = 0 : said_tree_pos++; +} + +#define SAID_NEXT_NODE said_next_node() + +static int said_leaf_node(tree_t pos, int value) { + said_tree[pos].type = PARSE_TREE_NODE_LEAF; + + if (value != VALUE_IGNORE) + said_tree[pos].content.value = value; + + return pos; +} + +static int said_branch_node(tree_t pos, int left, int right) { + said_tree[pos].type = PARSE_TREE_NODE_BRANCH; + + if (left != VALUE_IGNORE) + said_tree[pos].content.branches[0] = left; + + if (right != VALUE_IGNORE) + said_tree[pos].content.branches[1] = right; + + return pos; +} + +static tree_t said_paren(tree_t t1, tree_t t2) { + if (t1) + return said_branch_node(SAID_NEXT_NODE, t1, t2); + else + return t2; +} + +static tree_t said_value(int val, tree_t t) { + return said_branch_node(SAID_NEXT_NODE, said_leaf_node(SAID_NEXT_NODE, val), t); + +} + +static tree_t said_terminal(int val) { + return said_leaf_node(SAID_NEXT_NODE, val); +} + +static tree_t said_aug_branch(int n1, int n2, tree_t t1, tree_t t2) { + int retval; + + retval = said_branch_node(SAID_NEXT_NODE, + said_branch_node(SAID_NEXT_NODE, + said_leaf_node(SAID_NEXT_NODE, n1), + said_branch_node(SAID_NEXT_NODE, + said_leaf_node(SAID_NEXT_NODE, n2), + t1) + ), + t2); + +#ifdef SAID_DEBUG + fprintf(stderr, "AUG(0x%x, 0x%x, [%04x], [%04x]) = [%04x]\n", n1, n2, t1, t2, retval); +#endif + + return retval; +} + +static tree_t said_attach_branch(tree_t base, tree_t attacheant) { +#ifdef SAID_DEBUG + fprintf(stderr, "ATT2([%04x], [%04x]) = [%04x]\n", base, attacheant, base); +#endif + + if (!attacheant) + return base; + if (!base) + return attacheant; + + if (!base) + return 0; // Happens if we're out of space + + said_branch_node(base, VALUE_IGNORE, attacheant); + + return base; +} + +static said_spec_t said_top_branch(tree_t first) { +#ifdef SAID_DEBUG + fprintf(stderr, "TOP([%04x])\n", first); +#endif + said_branch_node(0, 1, 2); + said_leaf_node(1, 0x141); // Magic number #1 + said_branch_node(2, 3, first); + said_leaf_node(3, 0x13f); // Magic number #2 + + ++said_blessed; + + return 0; +} + +static int said_parse_spec(byte *spec) { + int nextitem; + + said_parse_error = NULL; + said_token = 0; + said_tokens_nr = 0; + said_blessed = 0; + + said_tree_pos = SAID_TREE_START; + + do { + nextitem = *spec++; + if (nextitem < SAID_FIRST) + said_tokens[said_tokens_nr++] = nextitem << 8 | *spec++; + else + said_tokens[said_tokens_nr++] = SAID_LONG(nextitem); + + } while ((nextitem != SAID_TERM) && (said_tokens_nr < MAX_SAID_TOKENS)); + + if (nextitem == SAID_TERM) + yyparse(); + else { + warning("SAID spec is too long"); + return 1; + } + + if (said_parse_error) { + warning("Error while parsing SAID spec: %s", said_parse_error); + free(said_parse_error); + return 1; + } + + if (said_tree_pos == 0) { + warning("Out of tree space while parsing SAID spec"); + return 1; + } + + if (said_blessed != 1) { + warning("Found multiple top branches"); + return 1; + } + + return 0; +} + +/**********************/ +/**** Augmentation ****/ +/**********************/ + +// primitive functions + +#define AUG_READ_BRANCH(a, br, p) \ + if (tree[p].type != PARSE_TREE_NODE_BRANCH) \ + return 0; \ + a = tree[p].content.branches[br]; + +#define AUG_READ_VALUE(a, p) \ + if (tree[p].type != PARSE_TREE_NODE_LEAF) \ + return 0; \ + a = tree[p].content.value; + +#define AUG_ASSERT(i) \ + if (!i) return 0; + +static int aug_get_next_sibling(parse_tree_node_t *tree, int pos, int *first, int *second) { + // Returns the next sibling relative to the specified position in 'tree', + // sets *first and *second to its augment node values, returns the new position + // or 0 if there was no next sibling + int seek, valpos; + + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(seek, 0, pos); + AUG_ASSERT(seek); + + // Now retrieve first value + AUG_READ_BRANCH(valpos, 0, seek); + AUG_ASSERT(valpos); + AUG_READ_VALUE(*first, valpos); + + // Get second value + AUG_READ_BRANCH(seek, 1, seek); + AUG_ASSERT(seek); + AUG_READ_BRANCH(valpos, 0, seek); + AUG_ASSERT(valpos); + AUG_READ_VALUE(*second, valpos); + + return pos; +} + +static int aug_get_wgroup(parse_tree_node_t *tree, int pos) { + // Returns 0 if pos in tree is not the root of a 3-element list, otherwise + // it returns the last element (which, in practice, is the word group + int val; + + AUG_READ_BRANCH(pos, 0, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + AUG_READ_VALUE(val, pos); + + return val; +} + +static int aug_get_base_node(parse_tree_node_t *tree) { + int startpos = 0; + AUG_READ_BRANCH(startpos, 1, startpos); + + return startpos; +} + +// semi-primitive functions + +static int aug_get_first_child(parse_tree_node_t *tree, int pos, int *first, int *second) { + // like aug_get_next_sibling, except that it recurses into the tree and + // finds the first child (usually *not* Ayanami Rei) of the current branch + // rather than its next sibling. + AUG_READ_BRANCH(pos, 0, pos); + AUG_ASSERT(pos); + AUG_READ_BRANCH(pos, 1, pos); + AUG_ASSERT(pos); + + return aug_get_next_sibling(tree, pos, first, second); +} + +static void aug_find_words_recursively(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, + int *ref_words, int *ref_words_nr, int maxwords, int refbranch) { + // Finds and lists all base (141) and reference (144) words */ + int major, minor; + int word; + int pos = aug_get_first_child(tree, startpos, &major, &minor); + + //if (major == WORD_TYPE_REF) + // refbranch = 1; + + while (pos) { + if ((word = aug_get_wgroup(tree, pos))) { // found a word + if (!refbranch && major == WORD_TYPE_BASE) { + if ((*base_words_nr) == maxwords) { + warning("Out of regular words"); + return; // return gracefully + } + + base_words[*base_words_nr] = word; // register word + ++(*base_words_nr); + + } + if (major == WORD_TYPE_REF || refbranch) { + if ((*ref_words_nr) == maxwords) { + warning("Out of reference words"); + return; // return gracefully + } + + ref_words[*ref_words_nr] = word; // register word + ++(*ref_words_nr); + + } + if (major != WORD_TYPE_SYNTACTIC_SUGAR && major != WORD_TYPE_BASE && major != WORD_TYPE_REF) + warning("aug_find_words_recursively(): Unknown word type %03x", major); + + } else // Did NOT find a word group: Attempt to recurse + aug_find_words_recursively(tree, pos, base_words, base_words_nr, + ref_words, ref_words_nr, maxwords, refbranch || major == WORD_TYPE_REF); + + pos = aug_get_next_sibling(tree, pos, &major, &minor); + } +} + + +static void aug_find_words(parse_tree_node_t *tree, int startpos, int *base_words, int *base_words_nr, + int *ref_words, int *ref_words_nr, int maxwords) { + // initializing wrapper for aug_find_words_recursively() + *base_words_nr = 0; + *ref_words_nr = 0; + + aug_find_words_recursively(tree, startpos, base_words, base_words_nr, ref_words, ref_words_nr, maxwords, 0); +} + + +static int aug_contains_word(int *list, int length, int word) { + int i; + + if (word == ANYWORD) + return (length); + + for (i = 0; i < length; i++) + if (list[i] == word) + return 1; + + return 0; +} + + +static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_branch, int major, int minor, int *base_words, int base_words_nr, + int *ref_words, int ref_words_nr); + +static int augment_match_expression_p(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_basepos, int major, int minor, + int *base_words, int base_words_nr, int *ref_words, int ref_words_nr) { + int cmajor, cminor, cpos; + cpos = aug_get_first_child(saidt, augment_pos, &cmajor, &cminor); + if (!cpos) { + warning("augment_match_expression_p(): Empty condition"); + return 1; + } + + scidprintf("Attempting to match (%03x %03x (%03x %03x\n", major, minor, cmajor, cminor); + + if ((major == WORD_TYPE_BASE) && (minor == AUGMENT_SENTENCE_MINOR_RECURSE)) + return augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr); + + switch (major) { + + case WORD_TYPE_BASE: + while (cpos) { + if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { + int word = aug_get_wgroup(saidt, cpos); + scidprintf("Looking for word %03x\n", word); + + if (aug_contains_word(base_words, base_words_nr, word)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { + if (augment_sentence_expression(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { + int gc_major, gc_minor; + int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); + + while (gchild) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, + minor, base_words, base_words_nr, + ref_words, ref_words_nr)) + return 1; + gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); + } + } else + warning("augment_match_expression_p(): Unknown type 141 minor number %3x", cminor); + + cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); + + } + break; + + case WORD_TYPE_REF: + while (cpos) { + if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_WORD) { + int word = aug_get_wgroup(saidt, cpos); + scidprintf("Looking for refword %03x\n", word); + + if (aug_contains_word(ref_words, ref_words_nr, word)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_MATCH_PHRASE) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + } else if (cminor == AUGMENT_SENTENCE_MINOR_PARENTHESES) { + int gc_major, gc_minor; + int gchild = aug_get_first_child(saidt, cpos, &gc_major, &gc_minor); + + while (gchild) { + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, major, + minor, base_words, base_words_nr, + ref_words, ref_words_nr)) + return 1; + gchild = aug_get_next_sibling(saidt, gchild, &gc_major, &gc_minor); + } + } else + warning("augment_match_expression_p(): Unknown type 144 minor number %3x", cminor); + + cpos = aug_get_next_sibling(saidt, cpos, &cmajor, &cminor); + + } + break; + + case AUGMENT_SENTENCE_PART_BRACKETS: + if (augment_match_expression_p(saidt, cpos, parset, parse_basepos, cmajor, cminor, + base_words, base_words_nr, ref_words, ref_words_nr)) + return 1; + + scidprintf("Didn't match subexpression; checking sub-bracked predicate %03x\n", cmajor); + + switch (cmajor) { + case WORD_TYPE_BASE: + if (!base_words_nr) + return 1; + break; + + case WORD_TYPE_REF: + if (!ref_words_nr) + return 1; + break; + + default: + warning("augment_match_expression_p(): (subp1) Unkonwn sub-bracket predicate %03x", cmajor); + } + + break; + + default: + warning("augment_match_expression_p(): Unknown predicate %03x", major); + + } + + scidprintf("augment_match_expression_p(): Generic failure\n"); + + return 0; +} + +static int augment_sentence_expression(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, + int parse_branch, int major, int minor, int *base_words, int base_words_nr, + int *ref_words, int ref_words_nr) { + int check_major, check_minor; + int check_pos = aug_get_first_child(saidt, augment_pos, &check_major, &check_minor); + do { + if (!(augment_match_expression_p(saidt, check_pos, parset, parse_branch, check_major, check_minor, + base_words, base_words_nr, ref_words, ref_words_nr))) + return 0; + } while ((check_pos = aug_get_next_sibling(saidt, check_pos, &check_major, &check_minor))); + + return 1; +} + +static int augment_sentence_part(parse_tree_node_t *saidt, int augment_pos, parse_tree_node_t *parset, int parse_basepos, int major, int minor) { + int pmajor, pminor; + int parse_branch = parse_basepos; + int optional = 0; + int foundwords = 0; + + scidprintf("Augmenting (%03x %03x\n", major, minor); + + if (major == AUGMENT_SENTENCE_PART_BRACKETS) { // '[/ foo]' is true if '/foo' or if there + // exists no x for which '/x' is true + if ((augment_pos = aug_get_first_child(saidt, augment_pos, &major, &minor))) { + scidprintf("Optional part: Now augmenting (%03x %03x\n", major, minor); + optional = 1; + } else { + scidprintf("Matched empty optional expression\n"); + return 1; + } + } + + if ((major < 0x141) || (major > 0x143)) { + scidprintf("augment_sentence_part(): Unexpected sentence part major number %03x\n", major); + return 0; + } + + while ((parse_branch = aug_get_next_sibling(parset, parse_branch, &pmajor, &pminor))) { + if (pmajor == major) { // found matching sentence part + int success; + int base_words_nr; + int ref_words_nr; + int base_words[AUGMENT_MAX_WORDS]; + int ref_words[AUGMENT_MAX_WORDS]; +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION + int i; +#endif + + scidprintf("Found match with pminor = %03x\n", pminor); + aug_find_words(parset, parse_branch, base_words, &base_words_nr, ref_words, &ref_words_nr, AUGMENT_MAX_WORDS); + foundwords |= (ref_words_nr | base_words_nr); +#ifdef SCI_DEBUG_PARSE_TREE_AUGMENTATION + printf("%d base words:", base_words_nr); + for (i = 0; i < base_words_nr; i++) + printf(" %03x", base_words[i]); + printf("\n%d reference words:", ref_words_nr); + for (i = 0; i < ref_words_nr; i++) + printf(" %03x", ref_words[i]); + printf("\n"); +#endif + + success = augment_sentence_expression(saidt, augment_pos, parset, parse_basepos, major, minor, + base_words, base_words_nr, ref_words, ref_words_nr); + + if (success) { + scidprintf("SUCCESS on augmenting (%03x %03x\n", major, minor); + return 1; + } + } + } + + if (optional && (foundwords == 0)) { + scidprintf("Found no words and optional branch => SUCCESS on augmenting (%03x %03x\n", major, minor); + return 1; + } + scidprintf("FAILURE on augmenting (%03x %03x\n", major, minor); + + return 0; +} + +static int augment_parse_nodes(parse_tree_node_t *parset, parse_tree_node_t *saidt) { + int augment_basepos = 0; + int parse_basepos; + int major, minor; + int dontclaim = 0; + + parse_basepos = aug_get_base_node(parset); + if (!parse_basepos) { + warning("augment_parse_nodes(): Parse tree is corrupt"); + return 0; + } + + augment_basepos = aug_get_base_node(saidt); + if (!augment_basepos) { + warning("augment_parse_nodes(): Said tree is corrupt"); + return 0; + } + + while ((augment_basepos = aug_get_next_sibling(saidt, augment_basepos, &major, &minor))) { + if ((major == 0x14b) && (minor == SAID_LONG(SAID_GT))) + dontclaim = 1; // special case + else // normal sentence part + if (!(augment_sentence_part(saidt, augment_basepos, parset, parse_basepos, major, minor))) { + scidprintf("Returning failure\n"); + return 0; // fail + } + } + + scidprintf("Returning success with dontclaim=%d\n", dontclaim); + + if (dontclaim) + return SAID_PARTIAL_MATCH; + else + return 1; // full match +} + + +/*******************/ +/**** Main code ****/ +/*******************/ + +int said(EngineState *s, byte *spec, bool verbose) { + int retval; + + parse_tree_node_t *parse_tree_ptr = s->_voc->_parser_nodes; + + if (s->parserIsValid) { + if (said_parse_spec(s, spec)) { + warning("Offending spec was: "); + s->_voc->decipherSaidBlock(spec); + return SAID_NO_MATCH; + } + + if (verbose) + vocab_dump_parse_tree("Said-tree", said_tree); // Nothing better to do yet + retval = augment_parse_nodes(parse_tree_ptr, &(said_tree[0])); + + if (!retval) + return SAID_NO_MATCH; + else if (retval != SAID_PARTIAL_MATCH) + return SAID_FULL_MATCH; + else + return SAID_PARTIAL_MATCH; + } + + return SAID_NO_MATCH; +} + + +#ifdef SAID_DEBUG_PROGRAM +int main (int argc, char *argv) { + byte block[] = {0x01, 0x00, 0xf8, 0xf5, 0x02, 0x01, 0xf6, 0xf2, 0x02, 0x01, 0xf2, 0x01, 0x03, 0xff}; + EngineState s; + + s.parser_valid = 1; + said(&s, block); +} +#endif + +} // End of namespace Sci diff --git a/engines/sci/parser/vocabulary.cpp b/engines/sci/parser/vocabulary.cpp new file mode 100644 index 0000000000..5ecea03899 --- /dev/null +++ b/engines/sci/parser/vocabulary.cpp @@ -0,0 +1,633 @@ +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +// Main vocabulary support functions and word lookup + +#include "sci/parser/vocabulary.h" +#include "sci/resource.h" +#include "sci/engine/state.h" +#include "sci/engine/kernel.h" +#include "sci/console.h" + +namespace Sci { + +#if 0 + +#define VOCAB_RESOURCE_CLASSES 996 +/** + * Vocabulary class names. + * These strange names were taken from an SCI01 interpreter. + */ +const char *class_names[] = {"", + "", + "conj", // conjunction + "ass", // ? + "pos", // preposition ? + "art", // article + "adj", // adjective + "pron", // pronoun + "noun", // noun + "auxv", // auxillary verb + "adv", // adverb + "verb", // verb + "", + "", + "", + "" + }; + +int *vocab_get_classes(ResourceManager *resMan, int* count) { + Resource* r; + int *c; + unsigned int i; + + if ((r = resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_CLASSES), 0)) == NULL) + return 0; + + c = (int *)malloc(sizeof(int) * r->size / 2); + for (i = 2; i < r->size; i += 4) { + c[i/4] = READ_LE_UINT16(r->data + i); + } + *count = r->size / 4; + + return c; +} + +int vocab_get_class_count(ResourceManager *resMan) { + Resource* r; + + if ((r = resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_CLASSES), 0)) == 0) + return 0; + + return r->size / 4; +} + +#endif + +Vocabulary::Vocabulary(ResourceManager *resMan) : _resMan(resMan) { + _parserRules = NULL; + _vocabVersion = kVocabularySCI0; + + memset(_parserNodes, 0, sizeof(_parserNodes)); + // Mark parse tree as unused + _parserNodes[0].type = kParseTreeLeafNode; + _parserNodes[0].content.value = 0; + + _synonyms.clear(); // No synonyms + + debug(2, "Initializing vocabulary"); + + if (getSciVersion() <= SCI_VERSION_1_EGA && loadParserWords()) { + loadSuffixes(); + if (loadBranches()) + // Now build a GNF grammar out of this + _parserRules = buildGNF(); + } else { + debug(2, "Assuming that this game does not use a parser."); + _parserRules = NULL; + } +} + +Vocabulary::~Vocabulary() { + freeRuleList(_parserRules); + freeSuffixes(); +} + +bool Vocabulary::loadParserWords() { + + char currentword[256] = ""; // They're not going to use words longer than 255 ;-) + int currentwordpos = 0; + + // First try to load the SCI0 vocab resource. + Resource *resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_MAIN_VOCAB), 0); + + if (!resource) { + warning("SCI0: Could not find a main vocabulary, trying SCI01"); + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_MAIN_VOCAB), 0); + _vocabVersion = kVocabularySCI1; + } + + if (!resource) { + warning("SCI1: Could not find a main vocabulary"); + return false; // NOT critical: SCI1 games and some demos don't have one! + } + + unsigned int seeker; + if (_vocabVersion == kVocabularySCI1) + seeker = 255 * 2; // vocab.900 starts with 255 16-bit pointers which we don't use + else + seeker = 26 * 2; // vocab.000 starts with 26 16-bit pointers which we don't use + + if (resource->size < seeker) { + warning("Invalid main vocabulary encountered: Too small"); + return false; + // Now this ought to be critical, but it'll just cause parse() and said() not to work + } + + _parserWords.clear(); + + while (seeker < resource->size) { + byte c; + + currentwordpos = resource->data[seeker++]; // Parts of previous words may be re-used + + if (_vocabVersion == kVocabularySCI1) { + c = 1; + while (seeker < resource->size && currentwordpos < 255 && c) { + c = resource->data[seeker++]; + currentword[currentwordpos++] = c; + } + if (seeker == resource->size) { + warning("SCI1: Vocabulary not usable, disabling"); + _parserWords.clear(); + return false; + } + } else { + do { + c = resource->data[seeker++]; + currentword[currentwordpos++] = c & 0x7f; // 0x80 is used to terminate the string + } while (c < 0x80); + } + + currentword[currentwordpos] = 0; + + // Now decode class and group: + c = resource->data[seeker + 1]; + ResultWord newWord; + newWord._class = ((resource->data[seeker]) << 4) | ((c & 0xf0) >> 4); + newWord._group = (resource->data[seeker + 2]) | ((c & 0x0f) << 8); + + // Add the word to the list + _parserWords[currentword] = newWord; + + seeker += 3; + } + + return true; +} + +const char *Vocabulary::getAnyWordFromGroup(int group) { + if (group == VOCAB_MAGIC_NUMBER_GROUP) + return "{number}"; + + for (WordMap::const_iterator i = _parserWords.begin(); i != _parserWords.end(); ++i) + if (i->_value._group == group) + return i->_key.c_str(); + + return "{invalid}"; +} + +bool Vocabulary::loadSuffixes() { + // Determine if we can find a SCI1 suffix vocabulary first + Resource* resource = NULL; + + if (_vocabVersion == kVocabularySCI0) + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB), 1); + else + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB), 1); + + if (!resource) + return false; // No vocabulary found + + unsigned int seeker = 1; + + while ((seeker < resource->size - 1) && (resource->data[seeker + 1] != 0xff)) { + suffix_t suffix; + + suffix.alt_suffix = (const char *)resource->data + seeker; + suffix.alt_suffix_length = strlen(suffix.alt_suffix); + seeker += suffix.alt_suffix_length + 1; // Hit end of string + + suffix.class_mask = (int16)READ_BE_UINT16(resource->data + seeker); + seeker += 2; + + // Beginning of next string - skip leading '*' + seeker++; + + suffix.word_suffix = (const char *)resource->data + seeker; + suffix.word_suffix_length = strlen(suffix.word_suffix); + seeker += suffix.word_suffix_length + 1; + + suffix.result_class = (int16)READ_BE_UINT16(resource->data + seeker); + seeker += 3; // Next entry + + _parserSuffixes.push_back(suffix); + } + + return true; +} + +void Vocabulary::freeSuffixes() { + Resource* resource = NULL; + + if (_vocabVersion == kVocabularySCI0) + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB), 0); + else + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB), 0); + + if (resource) + _resMan->unlockResource(resource); + + _parserSuffixes.clear(); +} + +bool Vocabulary::loadBranches() { + Resource *resource = NULL; + + if (_vocabVersion == kVocabularySCI0) + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_PARSE_TREE_BRANCHES), 0); + else + resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_PARSE_TREE_BRANCHES), 0); + + _parserBranches.clear(); + + if (!resource) + return false; // No parser tree data found + + int branches_nr = resource->size / 20; + + if (branches_nr == 0) { + warning("Parser tree data is empty"); + return false; + } + + _parserBranches.resize(branches_nr); + + for (int i = 0; i < branches_nr; i++) { + byte *base = resource->data + i * 20; + + _parserBranches[i].id = (int16)READ_LE_UINT16(base); + + for (int k = 0; k < 9; k++) + _parserBranches[i].data[k] = READ_LE_UINT16(base + 2 + 2 * k); + + _parserBranches[i].data[9] = 0; // Always terminate + } + + if (!_parserBranches[branches_nr - 1].id) // branch lists may be terminated by empty rules + _parserBranches.remove_at(branches_nr - 1); + + return true; +} + + +ResultWord Vocabulary::lookupWord(const char *word, int word_len) { + Common::String tempword(word, word_len); + + // Remove all dashes from tempword + for (uint i = 0; i < tempword.size(); ) { + if (tempword[i] == '-') + tempword.deleteChar(i); + else + ++i; + } + + // Look it up: + WordMap::iterator dict_word = _parserWords.find(tempword); + + // Match found? Return it! + if (dict_word != _parserWords.end()) { + return dict_word->_value; + } + + // Now try all suffixes + for (SuffixList::const_iterator suffix = _parserSuffixes.begin(); suffix != _parserSuffixes.end(); ++suffix) + if (suffix->alt_suffix_length <= word_len) { + + int suff_index = word_len - suffix->alt_suffix_length; + // Offset of the start of the suffix + + if (scumm_strnicmp(suffix->alt_suffix, word + suff_index, suffix->alt_suffix_length) == 0) { // Suffix matched! + // Terminate word at suffix start position...: + Common::String tempword2(word, MIN(word_len, suff_index)); + + // ...and append "correct" suffix + tempword2 += Common::String(suffix->word_suffix, suffix->word_suffix_length); + + dict_word = _parserWords.find(tempword2); + + if ((dict_word != _parserWords.end()) && (dict_word->_value._class & suffix->class_mask)) { // Found it? + // Use suffix class + ResultWord tmp = dict_word->_value; + tmp._class = suffix->result_class; + return tmp; + } + } + } + + // No match so far? Check if it's a number. + + ResultWord retval = { -1, -1 }; + char *tester; + if ((strtol(tempword.c_str(), &tester, 10) >= 0) && (*tester == '\0')) { // Do we have a complete number here? + ResultWord tmp = { VOCAB_CLASS_NUMBER, VOCAB_MAGIC_NUMBER_GROUP }; + retval = tmp; + } + + return retval; +} + +void Vocabulary::decipherSaidBlock(byte *addr) { + byte nextitem; + + do { + nextitem = *addr++; + + if (nextitem < 0xf0) { + nextitem = nextitem << 8 | *addr++; + printf(" %s[%03x]", getAnyWordFromGroup(nextitem), nextitem); + + nextitem = 42; // Make sure that group 0xff doesn't abort + } else switch (nextitem) { + case 0xf0: + printf(" ,"); + break; + case 0xf1: + printf(" &"); + break; + case 0xf2: + printf(" /"); + break; + case 0xf3: + printf(" ("); + break; + case 0xf4: + printf(" )"); + break; + case 0xf5: + printf(" ["); + break; + case 0xf6: + printf(" ]"); + break; + case 0xf7: + printf(" #"); + break; + case 0xf8: + printf(" <"); + break; + case 0xf9: + printf(" >"); + break; + case 0xff: + break; + } + } while (nextitem != 0xff); + + printf("\n"); +} + +bool Vocabulary::tokenizeString(ResultWordList &retval, const char *sentence, char **error) { + const char *lastword = sentence; + int pos_in_sentence = 0; + char c; + int wordlen = 0; + + *error = NULL; + + do { + + c = sentence[pos_in_sentence++]; + + if (isalnum(c) || (c == '-' && wordlen)) + ++wordlen; + // Continue on this word */ + // Words may contain a '-', but may not + // start with one. + else { + if (wordlen) { // Finished a word? + + ResultWord lookup_result = lookupWord(lastword, wordlen); + // Look it up + + if (lookup_result._class == -1) { // Not found? + *error = (char *)calloc(wordlen + 1, 1); + strncpy(*error, lastword, wordlen); // Set the offending word + retval.clear(); + return false; // And return with error + } + + // Copy into list + retval.push_back(lookup_result); + } + + lastword = sentence + pos_in_sentence; + wordlen = 0; + } + + } while (c); // Until terminator is hit + + return true; +} + +void Vocabulary::printSuffixes() const { + char word_buf[256], alt_buf[256]; + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + + int i = 0; + for (SuffixList::const_iterator suf = _parserSuffixes.begin(); suf != _parserSuffixes.end(); ++suf) { + strncpy(word_buf, suf->word_suffix, suf->word_suffix_length); + word_buf[suf->word_suffix_length] = 0; + strncpy(alt_buf, suf->alt_suffix, suf->alt_suffix_length); + alt_buf[suf->alt_suffix_length] = 0; + + con->DebugPrintf("%4d: (%03x) -%12s => -%12s (%03x)\n", i, suf->class_mask, word_buf, alt_buf, suf->result_class); + ++i; + } +} + +void Vocabulary::printParserWords() const { + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + + int j = 0; + for (WordMap::iterator i = _parserWords.begin(); i != _parserWords.end(); ++i) { + con->DebugPrintf("%4d: %03x [%03x] %20s |", j, i->_value._class, i->_value._group, i->_key.c_str()); + if (j % 3 == 0) + con->DebugPrintf("\n"); + j++; + } + + con->DebugPrintf("\n"); +} + +void _vocab_recursive_ptree_dump_treelike(parse_tree_node_t *nodes, int nr, int prevnr) { + if ((nr > VOCAB_TREE_NODES)/* || (nr < prevnr)*/) { + printf("Error(%04x)", nr); + return; + } + + if (nodes[nr].type == kParseTreeLeafNode) + //printf("[%03x]%04x", nr, nodes[nr].content.value); + printf("%x", nodes[nr].content.value); + else { + int lbranch = nodes[nr].content.branches[0]; + int rbranch = nodes[nr].content.branches[1]; + //printf("<[%03x]", nr); + printf("<"); + + if (lbranch) + _vocab_recursive_ptree_dump_treelike(nodes, lbranch, nr); + else + printf("NULL"); + + printf(","); + + if (rbranch) + _vocab_recursive_ptree_dump_treelike(nodes, rbranch, nr); + else + printf("NULL"); + + printf(">"); + } +} + +void _vocab_recursive_ptree_dump(parse_tree_node_t *nodes, int nr, int prevnr, int blanks) { + int lbranch = nodes[nr].content.branches[0]; + int rbranch = nodes[nr].content.branches[1]; + int i; + + if (nodes[nr].type == kParseTreeLeafNode) { + printf("vocab_dump_parse_tree: Error: consp is nil for element %03x\n", nr); + return; + } + + if ((nr > VOCAB_TREE_NODES)/* || (nr < prevnr)*/) { + printf("Error(%04x))", nr); + return; + } + + if (lbranch) { + if (nodes[lbranch].type == kParseTreeBranchNode) { + printf("\n"); + for (i = 0; i < blanks; i++) + printf(" "); + printf("("); + _vocab_recursive_ptree_dump(nodes, lbranch, nr, blanks + 1); + printf(")\n"); + for (i = 0; i < blanks; i++) + printf(" "); + } else + printf("%x", nodes[lbranch].content.value); + printf(" "); + }/* else printf ("nil");*/ + + if (rbranch) { + if (nodes[rbranch].type == kParseTreeBranchNode) + _vocab_recursive_ptree_dump(nodes, rbranch, nr, blanks); + else + printf("%x", nodes[rbranch].content.value); + }/* else printf("nil");*/ +} + +void vocab_dump_parse_tree(const char *tree_name, parse_tree_node_t *nodes) { + //_vocab_recursive_ptree_dump_treelike(nodes, 0, 0); + printf("(setq %s \n'(", tree_name); + _vocab_recursive_ptree_dump(nodes, 0, 0, 1); + printf("))\n"); +} + +void Vocabulary::dumpParseTree() { + //_vocab_recursive_ptree_dump_treelike(nodes, 0, 0); + printf("(setq parse-tree \n'("); + _vocab_recursive_ptree_dump(_parserNodes, 0, 0, 1); + printf("))\n"); +} + +void Vocabulary::synonymizeTokens(ResultWordList &words) { + if (_synonyms.empty()) + return; // No synonyms: Nothing to check + + for (ResultWordList::iterator i = words.begin(); i != words.end(); ++i) + for (SynonymList::const_iterator sync = _synonyms.begin(); sync != _synonyms.end(); ++sync) + if (i->_group == sync->replaceant) + i->_group = sync->replacement; +} + +void Vocabulary::printParserNodes(int num) { + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + + for (int i = 0; i < num; i++) { + con->DebugPrintf(" Node %03x: ", i); + if (_parserNodes[i].type == kParseTreeLeafNode) + con->DebugPrintf("Leaf: %04x\n", _parserNodes[i].content.value); + else + con->DebugPrintf("Branch: ->%04x, ->%04x\n", _parserNodes[i].content.branches[0], + _parserNodes[i].content.branches[1]); + } +} + +int Vocabulary::parseNodes(int *i, int *pos, int type, int nr, int argc, const char **argv) { + int nextToken = 0, nextValue = 0, newPos = 0, oldPos = 0; + Console *con = ((SciEngine *)g_engine)->getSciDebugger(); + + if (type == kParseNil) + return 0; + + if (type == kParseNumber) { + _parserNodes[*pos += 1].type = kParseTreeLeafNode; + _parserNodes[*pos].content.value = nr; + return *pos; + } + if (type == kParseEndOfInput) { + con->DebugPrintf("Unbalanced parentheses\n"); + return -1; + } + if (type == kParseClosingParenthesis) { + con->DebugPrintf("Syntax error at token %d\n", *i); + return -1; + } + + _parserNodes[oldPos = ++(*pos)].type = kParseTreeBranchNode; + + for (int j = 0; j <= 1; j++) { + if (*i == argc) { + nextToken = kParseEndOfInput; + } else { + const char *token = argv[(*i)++]; + + if (!strcmp(token, "(")) { + nextToken = kParseOpeningParenthesis; + } else if (!strcmp(token, ")")) { + nextToken = kParseClosingParenthesis; + } else if (!strcmp(token, "nil")) { + nextToken = kParseNil; + } else { + nextValue = strtol(token, NULL, 0); + nextToken = kParseNumber; + } + } + + if ((newPos = _parserNodes[oldPos].content.branches[j] = parseNodes(i, pos, nextToken, nextValue, argc, argv)) == -1) + return -1; + } + + const char *token = argv[(*i)++]; + if (strcmp(token, ")")) + con->DebugPrintf("Expected ')' at token %d\n", *i); + + return oldPos; +} + +} // End of namespace Sci diff --git a/engines/sci/parser/vocabulary.h b/engines/sci/parser/vocabulary.h new file mode 100644 index 0000000000..00b8780d1c --- /dev/null +++ b/engines/sci/parser/vocabulary.h @@ -0,0 +1,338 @@ +/* ScummVM - Graphic Adventure Engine + * + * ScummVM is the legal property of its developers, whose names + * are too numerous to list here. Please refer to the COPYRIGHT + * file distributed with this source distribution. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * $URL$ + * $Id$ + * + */ + +#ifndef SCI_SCICORE_VOCABULARY_H +#define SCI_SCICORE_VOCABULARY_H + +#include "common/str.h" +#include "common/hashmap.h" +#include "common/hash-str.h" +#include "common/list.h" + +#include "sci/sci.h" + +namespace Sci { + +class ResourceManager; + +/*#define VOCABULARY_DEBUG */ + +/** Number of bytes allocated on the heap to store bad words if parsing fails */ +#define PARSE_HEAP_SIZE 64 + +enum { + VOCAB_RESOURCE_SELECTORS = 997, + + VOCAB_RESOURCE_SCI0_MAIN_VOCAB = 0, + VOCAB_RESOURCE_SCI0_PARSE_TREE_BRANCHES = 900, + VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB = 901, + + VOCAB_RESOURCE_SCI1_MAIN_VOCAB = 900, + VOCAB_RESOURCE_SCI1_PARSE_TREE_BRANCHES = 901, + VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB = 902 +}; + + +enum { + VOCAB_CLASS_PREPOSITION = 0x01, + VOCAB_CLASS_ARTICLE = 0x02, + VOCAB_CLASS_ADJECTIVE = 0x04, + VOCAB_CLASS_PRONOUN = 0x08, + VOCAB_CLASS_NOUN = 0x10, + VOCAB_CLASS_INDICATIVE_VERB = 0x20, + VOCAB_CLASS_ADVERB = 0x40, + VOCAB_CLASS_IMPERATIVE_VERB = 0x80, + VOCAB_CLASS_NUMBER = 0x001 +}; + +enum { + kParseEndOfInput = 0, + kParseOpeningParenthesis = 1, + kParseClosingParenthesis = 2, + kParseNil = 3, + kParseNumber = 4 +}; + +/* Anywords are ignored by the parser */ +#define VOCAB_CLASS_ANYWORD 0xff + +/* This word class is used for numbers */ +#define VOCAB_MAGIC_NUMBER_GROUP 0xffd /* 0xffe ? */ + +/* Number of nodes for each parse_tree_node structure */ +#define VOCAB_TREE_NODES 500 + +#define VOCAB_TREE_NODE_LAST_WORD_STORAGE 0x140 +#define VOCAB_TREE_NODE_COMPARE_TYPE 0x146 +#define VOCAB_TREE_NODE_COMPARE_GROUP 0x14d +#define VOCAB_TREE_NODE_FORCE_STORAGE 0x154 + +#define SAID_COMMA 0xf0 +#define SAID_AMP 0xf1 +#define SAID_SLASH 0xf2 +#define SAID_PARENO 0xf3 +#define SAID_PARENC 0xf4 +#define SAID_BRACKO 0xf5 +#define SAID_BRACKC 0xf6 +#define SAID_HASH 0xf7 +#define SAID_LT 0xf8 +#define SAID_GT 0xf9 +#define SAID_TERM 0xff + +#define SAID_FIRST SAID_COMMA + +/* There was no 'last matching word': */ +#define SAID_FULL_MATCH 0xffff +#define SAID_NO_MATCH 0xfffe +#define SAID_PARTIAL_MATCH 0xfffd + +#define SAID_LONG(x) ((x) << 8) + +struct ResultWord { + int _class; /**< Word class */ + int _group; /**< Word group */ +}; + +typedef Common::List ResultWordList; + +typedef Common::HashMap WordMap; + + +struct ParseRuleList; + +struct suffix_t { + + int class_mask; /**< the word class this suffix applies to */ + int result_class; /**< the word class a word is morphed to if it doesn't fail this check */ + + int alt_suffix_length; /**< String length of the suffix */ + int word_suffix_length; /**< String length of the other suffix */ + + const char *alt_suffix; /**< The alternative suffix */ + const char *word_suffix; /**< The suffix as used in the word vocabulary */ + +}; + +typedef Common::List SuffixList; + + +struct synonym_t { + int replaceant; /**< The word group to replace */ + int replacement; /**< The replacement word group for this one */ +}; + +typedef Common::List SynonymList; + +struct parse_tree_branch_t { + int id; + int data[10]; +}; + +enum ParseTypes { + kParseTreeLeafNode = 0, + kParseTreeBranchNode = 1 +}; + +struct parse_tree_node_t { + ParseTypes type; /**< leaf or branch */ + union { + int value; /**< For leaves */ + short branches[2]; /**< For branches */ + } content; +}; + +enum VocabularyVersions { + kVocabularySCI0 = 0, + kVocabularySCI1 = 1 +}; + +class Vocabulary { +public: + Vocabulary(ResourceManager *resMan); + ~Vocabulary(); + + /** + * Gets any word from the specified group. For debugging only. + * @param group Group number + */ + const char *getAnyWordFromGroup(int group); + + + /** + * Looks up a single word in the words and suffixes list. + * @param word pointer to the word to look up + * @param word_len length of the word to look up + * @return the matching word (or (-1,-1) if there was no match) + */ + ResultWord lookupWord(const char *word, int word_len); + + + /** + * Tokenizes a string and compiles it into word_ts. + * @param[in] retval A list of words which will be set to the result + * @param[out] sentence The sentence to examine + * @param[out] error Points to a malloc'd copy of the offending text or to NULL on error + * @return true on success, false on failure + * + * On error, false is returned. If *error is NULL, the sentence did not + * contain any useful words; if not, *error points to a malloc'd copy of + * the offending word. The returned list may contain anywords. + */ + bool tokenizeString(ResultWordList &retval, const char *sentence, char **error); + + /** + * Builds a parse tree from a list of words, using a set of Greibach Normal + * Form rules. + * @param words The words to build the tree from + * @param verbose Set to true for debugging + * @return 0 on success, 1 if the tree couldn't be built in VOCAB_TREE_NODES + * nodes or if the sentence structure in 'words' is not part of the + * language described by the grammar passed in 'rules'. + */ + int parseGNF(const ResultWordList &words, bool verbose = false); + + /** + * Constructs the Greibach Normal Form of the grammar supplied in 'branches'. + * @param verbose Set to true for debugging. If true, the list is + * freed before the function ends + * @return Pointer to a list of singly linked GNF rules describing the same + * language that was described by 'branches' + * + * The original SCI rules are in almost-CNF (Chomsky Normal Form). Note that + * branch[0] is used only for a few magical incantations, as it is treated + * specially by the SCI parser. + */ + ParseRuleList *buildGNF(bool verbose = false); + + /** + * Deciphers a said block and dumps its content via printf. + * For debugging only. + * @param pos pointer to the data to dump + */ + void decipherSaidBlock(byte *pos); + + /** + * Prints the parser suffixes to the debug console. + */ + void printSuffixes() const; + + /** + * Prints the parser words to the debug console. + */ + void printParserWords() const; + + uint getParserBranchesSize() const { return _parserBranches.size(); } + const parse_tree_branch_t &getParseTreeBranch(int number) const { return _parserBranches[number]; } + + /** + * Adds a new synonym to the list + */ + void addSynonym(synonym_t syn) { _synonyms.push_back(syn); } + + /** + * Clears the list of synonyms + */ + void clearSynonyms() { _synonyms.clear(); } + + /** + * Synonymizes a token list + * Parameters: (ResultWordList &) words: The word list to synonymize + */ + void synonymizeTokens(ResultWordList &words); + + void printParserNodes(int num); + + void dumpParseTree(); + + int parseNodes(int *i, int *pos, int type, int nr, int argc, const char **argv); + +private: + /** + * Loads all words from the main vocabulary. + * @return true on success, false on failure + */ + bool loadParserWords(); + + /** + * Loads all suffixes from the suffix vocabulary. + * @return true on success, false on failure + */ + bool loadSuffixes(); + + /** + * Frees all suffixes in the given list. + * @param suffixes: The suffixes to free + */ + void freeSuffixes(); + + /** + * Retrieves all grammar rules from the resource data. + * @param branches The rules are stored into this Array + * @return true on success, false on error + */ + bool loadBranches(); + + /** + * Frees a parser rule list as returned by vocab_build_gnf(). + * @param rule_list the rule list to free + */ + void freeRuleList(ParseRuleList *rule_list); + + ResourceManager *_resMan; + VocabularyVersions _vocabVersion; + + // Parser-related lists + SuffixList _parserSuffixes; + ParseRuleList *_parserRules; /**< GNF rules used in the parser algorithm */ + Common::Array _parserBranches; + WordMap _parserWords; + SynonymList _synonyms; /**< The list of synonyms */ + +public: + // Accessed by said() + parse_tree_node_t _parserNodes[VOCAB_TREE_NODES]; /**< The parse tree */ +}; + +/** + * Prints a parse tree. + * @param tree_name Name of the tree to dump (free-form) + * @param nodes The nodes containing the parse tree + */ +void vocab_dump_parse_tree(const char *tree_name, parse_tree_node_t *nodes); + + + +/** + * Builds a parse tree from a spec and compares it to a parse tree. + * @param s The affected state + * @param spec Pointer to the spec to build + * @param verbose Whether to display the parse tree after building it + * @return 1 on a match, 0 otherwise + */ +int said(EngineState *s, byte *spec, bool verbose); + +} // End of namespace Sci + +#endif // SCI_SCICORE_VOCABULARY_H diff --git a/engines/sci/resource.cpp b/engines/sci/resource.cpp index 7a09041287..e4cd9b7d70 100644 --- a/engines/sci/resource.cpp +++ b/engines/sci/resource.cpp @@ -32,7 +32,7 @@ #include "sci/engine/state.h" #include "sci/engine/kernel.h" #include "sci/resource.h" -#include "sci/vocabulary.h" +#include "sci/parser/vocabulary.h" #include "sci/decompressor.h" namespace Sci { diff --git a/engines/sci/vocabulary.cpp b/engines/sci/vocabulary.cpp deleted file mode 100644 index 384530fb3e..0000000000 --- a/engines/sci/vocabulary.cpp +++ /dev/null @@ -1,633 +0,0 @@ -/* ScummVM - Graphic Adventure Engine - * - * ScummVM is the legal property of its developers, whose names - * are too numerous to list here. Please refer to the COPYRIGHT - * file distributed with this source distribution. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - * - * $URL$ - * $Id$ - * - */ - -// Main vocabulary support functions and word lookup - -#include "sci/vocabulary.h" -#include "sci/resource.h" -#include "sci/engine/state.h" -#include "sci/engine/kernel.h" -#include "sci/console.h" - -namespace Sci { - -#if 0 - -#define VOCAB_RESOURCE_CLASSES 996 -/** - * Vocabulary class names. - * These strange names were taken from an SCI01 interpreter. - */ -const char *class_names[] = {"", - "", - "conj", // conjunction - "ass", // ? - "pos", // preposition ? - "art", // article - "adj", // adjective - "pron", // pronoun - "noun", // noun - "auxv", // auxillary verb - "adv", // adverb - "verb", // verb - "", - "", - "", - "" - }; - -int *vocab_get_classes(ResourceManager *resMan, int* count) { - Resource* r; - int *c; - unsigned int i; - - if ((r = resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_CLASSES), 0)) == NULL) - return 0; - - c = (int *)malloc(sizeof(int) * r->size / 2); - for (i = 2; i < r->size; i += 4) { - c[i/4] = READ_LE_UINT16(r->data + i); - } - *count = r->size / 4; - - return c; -} - -int vocab_get_class_count(ResourceManager *resMan) { - Resource* r; - - if ((r = resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_CLASSES), 0)) == 0) - return 0; - - return r->size / 4; -} - -#endif - -Vocabulary::Vocabulary(ResourceManager *resMan) : _resMan(resMan) { - _parserRules = NULL; - _vocabVersion = kVocabularySCI0; - - memset(_parserNodes, 0, sizeof(_parserNodes)); - // Mark parse tree as unused - _parserNodes[0].type = kParseTreeLeafNode; - _parserNodes[0].content.value = 0; - - _synonyms.clear(); // No synonyms - - debug(2, "Initializing vocabulary"); - - if (getSciVersion() <= SCI_VERSION_1_EGA && loadParserWords()) { - loadSuffixes(); - if (loadBranches()) - // Now build a GNF grammar out of this - _parserRules = buildGNF(); - } else { - debug(2, "Assuming that this game does not use a parser."); - _parserRules = NULL; - } -} - -Vocabulary::~Vocabulary() { - freeRuleList(_parserRules); - freeSuffixes(); -} - -bool Vocabulary::loadParserWords() { - - char currentword[256] = ""; // They're not going to use words longer than 255 ;-) - int currentwordpos = 0; - - // First try to load the SCI0 vocab resource. - Resource *resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_MAIN_VOCAB), 0); - - if (!resource) { - warning("SCI0: Could not find a main vocabulary, trying SCI01"); - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_MAIN_VOCAB), 0); - _vocabVersion = kVocabularySCI1; - } - - if (!resource) { - warning("SCI1: Could not find a main vocabulary"); - return false; // NOT critical: SCI1 games and some demos don't have one! - } - - unsigned int seeker; - if (_vocabVersion == kVocabularySCI1) - seeker = 255 * 2; // vocab.900 starts with 255 16-bit pointers which we don't use - else - seeker = 26 * 2; // vocab.000 starts with 26 16-bit pointers which we don't use - - if (resource->size < seeker) { - warning("Invalid main vocabulary encountered: Too small"); - return false; - // Now this ought to be critical, but it'll just cause parse() and said() not to work - } - - _parserWords.clear(); - - while (seeker < resource->size) { - byte c; - - currentwordpos = resource->data[seeker++]; // Parts of previous words may be re-used - - if (_vocabVersion == kVocabularySCI1) { - c = 1; - while (seeker < resource->size && currentwordpos < 255 && c) { - c = resource->data[seeker++]; - currentword[currentwordpos++] = c; - } - if (seeker == resource->size) { - warning("SCI1: Vocabulary not usable, disabling"); - _parserWords.clear(); - return false; - } - } else { - do { - c = resource->data[seeker++]; - currentword[currentwordpos++] = c & 0x7f; // 0x80 is used to terminate the string - } while (c < 0x80); - } - - currentword[currentwordpos] = 0; - - // Now decode class and group: - c = resource->data[seeker + 1]; - ResultWord newWord; - newWord._class = ((resource->data[seeker]) << 4) | ((c & 0xf0) >> 4); - newWord._group = (resource->data[seeker + 2]) | ((c & 0x0f) << 8); - - // Add the word to the list - _parserWords[currentword] = newWord; - - seeker += 3; - } - - return true; -} - -const char *Vocabulary::getAnyWordFromGroup(int group) { - if (group == VOCAB_MAGIC_NUMBER_GROUP) - return "{number}"; - - for (WordMap::const_iterator i = _parserWords.begin(); i != _parserWords.end(); ++i) - if (i->_value._group == group) - return i->_key.c_str(); - - return "{invalid}"; -} - -bool Vocabulary::loadSuffixes() { - // Determine if we can find a SCI1 suffix vocabulary first - Resource* resource = NULL; - - if (_vocabVersion == kVocabularySCI0) - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB), 1); - else - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB), 1); - - if (!resource) - return false; // No vocabulary found - - unsigned int seeker = 1; - - while ((seeker < resource->size - 1) && (resource->data[seeker + 1] != 0xff)) { - suffix_t suffix; - - suffix.alt_suffix = (const char *)resource->data + seeker; - suffix.alt_suffix_length = strlen(suffix.alt_suffix); - seeker += suffix.alt_suffix_length + 1; // Hit end of string - - suffix.class_mask = (int16)READ_BE_UINT16(resource->data + seeker); - seeker += 2; - - // Beginning of next string - skip leading '*' - seeker++; - - suffix.word_suffix = (const char *)resource->data + seeker; - suffix.word_suffix_length = strlen(suffix.word_suffix); - seeker += suffix.word_suffix_length + 1; - - suffix.result_class = (int16)READ_BE_UINT16(resource->data + seeker); - seeker += 3; // Next entry - - _parserSuffixes.push_back(suffix); - } - - return true; -} - -void Vocabulary::freeSuffixes() { - Resource* resource = NULL; - - if (_vocabVersion == kVocabularySCI0) - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB), 0); - else - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB), 0); - - if (resource) - _resMan->unlockResource(resource); - - _parserSuffixes.clear(); -} - -bool Vocabulary::loadBranches() { - Resource *resource = NULL; - - if (_vocabVersion == kVocabularySCI0) - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI0_PARSE_TREE_BRANCHES), 0); - else - resource = _resMan->findResource(ResourceId(kResourceTypeVocab, VOCAB_RESOURCE_SCI1_PARSE_TREE_BRANCHES), 0); - - _parserBranches.clear(); - - if (!resource) - return false; // No parser tree data found - - int branches_nr = resource->size / 20; - - if (branches_nr == 0) { - warning("Parser tree data is empty"); - return false; - } - - _parserBranches.resize(branches_nr); - - for (int i = 0; i < branches_nr; i++) { - byte *base = resource->data + i * 20; - - _parserBranches[i].id = (int16)READ_LE_UINT16(base); - - for (int k = 0; k < 9; k++) - _parserBranches[i].data[k] = READ_LE_UINT16(base + 2 + 2 * k); - - _parserBranches[i].data[9] = 0; // Always terminate - } - - if (!_parserBranches[branches_nr - 1].id) // branch lists may be terminated by empty rules - _parserBranches.remove_at(branches_nr - 1); - - return true; -} - - -ResultWord Vocabulary::lookupWord(const char *word, int word_len) { - Common::String tempword(word, word_len); - - // Remove all dashes from tempword - for (uint i = 0; i < tempword.size(); ) { - if (tempword[i] == '-') - tempword.deleteChar(i); - else - ++i; - } - - // Look it up: - WordMap::iterator dict_word = _parserWords.find(tempword); - - // Match found? Return it! - if (dict_word != _parserWords.end()) { - return dict_word->_value; - } - - // Now try all suffixes - for (SuffixList::const_iterator suffix = _parserSuffixes.begin(); suffix != _parserSuffixes.end(); ++suffix) - if (suffix->alt_suffix_length <= word_len) { - - int suff_index = word_len - suffix->alt_suffix_length; - // Offset of the start of the suffix - - if (scumm_strnicmp(suffix->alt_suffix, word + suff_index, suffix->alt_suffix_length) == 0) { // Suffix matched! - // Terminate word at suffix start position...: - Common::String tempword2(word, MIN(word_len, suff_index)); - - // ...and append "correct" suffix - tempword2 += Common::String(suffix->word_suffix, suffix->word_suffix_length); - - dict_word = _parserWords.find(tempword2); - - if ((dict_word != _parserWords.end()) && (dict_word->_value._class & suffix->class_mask)) { // Found it? - // Use suffix class - ResultWord tmp = dict_word->_value; - tmp._class = suffix->result_class; - return tmp; - } - } - } - - // No match so far? Check if it's a number. - - ResultWord retval = { -1, -1 }; - char *tester; - if ((strtol(tempword.c_str(), &tester, 10) >= 0) && (*tester == '\0')) { // Do we have a complete number here? - ResultWord tmp = { VOCAB_CLASS_NUMBER, VOCAB_MAGIC_NUMBER_GROUP }; - retval = tmp; - } - - return retval; -} - -void Vocabulary::decipherSaidBlock(byte *addr) { - byte nextitem; - - do { - nextitem = *addr++; - - if (nextitem < 0xf0) { - nextitem = nextitem << 8 | *addr++; - printf(" %s[%03x]", getAnyWordFromGroup(nextitem), nextitem); - - nextitem = 42; // Make sure that group 0xff doesn't abort - } else switch (nextitem) { - case 0xf0: - printf(" ,"); - break; - case 0xf1: - printf(" &"); - break; - case 0xf2: - printf(" /"); - break; - case 0xf3: - printf(" ("); - break; - case 0xf4: - printf(" )"); - break; - case 0xf5: - printf(" ["); - break; - case 0xf6: - printf(" ]"); - break; - case 0xf7: - printf(" #"); - break; - case 0xf8: - printf(" <"); - break; - case 0xf9: - printf(" >"); - break; - case 0xff: - break; - } - } while (nextitem != 0xff); - - printf("\n"); -} - -bool Vocabulary::tokenizeString(ResultWordList &retval, const char *sentence, char **error) { - const char *lastword = sentence; - int pos_in_sentence = 0; - char c; - int wordlen = 0; - - *error = NULL; - - do { - - c = sentence[pos_in_sentence++]; - - if (isalnum(c) || (c == '-' && wordlen)) - ++wordlen; - // Continue on this word */ - // Words may contain a '-', but may not - // start with one. - else { - if (wordlen) { // Finished a word? - - ResultWord lookup_result = lookupWord(lastword, wordlen); - // Look it up - - if (lookup_result._class == -1) { // Not found? - *error = (char *)calloc(wordlen + 1, 1); - strncpy(*error, lastword, wordlen); // Set the offending word - retval.clear(); - return false; // And return with error - } - - // Copy into list - retval.push_back(lookup_result); - } - - lastword = sentence + pos_in_sentence; - wordlen = 0; - } - - } while (c); // Until terminator is hit - - return true; -} - -void Vocabulary::printSuffixes() const { - char word_buf[256], alt_buf[256]; - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - - int i = 0; - for (SuffixList::const_iterator suf = _parserSuffixes.begin(); suf != _parserSuffixes.end(); ++suf) { - strncpy(word_buf, suf->word_suffix, suf->word_suffix_length); - word_buf[suf->word_suffix_length] = 0; - strncpy(alt_buf, suf->alt_suffix, suf->alt_suffix_length); - alt_buf[suf->alt_suffix_length] = 0; - - con->DebugPrintf("%4d: (%03x) -%12s => -%12s (%03x)\n", i, suf->class_mask, word_buf, alt_buf, suf->result_class); - ++i; - } -} - -void Vocabulary::printParserWords() const { - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - - int j = 0; - for (WordMap::iterator i = _parserWords.begin(); i != _parserWords.end(); ++i) { - con->DebugPrintf("%4d: %03x [%03x] %20s |", j, i->_value._class, i->_value._group, i->_key.c_str()); - if (j % 3 == 0) - con->DebugPrintf("\n"); - j++; - } - - con->DebugPrintf("\n"); -} - -void _vocab_recursive_ptree_dump_treelike(parse_tree_node_t *nodes, int nr, int prevnr) { - if ((nr > VOCAB_TREE_NODES)/* || (nr < prevnr)*/) { - printf("Error(%04x)", nr); - return; - } - - if (nodes[nr].type == kParseTreeLeafNode) - //printf("[%03x]%04x", nr, nodes[nr].content.value); - printf("%x", nodes[nr].content.value); - else { - int lbranch = nodes[nr].content.branches[0]; - int rbranch = nodes[nr].content.branches[1]; - //printf("<[%03x]", nr); - printf("<"); - - if (lbranch) - _vocab_recursive_ptree_dump_treelike(nodes, lbranch, nr); - else - printf("NULL"); - - printf(","); - - if (rbranch) - _vocab_recursive_ptree_dump_treelike(nodes, rbranch, nr); - else - printf("NULL"); - - printf(">"); - } -} - -void _vocab_recursive_ptree_dump(parse_tree_node_t *nodes, int nr, int prevnr, int blanks) { - int lbranch = nodes[nr].content.branches[0]; - int rbranch = nodes[nr].content.branches[1]; - int i; - - if (nodes[nr].type == kParseTreeLeafNode) { - printf("vocab_dump_parse_tree: Error: consp is nil for element %03x\n", nr); - return; - } - - if ((nr > VOCAB_TREE_NODES)/* || (nr < prevnr)*/) { - printf("Error(%04x))", nr); - return; - } - - if (lbranch) { - if (nodes[lbranch].type == kParseTreeBranchNode) { - printf("\n"); - for (i = 0; i < blanks; i++) - printf(" "); - printf("("); - _vocab_recursive_ptree_dump(nodes, lbranch, nr, blanks + 1); - printf(")\n"); - for (i = 0; i < blanks; i++) - printf(" "); - } else - printf("%x", nodes[lbranch].content.value); - printf(" "); - }/* else printf ("nil");*/ - - if (rbranch) { - if (nodes[rbranch].type == kParseTreeBranchNode) - _vocab_recursive_ptree_dump(nodes, rbranch, nr, blanks); - else - printf("%x", nodes[rbranch].content.value); - }/* else printf("nil");*/ -} - -void vocab_dump_parse_tree(const char *tree_name, parse_tree_node_t *nodes) { - //_vocab_recursive_ptree_dump_treelike(nodes, 0, 0); - printf("(setq %s \n'(", tree_name); - _vocab_recursive_ptree_dump(nodes, 0, 0, 1); - printf("))\n"); -} - -void Vocabulary::dumpParseTree() { - //_vocab_recursive_ptree_dump_treelike(nodes, 0, 0); - printf("(setq parse-tree \n'("); - _vocab_recursive_ptree_dump(_parserNodes, 0, 0, 1); - printf("))\n"); -} - -void Vocabulary::synonymizeTokens(ResultWordList &words) { - if (_synonyms.empty()) - return; // No synonyms: Nothing to check - - for (ResultWordList::iterator i = words.begin(); i != words.end(); ++i) - for (SynonymList::const_iterator sync = _synonyms.begin(); sync != _synonyms.end(); ++sync) - if (i->_group == sync->replaceant) - i->_group = sync->replacement; -} - -void Vocabulary::printParserNodes(int num) { - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - - for (int i = 0; i < num; i++) { - con->DebugPrintf(" Node %03x: ", i); - if (_parserNodes[i].type == kParseTreeLeafNode) - con->DebugPrintf("Leaf: %04x\n", _parserNodes[i].content.value); - else - con->DebugPrintf("Branch: ->%04x, ->%04x\n", _parserNodes[i].content.branches[0], - _parserNodes[i].content.branches[1]); - } -} - -int Vocabulary::parseNodes(int *i, int *pos, int type, int nr, int argc, const char **argv) { - int nextToken = 0, nextValue = 0, newPos = 0, oldPos = 0; - Console *con = ((SciEngine *)g_engine)->getSciDebugger(); - - if (type == kParseNil) - return 0; - - if (type == kParseNumber) { - _parserNodes[*pos += 1].type = kParseTreeLeafNode; - _parserNodes[*pos].content.value = nr; - return *pos; - } - if (type == kParseEndOfInput) { - con->DebugPrintf("Unbalanced parentheses\n"); - return -1; - } - if (type == kParseClosingParenthesis) { - con->DebugPrintf("Syntax error at token %d\n", *i); - return -1; - } - - _parserNodes[oldPos = ++(*pos)].type = kParseTreeBranchNode; - - for (int j = 0; j <= 1; j++) { - if (*i == argc) { - nextToken = kParseEndOfInput; - } else { - const char *token = argv[(*i)++]; - - if (!strcmp(token, "(")) { - nextToken = kParseOpeningParenthesis; - } else if (!strcmp(token, ")")) { - nextToken = kParseClosingParenthesis; - } else if (!strcmp(token, "nil")) { - nextToken = kParseNil; - } else { - nextValue = strtol(token, NULL, 0); - nextToken = kParseNumber; - } - } - - if ((newPos = _parserNodes[oldPos].content.branches[j] = parseNodes(i, pos, nextToken, nextValue, argc, argv)) == -1) - return -1; - } - - const char *token = argv[(*i)++]; - if (strcmp(token, ")")) - con->DebugPrintf("Expected ')' at token %d\n", *i); - - return oldPos; -} - -} // End of namespace Sci diff --git a/engines/sci/vocabulary.h b/engines/sci/vocabulary.h deleted file mode 100644 index 00b8780d1c..0000000000 --- a/engines/sci/vocabulary.h +++ /dev/null @@ -1,338 +0,0 @@ -/* ScummVM - Graphic Adventure Engine - * - * ScummVM is the legal property of its developers, whose names - * are too numerous to list here. Please refer to the COPYRIGHT - * file distributed with this source distribution. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - * - * $URL$ - * $Id$ - * - */ - -#ifndef SCI_SCICORE_VOCABULARY_H -#define SCI_SCICORE_VOCABULARY_H - -#include "common/str.h" -#include "common/hashmap.h" -#include "common/hash-str.h" -#include "common/list.h" - -#include "sci/sci.h" - -namespace Sci { - -class ResourceManager; - -/*#define VOCABULARY_DEBUG */ - -/** Number of bytes allocated on the heap to store bad words if parsing fails */ -#define PARSE_HEAP_SIZE 64 - -enum { - VOCAB_RESOURCE_SELECTORS = 997, - - VOCAB_RESOURCE_SCI0_MAIN_VOCAB = 0, - VOCAB_RESOURCE_SCI0_PARSE_TREE_BRANCHES = 900, - VOCAB_RESOURCE_SCI0_SUFFIX_VOCAB = 901, - - VOCAB_RESOURCE_SCI1_MAIN_VOCAB = 900, - VOCAB_RESOURCE_SCI1_PARSE_TREE_BRANCHES = 901, - VOCAB_RESOURCE_SCI1_SUFFIX_VOCAB = 902 -}; - - -enum { - VOCAB_CLASS_PREPOSITION = 0x01, - VOCAB_CLASS_ARTICLE = 0x02, - VOCAB_CLASS_ADJECTIVE = 0x04, - VOCAB_CLASS_PRONOUN = 0x08, - VOCAB_CLASS_NOUN = 0x10, - VOCAB_CLASS_INDICATIVE_VERB = 0x20, - VOCAB_CLASS_ADVERB = 0x40, - VOCAB_CLASS_IMPERATIVE_VERB = 0x80, - VOCAB_CLASS_NUMBER = 0x001 -}; - -enum { - kParseEndOfInput = 0, - kParseOpeningParenthesis = 1, - kParseClosingParenthesis = 2, - kParseNil = 3, - kParseNumber = 4 -}; - -/* Anywords are ignored by the parser */ -#define VOCAB_CLASS_ANYWORD 0xff - -/* This word class is used for numbers */ -#define VOCAB_MAGIC_NUMBER_GROUP 0xffd /* 0xffe ? */ - -/* Number of nodes for each parse_tree_node structure */ -#define VOCAB_TREE_NODES 500 - -#define VOCAB_TREE_NODE_LAST_WORD_STORAGE 0x140 -#define VOCAB_TREE_NODE_COMPARE_TYPE 0x146 -#define VOCAB_TREE_NODE_COMPARE_GROUP 0x14d -#define VOCAB_TREE_NODE_FORCE_STORAGE 0x154 - -#define SAID_COMMA 0xf0 -#define SAID_AMP 0xf1 -#define SAID_SLASH 0xf2 -#define SAID_PARENO 0xf3 -#define SAID_PARENC 0xf4 -#define SAID_BRACKO 0xf5 -#define SAID_BRACKC 0xf6 -#define SAID_HASH 0xf7 -#define SAID_LT 0xf8 -#define SAID_GT 0xf9 -#define SAID_TERM 0xff - -#define SAID_FIRST SAID_COMMA - -/* There was no 'last matching word': */ -#define SAID_FULL_MATCH 0xffff -#define SAID_NO_MATCH 0xfffe -#define SAID_PARTIAL_MATCH 0xfffd - -#define SAID_LONG(x) ((x) << 8) - -struct ResultWord { - int _class; /**< Word class */ - int _group; /**< Word group */ -}; - -typedef Common::List ResultWordList; - -typedef Common::HashMap WordMap; - - -struct ParseRuleList; - -struct suffix_t { - - int class_mask; /**< the word class this suffix applies to */ - int result_class; /**< the word class a word is morphed to if it doesn't fail this check */ - - int alt_suffix_length; /**< String length of the suffix */ - int word_suffix_length; /**< String length of the other suffix */ - - const char *alt_suffix; /**< The alternative suffix */ - const char *word_suffix; /**< The suffix as used in the word vocabulary */ - -}; - -typedef Common::List SuffixList; - - -struct synonym_t { - int replaceant; /**< The word group to replace */ - int replacement; /**< The replacement word group for this one */ -}; - -typedef Common::List SynonymList; - -struct parse_tree_branch_t { - int id; - int data[10]; -}; - -enum ParseTypes { - kParseTreeLeafNode = 0, - kParseTreeBranchNode = 1 -}; - -struct parse_tree_node_t { - ParseTypes type; /**< leaf or branch */ - union { - int value; /**< For leaves */ - short branches[2]; /**< For branches */ - } content; -}; - -enum VocabularyVersions { - kVocabularySCI0 = 0, - kVocabularySCI1 = 1 -}; - -class Vocabulary { -public: - Vocabulary(ResourceManager *resMan); - ~Vocabulary(); - - /** - * Gets any word from the specified group. For debugging only. - * @param group Group number - */ - const char *getAnyWordFromGroup(int group); - - - /** - * Looks up a single word in the words and suffixes list. - * @param word pointer to the word to look up - * @param word_len length of the word to look up - * @return the matching word (or (-1,-1) if there was no match) - */ - ResultWord lookupWord(const char *word, int word_len); - - - /** - * Tokenizes a string and compiles it into word_ts. - * @param[in] retval A list of words which will be set to the result - * @param[out] sentence The sentence to examine - * @param[out] error Points to a malloc'd copy of the offending text or to NULL on error - * @return true on success, false on failure - * - * On error, false is returned. If *error is NULL, the sentence did not - * contain any useful words; if not, *error points to a malloc'd copy of - * the offending word. The returned list may contain anywords. - */ - bool tokenizeString(ResultWordList &retval, const char *sentence, char **error); - - /** - * Builds a parse tree from a list of words, using a set of Greibach Normal - * Form rules. - * @param words The words to build the tree from - * @param verbose Set to true for debugging - * @return 0 on success, 1 if the tree couldn't be built in VOCAB_TREE_NODES - * nodes or if the sentence structure in 'words' is not part of the - * language described by the grammar passed in 'rules'. - */ - int parseGNF(const ResultWordList &words, bool verbose = false); - - /** - * Constructs the Greibach Normal Form of the grammar supplied in 'branches'. - * @param verbose Set to true for debugging. If true, the list is - * freed before the function ends - * @return Pointer to a list of singly linked GNF rules describing the same - * language that was described by 'branches' - * - * The original SCI rules are in almost-CNF (Chomsky Normal Form). Note that - * branch[0] is used only for a few magical incantations, as it is treated - * specially by the SCI parser. - */ - ParseRuleList *buildGNF(bool verbose = false); - - /** - * Deciphers a said block and dumps its content via printf. - * For debugging only. - * @param pos pointer to the data to dump - */ - void decipherSaidBlock(byte *pos); - - /** - * Prints the parser suffixes to the debug console. - */ - void printSuffixes() const; - - /** - * Prints the parser words to the debug console. - */ - void printParserWords() const; - - uint getParserBranchesSize() const { return _parserBranches.size(); } - const parse_tree_branch_t &getParseTreeBranch(int number) const { return _parserBranches[number]; } - - /** - * Adds a new synonym to the list - */ - void addSynonym(synonym_t syn) { _synonyms.push_back(syn); } - - /** - * Clears the list of synonyms - */ - void clearSynonyms() { _synonyms.clear(); } - - /** - * Synonymizes a token list - * Parameters: (ResultWordList &) words: The word list to synonymize - */ - void synonymizeTokens(ResultWordList &words); - - void printParserNodes(int num); - - void dumpParseTree(); - - int parseNodes(int *i, int *pos, int type, int nr, int argc, const char **argv); - -private: - /** - * Loads all words from the main vocabulary. - * @return true on success, false on failure - */ - bool loadParserWords(); - - /** - * Loads all suffixes from the suffix vocabulary. - * @return true on success, false on failure - */ - bool loadSuffixes(); - - /** - * Frees all suffixes in the given list. - * @param suffixes: The suffixes to free - */ - void freeSuffixes(); - - /** - * Retrieves all grammar rules from the resource data. - * @param branches The rules are stored into this Array - * @return true on success, false on error - */ - bool loadBranches(); - - /** - * Frees a parser rule list as returned by vocab_build_gnf(). - * @param rule_list the rule list to free - */ - void freeRuleList(ParseRuleList *rule_list); - - ResourceManager *_resMan; - VocabularyVersions _vocabVersion; - - // Parser-related lists - SuffixList _parserSuffixes; - ParseRuleList *_parserRules; /**< GNF rules used in the parser algorithm */ - Common::Array _parserBranches; - WordMap _parserWords; - SynonymList _synonyms; /**< The list of synonyms */ - -public: - // Accessed by said() - parse_tree_node_t _parserNodes[VOCAB_TREE_NODES]; /**< The parse tree */ -}; - -/** - * Prints a parse tree. - * @param tree_name Name of the tree to dump (free-form) - * @param nodes The nodes containing the parse tree - */ -void vocab_dump_parse_tree(const char *tree_name, parse_tree_node_t *nodes); - - - -/** - * Builds a parse tree from a spec and compares it to a parse tree. - * @param s The affected state - * @param spec Pointer to the spec to build - * @param verbose Whether to display the parse tree after building it - * @return 1 on a match, 0 otherwise - */ -int said(EngineState *s, byte *spec, bool verbose); - -} // End of namespace Sci - -#endif // SCI_SCICORE_VOCABULARY_H -- cgit v1.2.3