DEVTOOLS: Rename blade_runner to create_bladerunner

1 files changed, 1138 insertions, 0 deletions

diff --git a/devtools/create_bladerunner/subtitles/fontCreator/grabberFromPNG17BR.py b/devtools/create_bladerunner/subtitles/fontCreator/grabberFromPNG17BR.py
new file mode 100644
index 0000000000..a02cd5b7c9
--- /dev/null
+++ b/devtools/create_bladerunner/subtitles/fontCreator/grabberFromPNG17BR.py
@@ -0,0 +1,1138 @@
+#!/usr/bin/python
+# -*- coding: UTF-8 -*-
+#-------------------------------------------------------------------------------
+# Name:		   grabberFromPNG15BR
+# Purpose:	   Parse the character fonts from a PNG file in order to create a
+#			   FON file for the Westwood Blade Runner PC game.
+#
+# Author:	   antoniou
+#
+# Created:	   16-05-2018
+# Copyright:   (c) antoniou 2018
+# Licence:
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the "Software"),
+# to deal in the Software without restriction, including without limitation
+# the rights to use, copy, modify, merge, publish, distribute, sublicense,
+# and/or sell copies of the Software, and to permit persons to whom the
+# Software is furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+# DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE
+# OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+#-------------------------------------------------------------------------------
+
+#
+# Let's assume an Input image with only a row of all character glyphs (no double rows...)
+# BLADE RUNNER:
+# TODO: A way to differentiate between corrupted TAHOMA18 and extended TAHOMA18 (user created)
+# TODO: Maybe instead of removing transparency from glyph ROW image (transparency threshold= 0), use a grey gradient for semitransparent pixels for the cases of TAHOMA fonts (and maybe the subtitle ones too)
+#           Create the font row in GIMP - add layers for shadows if necessary
+#           Merge all layers (maintaining an alpha channel for the background)
+#           Select all and float, which should create a floating selection with all the letters
+#           Promote that to layer and duplicate it.
+#           Choose one of the duplicated layers and COLORIFY it to PITCH BLACK
+#               Set the transparency threshold of THAT black layer to 0
+#           Put this layer under the other and merge the visible layers.
+#           This should get rid of semi-transparent pixels while maintaining the "aliasing" effect. There could be a better way but this should work ok.
+
+
+# TODO: A way to export game fonts to png image (in order to be able to create new expanded fonts keeping glyph consistency!)
+#           Also override the corruption in TAHOMA18 while exporting
+# TODO: Re-Check the order of fonts in (in-game resource font files) TAHOMA18 (stored corrupted) and TAHOMA24 (in good condition).
+# TODO: print a warning for mismatch of number of letters in encoding override (or internal) and detected fonts in ROW IMAGE (especially if we expect a double exclamation mark at the start - and we ignoring one of the two)
+# TODO: Don't crash if mismatch in detected letter (fewer) in row image vs the letters in ascii list of overrideEncoding file
+# TODO: A more detailed readme for this tool and how to use it
+#
+# DONE: enforce overrideEncoding.txt -- this tool should no longer work without one
+# DONE: Test greek subs too
+# DONE: Letter fonts should be spaced by TAB when copied into GIMP or other app to create the image row of all character glyphs
+# DONE: First character should be repeated in the ROW file (but taken into consideration once) in order to get the pixels for the TAB space between letters (left-start column to left-start column)
+# DONE: Use the tab space pixels to calculate the KERNING for each letter (x offset)
+# DONE: update the image segment size bytes in the header after having completed populating the image segment
+# DONE: allow settin explicit kerning and width addon for cases like i and l characters
+# DONE: entrée (boiled dog question) - has an e like goose liver pate --> TESTED
+# DONE: Tested ok "si senor" from peruvian lady / insect dealer too!
+# DONE: ability to manually set kerning (x-offset) for fonts by letter like a list in parameters or in overrideEncoding.txt }  i:1,j:-1,l:1	 (no space or white line characters) - POSITIVE OR NEGATIVE VALUES BOTH ADMITTED
+# DONE: a value of '-' for this means ignore
+# DONE: ability to manually set extra width (additional columns at the end of glyph, with transparent color) for fonts by letter like a list in parameters or in overrideEncoding.txt }	 i:1,j:2,l:1 - POSITIVE VALUES ONLY
+# DONE: make space pixels (var spaceWidthInPixels) into an external param?
+# DONE: INFO NOTE IT IS NOT POSSIBLE TO have partial transparency
+
+import os, sys, shutil
+from os import walk, errno
+import Image
+from struct import *
+import re
+import os.path
+from fonFileLib import *
+
+company_email = "classic.adventures.in.greek@gmail.com"
+app_version = "0.50"
+app_name = "grabberFromPNGHHBR"
+app_name_spaced = "Extract or create Font Files (.FON) for Blade Runner"
+
+
+supportedMIXInputFiles = ['STARTUP.MIX']
+## 4 font files
+supportedExportedFONFiles = ['10PT.FON', 'TAHOMA18.FON', 'TAHOMA24.FON', 'KIA6PT.FON']
+
+def calculateFoldHash(strFileName):
+	i = 0
+	hash = 0
+	strParam = strFileName.upper()
+	lenFileName = len(strParam);
+	while i < lenFileName and i < 12:
+		groupSum = 0
+		# work in groups of 4 bytes
+		for j in range(0, 4):
+			# LSB first, so the four letters in the string are re-arranged (first letter goes to lower place)
+			groupSum >>= 8;
+			if (i < lenFileName):
+				groupSum |= (ord(strParam[i]) << 24)
+				i += 1
+			else: # if	i >= lenFileName  but still haven't completed the four byte loop add 0s
+				groupSum |= 0
+		hash = ((hash << 1) | ((hash >> 31) & 1)) + groupSum
+	hash &= 0xFFFFFFFF	   # mask here!
+	#print (strParam +': '  +''.join('{:08X}'.format(hash)))
+	return hash
+
+class grabberFromPNG:
+	origEncoding = 'windows-1252'
+	defaultTargetLang = "greek"
+	defaultTargetEncoding = 'windows-1253' #greek
+	defaultTargetEncodingUnicode = unicode(defaultTargetEncoding, 'utf-8')
+
+	targetEncoding = 'windows-1253'
+	targetEncodingUnicode = unicode(targetEncoding, 'utf-8')
+
+	overrideEncodingPath = ""
+	originalFontName = ''
+
+	BR_GameID = 3
+	BR_Desc = 'Blade Runner'
+	BR_CodeName = 'BLADERUNNER'
+	BR_DefaultFontFileName = 'SUBTLS_E.FON'
+
+	defaultSpaceWidthInPixelsConst = 0x0007 #0x0008	 #0x0006
+	spaceWidthInPixels = defaultSpaceWidthInPixelsConst
+	specialGlyphMode = True
+	autoTabCalculation = True
+
+	reconstructEntireFont = False # TODO: TRUE!!!
+	minSpaceBetweenLettersInRowLeftToLeft =0
+	minSpaceBetweenLettersInColumnTopToTop = 0
+	kerningForFirstDummyFontLetter = 0
+	yOffsetForAllGlyphsExceptFirstSpecialGamma = 0
+#	 deductKerningPixels = 0
+
+	inputFonMixPath = ""
+	targetFONFilename = BR_DefaultFontFileName
+#	 origFontFilename=""
+	origFontPropertiesTxt = ""
+#	 imageOriginalPNG=""
+	imageRowFilePNG=""
+	copyFontFileName=""
+	copyFontPropertiesTxt = ""
+	copyPNGFileName=""
+
+	lettersFound = 0
+	listOfBaselines = []
+	listOfWidths = []
+	listOfHeights = [] # new for Blade Runner support
+	listOfLetterBoxes = []
+	properListOfLetterBoxes = []
+	startColOfPrevFontLetter = 0 # new for Blade Runner support
+	tabSpaceWidth = 0
+	startOfAllLettersIncludingTheExtraDoubleAndWithKern = 0
+	maxAsciiValueInEncoding	 = 0
+
+	listOfXOffsets = [] # new for Blade Runner support
+	listOfYOffsets = [] # new for Blade Runner support
+
+	listOfExplicitKerning = []
+	listOfWidthIncrements = []
+	listOfOutOfOrderGlyphs = []
+
+	targetLangOrderAndListOfForeignLettersAsciiValues= None
+
+	def initOverrideEncoding(self):
+		##
+		## FOR INIT PURPOSES!!!!
+		##
+		overrideFailed = True
+		targetLangOrderAndListOfForeignLettersStrUnicode = None
+		targetLangOrderAndListOfForeignLettersStr = None
+		# Read from an override file if it exists. Filename should be overrideEncoding.txt (overrideEncodingTextFile)
+		if(not self.overrideEncodingPath):
+			overrideEncodingTextFile = u'overrideEncoding.txt'
+			relPath = u'.'
+			self.overrideEncodingPath = os.path.join(relPath,overrideEncodingTextFile)
+
+		if os.access(self.overrideEncodingPath, os.F_OK):
+			## debug
+			#print "Override encoding file found: {0}.".format(self.overrideEncodingPath)
+			overEncodFile = open(self.overrideEncodingPath, 'r')
+			linesLst = overEncodFile.readlines()
+			overEncodFile.close()
+			if linesLst is None or len(linesLst) == 0:
+				overrideFailed = True
+			else:
+				print "Override Encoding Info: "
+				involvedTokensLst =[]
+				del involvedTokensLst[:] # unneeded
+				for readEncodLine in linesLst:
+					tmplineTokens = re.findall("[^\t\n]+",readEncodLine )
+					for x in tmplineTokens:
+						involvedTokensLst.append(x)
+	#				 print involvedTokensLst
+					#break #only read first line
+				for tokenNameKeyPair in involvedTokensLst:
+					nameKeyTupl = tokenNameKeyPair.split('=', 1)
+					try:
+						if len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'targetEncoding' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '-' and nameKeyTupl[1] != '':
+							self.targetEncodingUnicode = unicode(nameKeyTupl[1], 'utf-8')
+							self.targetEncoding = unicode.encode("%s" % self.targetEncodingUnicode, self.origEncoding)
+						elif len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'asciiCharList' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '':
+							targetLangOrderAndListOfForeignLettersStrUnicode = unicode(nameKeyTupl[1], 'utf-8')
+						elif len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'explicitKerningList' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '-' and nameKeyTupl[1] != '':
+							# split at comma, then split at ':' and store tuples of character and explicit kerning
+							explicitKerningTokenUnicode = unicode(nameKeyTupl[1], 'utf-8')
+							explicitKerningTokenStr = unicode.encode("%s" % explicitKerningTokenUnicode, self.targetEncoding)
+							tokensOfExplicitKerningTokenStrList = explicitKerningTokenStr.split(',')
+							for tokenX in tokensOfExplicitKerningTokenStrList:
+								tokensOfTupleList = tokenX.split(':')
+								self.listOfExplicitKerning.append((ord(tokensOfTupleList[0]), int(tokensOfTupleList[1])) )
+						elif len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'explicitWidthIncrement' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '-' and nameKeyTupl[1] != '':
+							# split at comma, then split at ':' and store tuples of character and explicit additional width - POSITIVE VALUES ONLY
+							explicitWidthIncrementTokenUnicode = unicode(nameKeyTupl[1], 'utf-8')
+							explicitWidthIncrementTokenStr =  unicode.encode("%s" % explicitWidthIncrementTokenUnicode, self.targetEncoding)
+							tokensOfWidthIncrementStrList = explicitWidthIncrementTokenStr.split(',')
+							for tokenX in tokensOfWidthIncrementStrList:
+								tokensOfTupleList = tokenX.split(':')
+								self.listOfWidthIncrements.append((ord(tokensOfTupleList[0]), int(tokensOfTupleList[1])))
+						elif len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'specialOutOfOrderGlyphsUTF8ToAsciiTargetEncoding' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '-' and nameKeyTupl[1] != '':
+							# split at comma, then split at ':' and store tuples of character
+							explicitOutOfOrderGlyphsTokenUnicode = unicode(nameKeyTupl[1], 'utf-8') # unicode(nameKeyTupl[1], 'utf-8')
+							#explicitOutOfOrderGlyphsTokenStr =  unicode.encode("%s" % explicitOutOfOrderGlyphsTokenUnicode, self.targetEncoding)
+							#explicitOutOfOrderGlyphsTokenStr =  explicitOutOfOrderGlyphsTokenUnicode.decode(self.targetEncoding) # unicode.encode("%s" % explicitOutOfOrderGlyphsTokenUnicode, 'utf-8')
+							tokensOfOutOfOrderGlyphsStrList = explicitOutOfOrderGlyphsTokenUnicode.split(',')
+							for tokenX in tokensOfOutOfOrderGlyphsStrList:
+								tokensOfTupleList = tokenX.split(':')
+								self.listOfOutOfOrderGlyphs.append( (unichr(ord(tokensOfTupleList[0])), unichr(ord(tokensOfTupleList[1]))) )
+						elif len(nameKeyTupl) == 2 and nameKeyTupl[0] == 'originalFontName' and nameKeyTupl[1] is not None and nameKeyTupl[1] != '-' and nameKeyTupl[1] != '':
+							self.originalFontName =  nameKeyTupl[1]
+					except:
+						overrideFailed = True
+						raise
+				if not (self.targetEncoding is None or self.targetEncoding  == '' or targetLangOrderAndListOfForeignLettersStrUnicode is None or len(targetLangOrderAndListOfForeignLettersStrUnicode) == 0 or self.originalFontName is None or self.originalFontName == ''):
+					overrideFailed = False
+
+				print "Target Encoding: " , self.targetEncoding
+				#print "Lang Order: " , targetLangOrderAndListOfForeignLettersStrUnicode
+				print "Explicit Kern List: " , self.listOfExplicitKerning
+				print "Explicit Width Increment List: " , self.listOfWidthIncrements
+				print "Original Font Name: " , self.originalFontName
+
+			if(len(self.listOfOutOfOrderGlyphs) == 0 and self.specialGlyphMode == True):
+				# Just keep those that are needed
+				if self.originalFontName == 'SUBTLS_E':
+					self.listOfOutOfOrderGlyphs.append((u'\xed', u'\u0386')) # spanish i (si)
+					self.listOfOutOfOrderGlyphs.append((u'\xf1', u'\xa5')) # spanish n (senor)
+					self.listOfOutOfOrderGlyphs.append((u'\xe2', u'\xa6')) # a for (liver) pate
+					self.listOfOutOfOrderGlyphs.append((u'\xe9', u'\xa7')) # e for (liver) pate
+				elif self.originalFontName == 'TAHOMA': # treat TAHOMA18 and TAHOMA24 similarily here
+					self.listOfOutOfOrderGlyphs.append((u'\xe9', u'\u0192')) 	# french e punctuated
+					self.listOfOutOfOrderGlyphs.append((u'\xfc', u'\u2013')) 	# u umlaut
+			print "Explicit Out Of Order Glyphs List: " , self.listOfOutOfOrderGlyphs
+		else:
+			## debug
+			print "Error: Override encoding file not found: {0}.".format(self.overrideEncodingPath)
+			#print "Override encoding file not found: {0}.".format(self.overrideEncodingFileRelPath)
+			#print "To override the default encoding {0} use an override encoding file with two tab separated entries: encoding (ascii) and characters-list. Convert to UTF-8 without BOM and save. For example:".format(defaultTargetEncoding)
+			#print "windows-1252\t!!\"#$%&\'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}"
+			pass
+
+		if overrideFailed:
+			## debug
+			print "Override encoding file FAILED-1." #" Initializing for {0}...".format(defaultTargetLang)
+	#		 self.targetEncoding = self.defaultTargetEncoding
+	#		 self.targetEncodingUnicode = self.defaultTargetEncodingUnicode
+	#		 targetLangOrderAndListOfForeignLettersStrUnicode = unicode(allOfGreekChars, 'utf-8')
+	#		 #print targetLangOrderAndListOfForeignLettersStrUnicode
+			sys.exit()	# terminate if override Failed (Blade Runner)
+
+		try:
+			targetLangOrderAndListOfForeignLettersStr = unicode.encode("%s" % targetLangOrderAndListOfForeignLettersStrUnicode, self.targetEncoding)
+		except:
+			## debug
+			print "Override encoding file FAILED-2." #"Initializing for {0}...".format(defaultTargetLang)
+	#		 self.targetEncoding = self.defaultTargetEncoding
+	#		 self.targetEncodingUnicode = self.defaultTargetEncodingUnicode
+	#		 targetLangOrderAndListOfForeignLettersStrUnicode = unicode(allOfGreekChars, 'utf-8')
+	#		 targetLangOrderAndListOfForeignLettersStr = unicode.encode("%s" % targetLangOrderAndListOfForeignLettersStrUnicode, self.targetEncoding)
+	#		 #print targetLangOrderAndListOfForeignLettersStrUnicode
+			sys.exit()	# terminate if override Failed (Blade Runner)
+
+		targetLangOrderAndListOfForeignLetters = list(targetLangOrderAndListOfForeignLettersStr)
+		print  targetLangOrderAndListOfForeignLetters, len(targetLangOrderAndListOfForeignLetters)	   # new
+		self.targetLangOrderAndListOfForeignLettersAsciiValues = [ord(i) for i in targetLangOrderAndListOfForeignLetters]
+		print self.targetLangOrderAndListOfForeignLettersAsciiValues, len(self.targetLangOrderAndListOfForeignLettersAsciiValues)
+		self.maxAsciiValueInEncoding = max(self.targetLangOrderAndListOfForeignLettersAsciiValues)
+
+#	 for charAsciiValue in targetLangOrderAndListOfForeignLetters:
+#		 print "Ord of chars: %d" % ord(charAsciiValue)
+
+	##
+	## END OF INIT CODE
+	##
+
+#
+# TODO: warning: assumes that there is a margin on top and bellow the letters
+# (especially on top; if a letter starts from the first pixel row,
+# it might not detect it!) <---to fix
+#
+	def __init__(self, pselectedEncoding=None):
+		self.minSpaceBetweenLettersInRowLeftToLeft = 0
+		self.minSpaceBetweenLettersInColumnTopToTop = 0
+		self.kerningForFirstDummyFontLetter = 0
+		self.yOffsetForAllGlyphsExceptFirstSpecialGamma = 0
+		self.spaceWidthInPixels = self.defaultSpaceWidthInPixelsConst
+#		 self.deductKerningPixels = 0
+		self.reconstructEntireFont = False # TODO : True?
+		#self.origFontFilename=porigFontFilename
+		self.inputFonMixPath = ""
+		self.targetFONFilename = self.BR_DefaultFontFileName
+		self.copyFontFileName = ""
+		self.copyPNGFileName=""
+		#self.imageOriginalPNG=pimageOriginalPNG
+		self.imageRowFilePNG = ""
+		self.baselineOffset = 0
+
+		self.lettersFound = 0
+		self.origFontPropertiesTxt = ""
+		self.copyFontPropertiesTxt = ""
+		self.cleanup() # for good practice (this should not be here, but it's here too due to quick/sloppy :) coding (TODO: fix it)
+		#debug
+		#self.DBinit() # TODO REMOVE?
+		if pselectedEncoding == None:
+			pselectedEncoding = self.targetEncoding
+
+		self.selectedGameID = self.BR_GameID
+		self.activeEncoding = pselectedEncoding
+		self.lettersInOriginalFontFile = 0 #initialization
+
+		# TODO: we should get from the DB the encoding and lettersString
+		# and the Empty slots for the selected Game and Calculate the rest of
+		# the lists/dictionaries on the fly.
+		# IF no lettersString or encoding has been defined...
+		# then issue error? or continue with hard-coded
+		# (insert to db as well and inform the GUI?)
+		#self.calcFromDB() # TODO REMOVE?
+		return
+
+
+	def cleanup(self):
+		self.lettersFound = 0
+		self.startColOfPrevFontLetter = 0
+		self.tabSpaceWidth = 0
+		self.startOfAllLettersIncludingTheExtraDoubleAndWithKern = 0
+		del self.listOfBaselines[:]
+		del self.listOfWidths[:]
+		del self.listOfHeights[:]
+		del self.listOfLetterBoxes[:]
+		del self.properListOfLetterBoxes[:]
+		del self.listOfXOffsets[:] # new for Blade Runner support
+		del self.listOfYOffsets[:] # new for Blade Runner support
+#		 del self.listOfExplicitKerning[:] # don't clean these up
+#		 del self.listOfWidthIncrements[:] # don't clean these up
+
+		self.origFontPropertiesTxt = ""
+		self.copyFontPropertiesTxt = ""
+		return
+
+##
+## SETTERS
+##
+	def setInputPathForFonMix(self, pInputMixPath):
+		self.inputFonMixPath = pInputMixPath
+		return
+
+	def setImageRowFilePNG(self, pimageRowFilePNG):
+		self.imageRowFilePNG = pimageRowFilePNG
+		return
+
+	def setTargetFONFilename(self, pTargetFONFilename):
+		self.targetFONFilename = pTargetFONFilename
+		return
+
+	def setMinSpaceBetweenLettersInRowLeftToLeft(self, pminSpaceBetweenLettersInRowLeftToLeft):
+		self.minSpaceBetweenLettersInRowLeftToLeft = pminSpaceBetweenLettersInRowLeftToLeft
+		return
+	def setMinSpaceBetweenLettersInColumnTopToTop(self, pminSpaceBetweenLettersInColumnTopToTop):
+		self.minSpaceBetweenLettersInColumnTopToTop = pminSpaceBetweenLettersInColumnTopToTop
+		return
+
+	def setKerningForFirstDummyFontLetter(self, pKerningForFirstDummyFontLetter):
+		self.kerningForFirstDummyFontLetter = pKerningForFirstDummyFontLetter
+		return
+
+	def setYOffsetForAllGlyphsExceptFirstSpecialGamma(self, pYOffsetForAllGlyphsExceptFirstSpecialGamma):
+		self.yOffsetForAllGlyphsExceptFirstSpecialGamma = pYOffsetForAllGlyphsExceptFirstSpecialGamma
+		return
+
+#	 def setDeductKerningPixels(self, pDeductKerningPixels):
+#		 self.deductKerningPixels = pDeductKerningPixels
+#		 return
+
+	def setSpaceWidthInPixels(self, pSpaceWidthInPixels):
+		self.spaceWidthInPixels = pSpaceWidthInPixels
+		return
+
+	def setSpecialGlyphMode(self, pSpecialGlyphMode):
+		self.specialGlyphMode = pSpecialGlyphMode
+		return
+
+
+	def setAutoTabCalculation(self, pAutoTabCalculation):
+		self.autoTabCalculation = pAutoTabCalculation
+		return
+
+	def setOverrideEncodingPath(self, pOverrideEncodingPath):
+ 		self.overrideEncodingPath = pOverrideEncodingPath
+		return
+
+##
+## END OF SETTERS
+##
+
+
+
+	def parseImage(self,  loadedImag, imwidth, imheight, trimTopPixels=0, trimBottomPixels = 0, firstDoubleLetterIgnore = False):
+		""" parsing input image and detect one character font per run, and deleting the detected character font after calculating its specs (this is done in-memory; we are not writing back to the file)
+		"""
+		prevColStartForLetter = 0
+		prevRowStartForLetter = 0
+		startCol = 0
+		startRow = 0
+		endCol = 0
+		endRow = 0
+		for x in range(0, imwidth):		 # for each column
+			if startCol != 0:
+				break
+			for y in range(0, imheight):	 # we search all rows (for each column)
+				r1,g1,b1,a1 = loadedImag[x, y]
+				if a1 != 0:	 # if pixel not completely transparent -- this is not necessarily the *top* left pixel of a font letter though! -- the startRow is still to be determined.
+	#				 print loadedImag[x, y]
+					if prevColStartForLetter == 0:
+						prevColStartForLetter = x
+						prevRowStartForLetter = y
+						startCol = x
+	#					 print "Letter found"
+	#					 print "start col:%d" % startCol
+	# #					   print "hypothe row:%d" % y
+	#					 # starting from the first row of the row-image (to do optimize), we parse by rows to find the top point (row coordinate) of the character font
+	 #					 for y2 in range(0, y+1):
+						tmpSum = y + self.minSpaceBetweenLettersInColumnTopToTop
+						scanToRow = imheight		  # - explicitly set this to the whole image height -- assumes only one row of character fonts
+						if tmpSum < imheight:	   # TODO: WAS scanToRow < imheight but this doesn't get executed anymore due to explicitly setting scanToRow to imheight (assuming only one row of character fonts)
+													 # DONE: NEW changed check to if tmpSum < imheight which makes more sense
+							scanToRow = tmpSum
+						for y2 in range(0, scanToRow):		   # a loop to find the startRow  - Check by row (starting from the top of the image and the left column of the letter)
+							if startRow != 0:
+								break
+							tmpSum = startCol + self.minSpaceBetweenLettersInRowLeftToLeft
+							scanToCol = imwidth
+							if tmpSum < imwidth:
+								scanToCol = tmpSum
+							#print (startCol, scanToCol)
+							for x2 in range(startCol, scanToCol): # check all columns (for each row)
+								#print loadedImag[x2, y2]
+								r2,g2,b2,a2 = loadedImag[x2, y2]
+								if a2 != 0 and startRow == 0:
+									startRow = y2 + trimTopPixels
+	#								 print "start row: %d" % startRow
+									break
+		if startCol > 0 and startRow > 0:		   # WARNING: TODO NOTE: SO NEVER HAVE AN INPUT IMAGE WHERE THE FONT CHARACTERS ARE TOUCHING THE TOP OF THE IMAGE WITH NO EMPTY SPACE WHATSOEVER
+			tmpSum = startRow + self.minSpaceBetweenLettersInColumnTopToTop
+			scanToRow = imheight
+			if tmpSum < imheight:
+			   scanToRow = tmpSum
+			tmpSum = startCol + self.minSpaceBetweenLettersInRowLeftToLeft
+			scanToCol = imwidth
+			if tmpSum < imwidth:
+			   scanToCol = tmpSum
+			for y in range(startRow, scanToRow):		# now check per row (we go through all theoretical rows, no breaks)-- we want to find the bottom row
+				for x in range(startCol, scanToCol):	# check the columns for each row
+					r1,g1,b1,a1 = loadedImag[x, y]
+					if a1 != 0:
+						endRow = y
+			if endRow > 0:
+				endRow = endRow - trimBottomPixels
+	#		 print "end row:% d" %endRow
+
+		if startCol > 0 and startRow > 0 and endRow > 0:
+			tmpSum = startCol + self.minSpaceBetweenLettersInRowLeftToLeft
+			scanToCol = imwidth
+			if tmpSum < imwidth:
+			   scanToCol = tmpSum
+			for x in range(startCol, scanToCol):	# now check per column (we go through all theoretical columns, no breaks) -- we want to find the bottom column
+				for y in range(startRow, endRow+1): # check the rows for each column
+					r1,g1,b1,a1 = loadedImag[x, y]
+					#print	loadedImag[x, y]
+					if a1 != 0:
+						endCol = x
+	#		 print "end col:% d" %endCol
+		if startCol > 0 and startRow > 0 and endRow > 0 and endCol > 0:
+			# append deducted baseline
+			#
+			if firstDoubleLetterIgnore == True:
+				self.startOfAllLettersIncludingTheExtraDoubleAndWithKern = startCol - self.kerningForFirstDummyFontLetter
+			else:							  # firstDoubleLetterIgnore == False
+				if self.autoTabCalculation == True:
+					if self.tabSpaceWidth == 0:
+						#print "start startPre", startCol, self.startColOfPrevFontLetter
+						self.tabSpaceWidth = startCol - self.startColOfPrevFontLetter
+						print "Tab Space Width detected: %d " % (self.tabSpaceWidth)
+					# new if -- don't use else here, to include the case of when we first detected the tab space width
+					if self.tabSpaceWidth > 0:
+						self.listOfXOffsets.append(startCol - (self.startOfAllLettersIncludingTheExtraDoubleAndWithKern + (self.lettersFound + 1) * self.tabSpaceWidth) ) #	 + self.deductKerningPixels )
+						#print "xOffSet", startCol - (self.startOfAllLettersIncludingTheExtraDoubleAndWithKern + (self.lettersFound + 1) * self.tabSpaceWidth)
+				else:
+					self.listOfXOffsets.append(0)
+
+
+
+				self.listOfBaselines.append(endRow)
+				self.listOfWidths.append(endCol-startCol + 1) # includes the last col (TODO this was without the +1 for MI:SE translator -- possible bug? did we compensate?)
+				self.listOfHeights.append(endRow - startRow + 1) # +1 includes the last row
+				self.listOfLetterBoxes.append((startCol, startRow, endCol, endRow))
+
+			self.startColOfPrevFontLetter = startCol	  #update for next pass
+			#delete the letter - even in the case of ignoring the first double letter
+			for x in range(startCol, endCol+1):
+				for y in range(startRow - trimTopPixels, endRow+1 + trimBottomPixels):
+				   loadedImag[x, y] = 0, 0, 0, 0
+			return 0
+		else: return -1
+#
+#
+#
+	def generateModFiles(self, customBaselineOffs):
+		""" Generate font (FON) files (work on copies, not the originals). Return values: 0 no errors, -1 output font file has alrady new letters, -2 no fonts found in png (TODO: more error cases)
+		"""
+		#
+		# When a customBaselineOffs is set, we should expand the space for the letter (otherwise it will be overflown in the next line or truncated (outside the png)
+		# We can't expand the space for the letter downwards, because the engine will (com)press the new height to fit its expected and it will look bad.
+		# NO THAT WON'T WORK--> The font will remain in the wrong place: Should we	CUT from the top and hope that we don't trunctate!? Keeping the resulting height equal to the expected one?
+		# MAYBE: Do not alter the baseline of the original file, but the detected (popular) one of the line file!!!
+		retVal = 0
+		totalFontLetters = 0
+		importedNumOfLetters = 0
+		errMsg = ""
+		errorFound = False
+		im = None
+		pix = None
+		pixReloaded = None
+		#
+		# CONSTANTS
+		#
+		origGameFontDiakenoHeight = 0
+		interLetterSpacingInPNG = 4
+		origGameFontSizeEqBaseLine = 0
+		# offset for start of PNG index table
+#		 firstTableLineOffset = self.PNG_TABLE_STARTLINE_OFFSET
+		lettersInOriginalFontFile = self.lettersInOriginalFontFile
+		#
+		# detection of origGameFontSizeEqBaseLine
+		#
+		#origGameFontSizeEqBaseLine = self.findDetectedBaseline() ## NEW BR REMOVED
+		self.cleanup() # necessary after detection of baseline, because it fills up some of the	 lists used in the following!
+
+##		  self.origFontPropertiesTxt = self.getImagePropertiesInfo(True) # "%dx%dx%s" % (im.size[0],im.size[1], im.mode) # NEW REMOVED
+#		 print "WEEEE::: ", self.imageOriginalPNG, im.format, "%dx%d" % im.size, im.mode
+#		 print "BASELINE DETECTED:%d " % origGameFontSizeEqBaseLine
+
+		#
+		# OPEN THE IMAGE WITH THE ROW OF CHARACTER FONTS TO BE IMPORTED
+		#
+		if os.access(self.imageRowFilePNG, os.F_OK) :
+			try:
+				im = Image.open(self.imageRowFilePNG)
+			except:
+				errMsg = "No letters were found in input png - IO exception!"
+				print errMsg
+				retVal = -2
+				errorFound = True
+		else:
+			errMsg = "No letters were found in input png - IO fault!"
+			print errMsg
+			retVal = -2
+			errorFound = True
+		if not errorFound:
+			#debug
+			#print self.imageRowFilePNG, im.format, "%dx%d" % im.size, im.mode
+			w1, h1 = im.size
+			trimTopPixels = 0
+			trimBottomPixels = 0
+			italicsMode = False	  # will be set to true only if the prefix of the row file is itcrp_ or it_ in order to activate some extra settings for kerning and letter width!
+			# TODO the note about special handling of row PNG files with it_ or itcrp_ prefix, should be moved to the documentation
+			# TODO the special settings for handling italic native letters should be in the settings(?)
+			filepathSplitTbl = os.path.split(self.imageRowFilePNG)
+			sFilenameOnlyImageRowFilePNG = filepathSplitTbl[1]
+
+			if sFilenameOnlyImageRowFilePNG.startswith("itcrp_") or sFilenameOnlyImageRowFilePNG.startswith("it_"):
+				italicsMode = True
+
+			if sFilenameOnlyImageRowFilePNG.startswith("itcrp_"):
+				trimTopPixels = 1
+				trimBottomPixels = 1
+				print "Will trim upper line by %d pixels and bottom line by %d pixels" % (trimTopPixels, trimBottomPixels)
+			pix = im.load()
+			# pix argument is mutable (will be changed in the parseImage body)
+			if self.parseImage(pix, w1, h1, trimTopPixels, trimBottomPixels, True) == 0:	#first run, just get the start column, ignore the letter - don't store it . We need this for the tab space width calculation and eventually the kerning calc of the letters
+				# after the first call, we got an update on self.startColOfPrevFontLetter using the dummy double firstg letter font
+				while self.parseImage(pix, w1, h1, trimTopPixels, trimBottomPixels) == 0:
+					self.lettersFound = self.lettersFound + 1 # == 0 means one character font was detected so +1 to the counter
+		#	 print self.listOfBaselines
+			#debug
+			print "Font Letters Detected (not including the first double): %d" % (self.lettersFound)
+			if (self.lettersFound ) > 0 :
+				#print "widths: ", self.listOfWidths[0:]
+				#print "Plain x offsets:"
+				#print zip(self.targetLangOrderAndListOfForeignLettersAsciiValues[1:], self.listOfXOffsets)
+#				 # normalize x offsets
+#				 minXoffset = min(self.listOfXOffsets)
+#				 if(minXoffset < 0):
+#					 addNormalizer = minXoffset * (-1)
+#					 self.listOfXOffsets = [ x + addNormalizer	for x in self.listOfXOffsets]
+#				 print "Normalized x offsets: "
+#				 print self.listOfXOffsets
+				# calculate y offsets
+				(listOfStartCols, listOfStartRows, listOfEndCols, listOfEndRows) = zip(* self.listOfLetterBoxes)
+				minTopRow = min(listOfStartRows)
+				self.listOfYOffsets = [ x - minTopRow for x in listOfStartRows]
+				if (self.yOffsetForAllGlyphsExceptFirstSpecialGamma != 0):
+					self.listOfYOffsets =  [ x + self.yOffsetForAllGlyphsExceptFirstSpecialGamma for x in self.listOfYOffsets]
+				#print "Y offsets: "
+				#print self.listOfYOffsets
+				#
+				#
+				# # actually explicit Width setting could affect this so calc a new list here with final widths and get the max on that list!
+				#
+				listOfCalcWidths = []
+				kIncIndx = 1
+				for tmpWidth in self.listOfWidths:
+					explicitWidthIncrementVal = 0
+					if len(self.listOfWidthIncrements ) > 0:
+						tmpOrd = self.targetLangOrderAndListOfForeignLettersAsciiValues[kIncIndx]
+						keysOfWidthIncrements, valuesOfWidthIncrements = (zip(*self.listOfWidthIncrements))
+						if tmpOrd in keysOfWidthIncrements:
+							print "Explicit width increment for %d: %d" % (tmpOrd, valuesOfWidthIncrements[keysOfWidthIncrements.index(tmpOrd)])
+							explicitWidthIncrementVal = valuesOfWidthIncrements[keysOfWidthIncrements.index(tmpOrd)]
+							listOfCalcWidths.append(tmpWidth + explicitWidthIncrementVal )
+					if explicitWidthIncrementVal == 0:
+						listOfCalcWidths.append(tmpWidth)
+					kIncIndx = kIncIndx + 1
+				#maxFontWidth =	 max(self.listOfWidths)
+				maxFontWidth =	max(listOfCalcWidths)
+				maxFontHeight =	 max(self.listOfHeights)
+				print "Max Width, Max Height (not necessarily for the same character font): %d, %d" % (maxFontWidth, maxFontHeight)
+				#print "Index\tAsciiOrd\tX Offs\tY Offs\tWidth\tHeight"
+				#print zip(range(1, len(self.listOfXOffsets)), self.targetLangOrderAndListOfForeignLettersAsciiValues[1:], self.listOfXOffsets, self.listOfYOffsets, listOfCalcWidths, self.listOfHeights)
+				targetFontFile = None
+				try:
+					targetFontFile = open(self.targetFONFilename, 'wb')
+				except:
+					errorFound = True
+				if not errorFound:
+				# reopen the image with our Fonts because we deleted the letters in the in-mem copy
+					im = None
+					if os.access(self.imageRowFilePNG, os.F_OK) :
+						try:
+							im = Image.open(self.imageRowFilePNG)
+						except:
+							errorFound = True
+					else:
+						errorFound = True
+					if not errorFound:
+						pixReloaded = None
+						pixReloaded = im.load()
+
+						# first 4 bytes are the max ascii char value supported (it's basically the number of entries in the character index table)
+						# next 4 bytes are max font char width	(pixels)
+						# next 4 bytes are max font char height (pixels)
+						# next 4 bytes give the size of the graphic segment for the font characters (this size is in word units, so it needs *2 to get the byte size)
+						#				this size should be updated at the end (after filling the file with all font image data)
+						#
+						# pack 'I' unsigned int
+						print "NumberOfEntriesInFontTabl", (self.maxAsciiValueInEncoding + 1 + 1)
+						numberOfEntriesInFontTable = self.maxAsciiValueInEncoding + 1 + 1  # 0x0100 # This is actually the max ascii value + plus one (1) to get the font index value + plus another one (1) to get the count (since we have zero based indices)
+															# TODO ??? could be more than this if we need to keep other characters (not in our codeset) and expand the ascii table and offset the new characters
+						numberOfEntriesInFontTableInFile = pack('I', numberOfEntriesInFontTable )
+						targetFontFile.write(numberOfEntriesInFontTableInFile)
+						maxFontWidthPixelsToWrite = pack('I', maxFontWidth)
+						targetFontFile.write(maxFontWidthPixelsToWrite)
+						maxFontHeightPixelsToWrite = pack('I', maxFontHeight)
+						targetFontFile.write(maxFontHeightPixelsToWrite)
+						fontImagesSegmentSize = pack('I', 0x0000) #	 - to be updated at the end!
+						targetFontFile.write(fontImagesSegmentSize)
+
+						startOfImageSegmentAbs = 0x10 + 20 * numberOfEntriesInFontTable # header is 0x10 bytes. Then table of  20 bytes *	numberOfEntriesInFontTable and then the data.
+						lastImageSegmentOffset = 0
+	#					 targetFontFile.close() # don't close here
+						#
+						# Fonts index table - properties and offset in image segment
+						# TODO - REVISE WHEN FINISHED WITH COMPLETE TRANSCRIPT for special glyphs
+						# So far additional required characters (not included in the standard ASCII (127 chars) are:
+						# the spanish i (put it in ASCII value 0xA2 (162), font index 0xA3)? todo verify -- actual ASCII value in codepage 1252 is 0xED (237)
+						# the spanish n (put it in ASCII value 0xA5 (165), font index 0xA6)? todo verify -- actual ASCII value in codepage 1252 is 0xF1 (241)
+						# DONE we also need special fonts for liver	 pâté
+						#											a actual ASCII value is 0xE2 (226) in codepage 1252 -- put it in ASCII value 0xA6 (166) -- font index 0xA7
+						#											e actual ASCII value is 0xE9 (233) in codepage 1252 -- put it in ASCII value 0xA7 (167) -- font index 0xA8
+						# In the row png font glyphs, the letter glyphs (images) should be the actual character fonts (spanish n, i etc)
+						#				but in the overrideEncoding.txt we need the corresponding ASCII characters for the particular codepage of the text (eg here the greek windows-1253)
+						#
+						# NOTE! WARNING: We need to add the corresponding ASCII characters for our codepage (eg for Windows 1253 the characters with value 0xA2 and 0xA5 which are not the spanish characters but will act as delegates for them)
+						# the greek Ά (alpha tonoumeno) character has ASCII value 0xA2 (162) (in codeset Windows 1253) so conflict with spanish i in in-game Tahoma -- put it in 0xA3 (163) (font index 0xA4)
+						# We should fill all unused characters with "space" placeholder. Probably all of them should also point to the same area (second block) of the image segment too.
+						# First block of the image area (font index = 0) is reserved and should be the "border" gamma-like character.
+						#
+						# Kerning of the first letter font is '1' for Tahoma18 (when shadowed from every side (the left side shadow reduces the kerning), otherwise it would be 2) -- TODO for now this should be a launch parameter
+						# Y offset should be calculated from the top row of the heighest font
+						#kIncIndx = 0
+						## aux list because the order of the fonts in our overrideEncoding may not match the real order in the ascii table
+                        #listOfWriteableEntries = [ (0,0,0,0,0) for i in range(0, numberOfEntriesInFontTable)] # create a list of placeholders for the number of entries we need to write
+						#print " *************** DBG **************"
+						#print listOfWriteableEntries
+						#print " *************** DBG **************"
+						del self.properListOfLetterBoxes[:]
+						for i in range(0, numberOfEntriesInFontTable):	# blocks of 20 bytes
+							# 20 byte block
+							# 4 bytes x offset (from what ref point? is this for kerning ) - CAN THIS BE NEGATIVE?
+							# 4 bytes y offset (from what ref point? is this for the baseline?) - CAN THIS BE NEGATIVE?
+							# 4 bytes char width
+							# 4 bytes char height
+							# 4 bytes offset in image segment (units in words (2 bytes))
+
+							# TODO add all standard ascii characters in the ROW IMAGE before the additional required spanish and then GREEK alphabet characters --
+							#											-- greek Ά should be at its proper place (between spanish i and spanish n).
+							# TODO check possible support issues for ώ greek character
+							if i == 0:
+								# the first entry is a special font character of max width and max height with a horizontal line across the top-most row and a vertical line across the left-most column
+								tmpXOffsetToWrite = pack('I', 0x0000)
+								targetFontFile.write(tmpXOffsetToWrite)
+								tmpYOffsetToWrite = pack('I', 0x0000)
+								targetFontFile.write(tmpYOffsetToWrite)
+								tmpWidthToWrite = pack('I', maxFontWidth)
+								targetFontFile.write(tmpWidthToWrite)
+								tmpHeightToWrite = pack('I', maxFontHeight)
+								targetFontFile.write(tmpHeightToWrite)
+								tmpDataOffsetToWrite = pack('I', 0x0000) # start of image segment means 0 offset
+								targetFontFile.write(tmpDataOffsetToWrite)
+								# TODO maybe conform more with game's format: Eg Tahoma24.fon (native game resource) does not always point to the second character font offset for dummy entries, but to the latest offset and only additionally sets the x-offset property (all others are 0) - eg look for 0x74c9 offsets (byte sequence 0xc9 0x74)
+								dummyCharFontImageConstOffset = maxFontWidth * maxFontHeight; # const. actual offset in bytes is twice that. This counts in words (2-bytes)	 - This points to the first valid entry but with properties that make it translate as a space or dummy(?)
+								lastImageSegmentOffset = maxFontWidth * maxFontHeight; # actual offset in bytes is twice that. This counts in words (2-bytes)
+                                #listOfWriteableEntries[0] = (tmpXOffsetToWrite, tmpYOffsetToWrite, tmpWidthToWrite, tmpHeightToWrite, tmpDataOffsetToWrite)
+							else:
+								if (i-1) in self.targetLangOrderAndListOfForeignLettersAsciiValues:
+									# then this is an actual entry
+									# i-1 is the order of an ascii character, that should be placed in the next slot in the output FON file
+									# but this ascii character in the input overrideEncoding could be not it the same i spot -- because of the correspondance to an out-of-order PNG row file
+									kIncIndxLst = [item for item in enumerate(self.targetLangOrderAndListOfForeignLettersAsciiValues[1:], 0) if item[1] == (i-1) ]
+									kIncIndx = kIncIndxLst[0][0]
+									#kIncIndx = self.targetLangOrderAndListOfForeignLettersAsciiValues.index(i-1)
+									#print kIncIndxLst
+									#print kIncIndx, i-1
+									#print i, ": actual entry index of ascii char", (i-1)," width:", self.listOfWidths[kIncIndx]
+									#print "Self explicit kerning list: " , self.listOfExplicitKerning
+									if len(self.listOfExplicitKerning ) > 0:
+										keysOfExplicitKerning, valuesOfExplicitKerning = (zip(*self.listOfExplicitKerning))
+										if (i - 1) in keysOfExplicitKerning:
+											# found explicit kerning for this
+											print "Explicit kerning for %d " % (i-1)
+											self.listOfXOffsets[kIncIndx] = valuesOfExplicitKerning[keysOfExplicitKerning.index(i-1)] # explicit X offset
+
+									tmpXOffsetToWrite = pack('i', self.listOfXOffsets[kIncIndx]) # x offset - from left			# TODO check if ok. Changed to signed int since it can be negative sometimes!
+									targetFontFile.write(tmpXOffsetToWrite)
+									tmpYOffsetToWrite = pack('I', self.listOfYOffsets[kIncIndx])   # y offset from topmost row
+									targetFontFile.write(tmpYOffsetToWrite)
+
+									if len(self.listOfWidthIncrements ) > 0:
+										keysOfWidthIncrements, valuesOfWidthIncrements = (zip(*self.listOfWidthIncrements))
+										if (i - 1) in keysOfWidthIncrements:
+											print "Explicit width increment for %d " % (i-1)
+											foundExplicitWidthIncrement = True
+			 								self.listOfWidths[kIncIndx] = self.listOfWidths[kIncIndx] + valuesOfWidthIncrements[keysOfWidthIncrements.index(i-1)]
+
+									tmpWidthToWrite = pack('I', self.listOfWidths[kIncIndx] )
+									targetFontFile.write(tmpWidthToWrite)
+									tmpHeightToWrite = pack('I', self.listOfHeights[kIncIndx])
+									targetFontFile.write(tmpHeightToWrite)
+									tmpDataOffsetToWrite = pack('I', lastImageSegmentOffset) #
+									targetFontFile.write(tmpDataOffsetToWrite)
+									lastImageSegmentOffset = lastImageSegmentOffset +  self.listOfWidths[kIncIndx]  * self.listOfHeights[kIncIndx]
+									#kIncIndx = kIncIndx + 1 # increases only for valid characters
+									#
+									# populate self.properListOfLetterBoxes here
+									#
+									self.properListOfLetterBoxes.append(self.listOfLetterBoxes[kIncIndx])
+								else:
+									#
+									#print i, ": phony entry"
+									# TODO in-game resource fonts don't point all to the first entry as dummy but to the last valid entry encountered
+									tmpXOffsetToWrite = pack('I', 0x0000) # 0 x offset
+									targetFontFile.write(tmpXOffsetToWrite)
+									tmpYOffsetToWrite = pack('I', 0x0000)	# 0 y offset
+									targetFontFile.write(tmpYOffsetToWrite)
+									tmpWidthToWrite = pack('I', self.spaceWidthInPixels) # font width set for some pixels of space
+									targetFontFile.write(tmpWidthToWrite)
+									tmpHeightToWrite = pack('I', 0x0000)
+									targetFontFile.write(tmpHeightToWrite)
+									tmpDataOffsetToWrite = pack('I', dummyCharFontImageConstOffset) #
+									targetFontFile.write(tmpDataOffsetToWrite)
+						# end of for loop over all possible ascii values contained in the fon file
+						# print the corrected properties per glyph font:
+						print "***** FINAL (Explicit Kern, width accounted) *****\nIndex\tAsciiOrd\tX Offs\tY Offs\tWidth\tHeight"
+						tmpListOfTuplesToPrintDbg = zip(range(1, len(self.listOfXOffsets)), self.targetLangOrderAndListOfForeignLettersAsciiValues[1:], self.listOfXOffsets, self.listOfYOffsets, listOfCalcWidths, self.listOfHeights)
+						for itemDbg in tmpListOfTuplesToPrintDbg:
+							print "%4d\t%8d\t%6d\t%6d\t%6d\t%6d" % (itemDbg[0], itemDbg[1], itemDbg[2], itemDbg[3], itemDbg[4], itemDbg[5])
+
+						#
+						#
+						# Now fill in the image segment
+						# Fonts are written from TOP to Bottom, Left to Right. Each pixel is 16 bit (2 bytes). Highest bit seems to determine transparency (on/off flag).
+						#
+						# There seem to be 5 bits per RGB channel and the value is the corresponding 8bit value (from the 24 bit pixel color) shifting out (right) the 3 LSBs
+						# NOTE: Since we can't have transparency at channel level(?), it's best to have the input PNG not have transparent colored pixels (in Gimp merge the font layers, foreground and shadow and then from Layer settings set transparency threshold to 0 for that layer)- keep the background transparent!
+						#
+						# First font image is the special character (border of top row and left column) - color of font pixels should be "0x7FFF" for filled and "0x8000" for transparent
+						#
+						#
+						# Then follow up with the image parts for each letter!
+						#
+						#
+						#
+						# START of First special character image segment
+						#
+						for i in range(0, maxFontWidth * maxFontHeight):
+							if(i < maxFontWidth or i % maxFontWidth == 0):
+								tmpPixelColorRGB555ToWrite = pack('H', 0x7FFF) #unsigned short - 2 bytes
+								targetFontFile.write(tmpPixelColorRGB555ToWrite)
+							else:
+								tmpPixelColorRGB555ToWrite = pack('H', 0x8000)
+								targetFontFile.write(tmpPixelColorRGB555ToWrite) # unsigned short - 2 bytes
+						#
+						# END of First special character image segment
+						#
+						#
+						# Start rest of the font characters image segments
+						#
+						#
+
+						#
+						# If we have a character with explicit width increment (y) we should add columns of transparent colored pixels at the end (so since this is done by row, we should add y number of transparent pixels at the end of each row)
+						kIncIndx = 1 # start after the first glyph (which is DOUBLE)
+						for (c_startCol, c_startRow, c_endCol, c_endRow) in self.properListOfLetterBoxes[0:]:
+							#print (c_startCol, c_startRow, c_endCol, c_endRow),'for letter ', self.targetLangOrderAndListOfForeignLettersAsciiValues[kIncIndx]
+							explicitWidthIncrementVal = 0
+							if len(self.listOfWidthIncrements ) > 0:
+								tmpOrd = self.targetLangOrderAndListOfForeignLettersAsciiValues[kIncIndx]
+								keysOfWidthIncrements, valuesOfWidthIncrements = (zip(*self.listOfWidthIncrements))
+								if tmpOrd in keysOfWidthIncrements:
+									#print "Explicit width increment for %d: %d" % (tmpOrd, valuesOfWidthIncrements[keysOfWidthIncrements.index(tmpOrd)])
+									explicitWidthIncrementVal = valuesOfWidthIncrements[keysOfWidthIncrements.index(tmpOrd)]
+
+
+							for tmpRowCur in range(c_startRow, c_endRow + 1):
+								for tmpColCur in range(c_startCol, c_endCol +1):
+									#print (tmpRowCur, tmpColCur)
+									r1,g1,b1,a1 = pixReloaded[tmpColCur, tmpRowCur] # Index col first, row second for image pixel array. TODO asdf this pix has been modified. All pixels would be transparent? - load image again?
+									if(a1 == 0):
+#										 print "with alpha 8bit:", (r1, g1, b1, a1)
+										#make completely transparent - write 0x8000
+										tmpPixelColorRGB555ToWrite = pack('H', 0x8000)
+										targetFontFile.write(tmpPixelColorRGB555ToWrite) # unsigned short - 2 bytes
+									else:	  # alpha should be 255 here really.
+										#print "8bit:", (r1, g1, b1)
+										tmp5bitR1 = (r1 >> 3) & 0x1f
+										tmp5bitG1 = (g1 >> 3) & 0x1f
+										tmp5bitB1 = (b1 >> 3) & 0x1f
+										#print "5bit:", (tmp5bitR1, tmp5bitG1, tmp5bitB1)
+										tmpPixelColorConvertedToRGB555 = (tmp5bitR1 << 10) | (tmp5bitG1 << 5) | (tmp5bitB1)
+										#print "16bit:", tmpPixelColorConvertedToRGB555
+										tmpPixelColorRGB555ToWrite = pack('H', tmpPixelColorConvertedToRGB555)
+										targetFontFile.write(tmpPixelColorRGB555ToWrite) # unsigned short - 2 bytes
+									if (tmpColCur == c_endCol and explicitWidthIncrementVal > 0):
+										for tmpExtraColCur in range (0, explicitWidthIncrementVal):
+											#make completely transparent - write 0x8000
+											tmpPixelColorRGB555ToWrite = pack('H', 0x8000)
+											targetFontFile.write(tmpPixelColorRGB555ToWrite) # unsigned short - 2 bytes
+							kIncIndx = kIncIndx + 1 # finally increase the kIncIndx for next glyph
+
+						#
+						# End rest of the font characters image segments
+						#
+						targetFontFile.close()
+						#
+						# Re -open and write the image segment
+						#
+						targetFontFile = None
+						try:
+							targetFontFile = open(self.targetFONFilename, 'r+b')
+						except:
+							errorFound = True
+						if not errorFound:
+							targetFontFile.seek(0x0C)	  # position to write imageSegmentSize
+							tmpImageSegmentToWrite = pack('I', lastImageSegmentOffset)
+							targetFontFile.write(tmpImageSegmentToWrite)
+							targetFontFile.close()
+
+			else: ## if (self.lettersFound ) <= 0
+				errMsg = "No letters were found in input png!"
+				print errMsg
+				retVal = -2
+		return (retVal, errMsg, origGameFontSizeEqBaseLine, totalFontLetters, importedNumOfLetters)
+
+	def extractFonFilesFromMix(self):
+		""" Generate PNG files out of FON files stores in a MIX resource
+		"""
+		print "Checking in %s for MIX files to extract FON's from" % (self.inputFonMixPath)
+		inputMIXFilesFound = []
+		# breaking after first for loop yields only the top directory files, which is what we want
+		for (dirpath, dirnames, filenames) in walk(self.inputFonMixPath):
+			for filename in filenames:
+				for mixFileName in supportedMIXInputFiles:
+					if filename.upper() == mixFileName:
+						inputMIXFilesFound.append(mixFileName)
+			break
+		for tmpMIXfileName in inputMIXFilesFound:
+			print "Found MIX: %s" % ('"' + self.inputFonMixPath + tmpMIXfileName + '"')
+			errorFound = False
+			inMIXFile = None
+			#
+			try:
+				inMIXFile = open(os.path.join(self.inputFonMixPath,tmpMIXfileName), 'rb')
+			except:
+				errorFound = True
+				print "Unexpected error:", sys.exc_info()[0]
+				raise
+			if not errorFound:
+				totalFONs = 0
+				tmpBuff = inMIXFile.read(2)
+				# H: unsigned short (2 bytes) followed by I: unsigned int (4 bytes)
+				mixFileEntriesNumTuple = struct.unpack('H', tmpBuff)
+				numOfEntriesToExtract = mixFileEntriesNumTuple[0]
+				tmpBuff = inMIXFile.read(4)
+				mixFileDataSegmentSizeTuple = struct.unpack('I', tmpBuff)
+				allMixFileSize = mixFileDataSegmentSizeTuple[0]
+				inMIXFile.seek(0, 2) # go to file end
+				allActualBytesInMixFile = inMIXFile.tell()
+				inMIXFile.seek(6, 0) # go to start of table of MIX file entries (right after the 6 bytes header)
+				# 2 + 4 = 6 bytes short MIX header
+				# 12 bytes per MIX entry in entries table
+				# quick size validation
+				print "Entries: %d, data segment %d bytes" % (numOfEntriesToExtract, allMixFileSize)
+				if allActualBytesInMixFile != 2 + 4 + 12 * numOfEntriesToExtract + allMixFileSize:
+					print "Error: MIX file size mismatch with reported size in header for %s!" % (tmpMIXfileName)
+				else:
+					#
+					# 12 bytes per entry
+					#		4 bytes: ID
+					#		4 bytes: Offset in data segment
+					#		4 bytes: Size of data
+					#
+					for i in range(0, numOfEntriesToExtract):
+						foundFONFile = False
+						currFonFileName = 'UNKNOWN.FON'
+						inMIXFile.seek(2 + 4 + 12*i)
+						tmpBuff = inMIXFile.read(4)
+						tmpRdTuple = struct.unpack('I', tmpBuff)
+						idOfMIXEntry = tmpRdTuple[0]
+						tmpBuff = inMIXFile.read(4)
+						tmpRdTuple = struct.unpack('I', tmpBuff)
+						offsetOfMIXEntry = tmpRdTuple[0]
+						tmpBuff = inMIXFile.read(4)
+						tmpRdTuple = struct.unpack('I', tmpBuff)
+						sizeOfMIXEntry = tmpRdTuple[0]
+
+						for suppFONFileName in supportedExportedFONFiles:
+							if(idOfMIXEntry == calculateFoldHash(suppFONFileName)):
+								foundFONFile = True
+								currFonFileName = suppFONFileName
+								break
+
+						if (foundFONFile == True):
+							print "Entry Name: %s, Entry ID: %s, offset %s, data segment %s bytes" % (currFonFileName, ''.join('{:08X}'.format(idOfMIXEntry)), ''.join('{:08X}'.format(offsetOfMIXEntry)),''.join('{:08X}'.format(sizeOfMIXEntry)))
+							#
+							# IF FON FILE:
+							# put file in FON object
+							#
+							#
+							inMIXFile.seek(2 + 4 + 12*numOfEntriesToExtract + offsetOfMIXEntry)
+							if(offsetOfMIXEntry + sizeOfMIXEntry > allMixFileSize):
+								print "Error: FON file size mismatch with reported size in entry header!"
+							else:
+								fonFileBuffer = inMIXFile.read(sizeOfMIXEntry)
+								if (len(fonFileBuffer) == sizeOfMIXEntry):
+								# load FON file
+									thisFonFile = fonFile()
+									if (thisFonFile.loadFonFile(fonFileBuffer, allMixFileSize, currFonFileName)):
+										print "FON file loaded"
+										thisFonFile.outputFonToPNG()
+										totalFONs = totalFONs + 1
+									else:
+										print "Error while LOADING FON file!"
+								else:
+									print "Error while reading FON file %s into mem buffer" % (''.join('{:08X}'.format(idOfMIXEntry)))
+
+				inMIXFile.close()
+				print "Total FONs: %d " % (totalFONs)
+		return
+
+#
+#
+# ########################
+# main
+#
+# #########################
+#
+if __name__ == '__main__':
+#	main()
+	invalidSyntax = False
+	extractFonMode = False
+
+	TMPSpecialGlyphMode = True
+	TMPAutoTabCalculation = True
+
+	TMPOverrideEncodingFilePath = ""
+	TMPinputPathForMixFiles = ""
+	TMPimageRowFilePNG = ""
+	TMPTargetFONfilename = ""
+	TMPminSpaceBetweenLettersInRowLeftToLeft = 0
+	TMPminSpaceBetweenLettersInColumnTopToTop = 0
+	TMPkerningForFirstDummyFontLetter = 0
+	TMPYOffsToApplyToAllGlyphsExceptFirstSpecialGamma = 0
+	TMPSpaceWidthInPixels = 10
+#	TMPdeductKerningPixels = 0
+	TMPcustomBaseLineOffset = 0
+
+#	 print "Len of sysargv = %s" % (len(sys.argv))
+	if len(sys.argv) == 2:
+		if(sys.argv[1] == '--help'or sys.argv[1] == '-h'):
+			print "%s %s supports Blade Runner (English version, CD edition)." % (app_name_spaced, app_version)
+			print "Created by Praetorian of the classic adventures in Greek team."
+			print "Always keep backups!"
+			print "--------------------"
+			print "Preparatory steps:"
+			print "1. Put overrideEncoding.txt file in the same folder with this tool. (Recommended, but not obligatory step)"
+			print "--------------------"
+			print "Valid syntax A - export game fonts:"
+			print "%s -ip [folderpath_for_MIX_Files]\n" % (app_name)
+			print "Valid syntax B - create subtitle font:"
+			print "%s -im [image_Row_PNG_Filename] -om [output_FON_filename] -pxLL [minSpaceBetweenLettersInRowLeftToLeft] -pxTT [minSpaceBetweenLettersInColumnTopToTop] -pxKn [kerningForFirstDummyFontLetter] -pxWS [whiteSpaceWidthInPixels]\n" % (app_name)    # deductKerningPixels"
+			print "The -ip switch has an argument that is the path for the input (MIX) files folder (can be the same as the Blade Runner installation folder)."
+			print "The -oe switch has an argument that is the input overrideEncoding file to use for the particular font creation."
+			print "The -im switch has an argument that is the input PNG image with a row of the font glyphs spaced apart."
+			print "The -om switch has an argument that is the output FON filename."
+			print "The -pxLL switch has an integer argument that specifies the minimum number of pixels between the left side of a glyph and the left side of the next glyph to its right in the line-row PNG."
+			print "The -pxTT switch has an integer argument that specifies the minimum number of pixels between the top side of a glyph and the top side of the glyph below (if there's a second row) in the row PNG. If there is only one row, this argument still should be set (as if there was another row) to define where the parser should stop checking for the limits of a glyph vertically."
+			print "The -pxKn switch has an integer argument that sets kerning for the first dummy font glyph."
+			print "The -pxYo switch has an integer argument that sets an offset to be added to all detected y offsets for the glyphs (except the special first one)."
+			print "The -pxWS switch has an integer argument that sets the white space width in pixels for this particular font."
+			print "The --noSpecialGlyphs switch removes consideration for special glyphs that exist out of their proper ascii order."
+			print "The --noAutoTabCalculation switch removes the detection of tab spacing between letters (use this switch if you didn't create the PNG row file using a tab spaced list of glyphs)."
+			print "--------------------"
+			print "Thank you for using this app."
+			print "Please provide any feedback to: %s " % (company_email)
+			sys.exit()
+		elif(sys.argv[1] == '--version' or sys.argv[1] == '-v'):
+			print "%s %s supports Blade Runner (English version, CD edition)." % (app_name_spaced, app_version)
+			print "Please provide any feedback to: %s " % (company_email)
+			sys.exit()
+		else:
+			invalidSyntax = True
+	elif len(sys.argv) > 2:
+		for i in range(1, len(sys.argv)):
+			if( i < (len(sys.argv) - 1) and sys.argv[i][:1] == '-' and sys.argv[i+1][:1] != '-'):
+				if (sys.argv[i] == '-ip'):
+					TMPinputPathForMixFiles = sys.argv[i+1]
+					extractFonMode = True
+					print "Original FON file extraction mode enabled."
+				elif (sys.argv[i] == '-oe'):
+					TMPOverrideEncodingFilePath = sys.argv[i+1]
+				elif (sys.argv[i] == '-im'):
+					TMPimageRowFilePNG = sys.argv[i+1]
+				elif (sys.argv[i] == '-om'):
+					TMPTargetFONfilename = sys.argv[i+1]
+				elif (sys.argv[i] == '-pxLL'):
+					TMPminSpaceBetweenLettersInRowLeftToLeft = int(sys.argv[i+1])
+				elif (sys.argv[i] == '-pxTT'):
+					TMPminSpaceBetweenLettersInColumnTopToTop = int(sys.argv[i+1])
+				elif (sys.argv[i] == '-pxKn'):
+					TMPkerningForFirstDummyFontLetter = int(sys.argv[i+1])
+				elif (sys.argv[i] == '-pxYo'):
+					TMPYOffsToApplyToAllGlyphsExceptFirstSpecialGamma = int(sys.argv[i+1])
+				elif (sys.argv[i] == '-pxWS'):
+					TMPSpaceWidthInPixels = int(sys.argv[i+1])
+			elif sys.argv[i] == '--noSpecialGlyphs':
+				print "No special out-of-order glyphs mode enabled."
+				TMPSpecialGlyphMode = False
+			elif sys.argv[i] == '--noAutoTabCalculation':
+				print "No automatic tab calculation between glyphs."
+				TMPAutoTabCalculation = False
+
+		if (extractFonMode == False) and (not TMPTargetFONfilename or not TMPimageRowFilePNG or TMPminSpaceBetweenLettersInRowLeftToLeft <= 0 or TMPminSpaceBetweenLettersInColumnTopToTop <= 0 or TMPkerningForFirstDummyFontLetter <= 0 or TMPSpaceWidthInPixels <= 0)  : # this argument is mandatory
+			invalidSyntax = True
+
+		if (extractFonMode == True) and ( (TMPinputPathForMixFiles == '')  or  not TMPOverrideEncodingFilePath ):
+			invalidSyntax = True
+	else:
+		invalidSyntax = True
+
+
+	if (invalidSyntax == False):
+		#myGrabInstance = grabberFromPNG('windows-1253') #, grabberFromPNG.BR_GameID)
+		myGrabInstance = grabberFromPNG()
+		myGrabInstance.setInputPathForFonMix(TMPinputPathForMixFiles)
+		myGrabInstance.setImageRowFilePNG(TMPimageRowFilePNG)
+		myGrabInstance.setTargetFONFilename(TMPTargetFONfilename)
+		myGrabInstance.setMinSpaceBetweenLettersInRowLeftToLeft(TMPminSpaceBetweenLettersInRowLeftToLeft)
+		myGrabInstance.setMinSpaceBetweenLettersInColumnTopToTop(TMPminSpaceBetweenLettersInColumnTopToTop)
+		myGrabInstance.setKerningForFirstDummyFontLetter(TMPkerningForFirstDummyFontLetter)
+		myGrabInstance.setYOffsetForAllGlyphsExceptFirstSpecialGamma(TMPYOffsToApplyToAllGlyphsExceptFirstSpecialGamma)
+		myGrabInstance.setSpaceWidthInPixels(TMPSpaceWidthInPixels)
+		myGrabInstance.setSpecialGlyphMode(TMPSpecialGlyphMode)
+		myGrabInstance.setAutoTabCalculation(TMPAutoTabCalculation)
+		myGrabInstance.setOverrideEncodingPath(TMPOverrideEncodingFilePath)
+		myGrabInstance.initOverrideEncoding()
+#		myGrabInstance.setDeductKerningPixels(TMPdeductKerningPixels)
+		if extractFonMode:
+			myGrabInstance.extractFonFilesFromMix()
+		else:
+			myGrabInstance.generateModFiles(TMPcustomBaseLineOffset)
+	else:
+		invalidSyntax = True
+
+	if invalidSyntax == True:
+		print "Invalid syntax\n Try: \n %s -op [folderpath_for_extracted_wav_Files] \n %s --help for more info \n %s --version for version info " % (app_name, app_name, app_name)
+		tmpi = 0
+		for tmpArg in sys.argv:
+			if tmpi==0: #skip first argument
+				tmpi+=1
+				continue
+			print "\nArgument: %s" % (tmpArg)
+			tmpi+=1
+else:
+	#debug
+	#print 'font grabber imported from another module'
+	pass