[pywin32-checkins] pywin32/Pythonwin/pywin/scintilla formatter.py, 1.17, 1.18

[Mark H. <mha...@us...>]

Update of /cvsroot/pywin32/pywin32/Pythonwin/pywin/scintilla
In directory ddv4jf1.ch3.sourceforge.com:/tmp/cvs-serv22805/pywin/scintilla

Modified Files:
	formatter.py 
Log Message:
do syntax coloring on the raw utf8 stream

Index: formatter.py
===================================================================
RCS file: /cvsroot/pywin32/pywin32/Pythonwin/pywin/scintilla/formatter.py,v
retrieving revision 1.17
retrieving revision 1.18
diff -C2 -d -r1.17 -r1.18
*** formatter.py	3 Jan 2009 05:52:28 -0000	1.17
--- formatter.py	31 Jan 2009 03:52:39 -0000	1.18
***************
*** 240,244 ****
  	def Colorize(self, start=0, end=-1):
  		scintilla = self.scintilla
! 		stringVal = scintilla.GetTextRange(start, end)
  		if start > 0:
  			stylenum = scintilla.SCIGetStyleAt(start - 1)
--- 240,248 ----
  	def Colorize(self, start=0, end=-1):
  		scintilla = self.scintilla
! 		# scintilla's formatting is all done in terms of utf, so
! 		# we work with utf8 bytes instead of unicode.  This magically
! 		# works as any extended chars found in the utf8 don't change
! 		# the semantics.
! 		stringVal = scintilla.GetTextRange(start, end, decode=False)
  		if start > 0:
  			stylenum = scintilla.SCIGetStyleAt(start - 1)
***************
*** 359,363 ****
  
  	def ClassifyWord(self, cdoc, start, end, prevWord):
! 		word = cdoc[start:end+1]
  		attr = STYLE_IDENTIFIER
  		if prevWord == "class":
--- 363,367 ----
  
  	def ClassifyWord(self, cdoc, start, end, prevWord):
! 		word = cdoc[start:end+1].decode('latin-1')
  		attr = STYLE_IDENTIFIER
  		if prevWord == "class":
***************
*** 365,369 ****
  		elif prevWord == "def":
  			attr = STYLE_METHOD
! 		elif cdoc[start] in string.digits:
  			attr = STYLE_NUMBER
  		elif iskeyword(word):
--- 369,373 ----
  		elif prevWord == "def":
  			attr = STYLE_METHOD
! 		elif word[0] in string.digits:
  			attr = STYLE_NUMBER
  		elif iskeyword(word):
***************
*** 383,395 ****
  		state = styleStart
  		chPrev = chPrev2 = chPrev3 = ' '
! 		chNext = cdoc[charStart]
! 		chNext2 = cdoc[charStart]
  		startSeg = i = charStart
  		while i < lengthDoc:
  			ch = chNext
  			chNext = ' '
! 			if i+1 < lengthDoc: chNext = cdoc[i+1]
  			chNext2 = ' '
! 			if i+2 < lengthDoc: chNext2 = cdoc[i+2]
  			if state == STYLE_DEFAULT:
  				if ch in wordstarts:
--- 387,398 ----
  		state = styleStart
  		chPrev = chPrev2 = chPrev3 = ' '
! 		chNext2 = chNext = cdoc[charStart:charStart+1].decode('latin-1')
  		startSeg = i = charStart
  		while i < lengthDoc:
  			ch = chNext
  			chNext = ' '
! 			if i+1 < lengthDoc: chNext = cdoc[i+1:i+2].decode('latin-1')
  			chNext2 = ' '
! 			if i+2 < lengthDoc: chNext2 = cdoc[i+2:i+3].decode('latin-1')
  			if state == STYLE_DEFAULT:
  				if ch in wordstarts: