zacc806
/
pdfminer.six
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

cmap bytes and decode

pull/1/head
cybjit 2014-09-07 18:41:04 +02:00
parent cc733c8217
commit a6f31a713d
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
pdfminer/cmapdb.py
View File

@ -180,11 +180,11 @@ class FileUnicodeMap(UnicodeMap):
if isinstance(code, PSLiteral): if isinstance(code, PSLiteral):
# Interpret as an Adobe glyph name. # Interpret as an Adobe glyph name.
self.cid2unichr[cid] = name2unicode(code.name) self.cid2unichr[cid] = name2unicode(code.name)
elif isinstance(code, str): elif isinstance(code, bytes):
# Interpret as UTF-16BE. # Interpret as UTF-16BE.
self.cid2unichr[cid] = unicode(code, 'UTF-16BE', 'ignore') self.cid2unichr[cid] = code.decode('UTF-16BE', 'ignore')
elif isinstance(code, int): elif isinstance(code, int):
self.cid2unichr[cid] = unichr(code) self.cid2unichr[cid] = six.unichr(code)
else: else:
raise TypeError(code) raise TypeError(code)
return return
@ -379,7 +379,7 @@ class CMapParser(PSStackParser):
if token is self.KEYWORD_ENDBFRANGE: if token is self.KEYWORD_ENDBFRANGE:
objs = [obj for (__, obj) in self.popall()] objs = [obj for (__, obj) in self.popall()]
for (s, e, code) in choplist(3, objs): for (s, e, code) in choplist(3, objs):
if (not isinstance(s, str) or not isinstance(e, str) or if (not isinstance(s, bytes) or not isinstance(e, bytes) or
len(s) != len(e)): len(s) != len(e)):
continue continue
s1 = nunpack(s) s1 = nunpack(s)
@ -404,7 +404,7 @@ class CMapParser(PSStackParser):
if token is self.KEYWORD_ENDBFCHAR: if token is self.KEYWORD_ENDBFCHAR:
objs = [obj for (__, obj) in self.popall()] objs = [obj for (__, obj) in self.popall()]
for (cid, code) in choplist(2, objs): for (cid, code) in choplist(2, objs):
if isinstance(cid, str) and isinstance(code, str): if isinstance(cid, bytes) and isinstance(code, bytes):
self.cmap.add_cid2unichr(nunpack(cid), code) self.cmap.add_cid2unichr(nunpack(cid), code)
return return