def load(self, parser, debug=0):
(_, objid) = parser.nexttoken() # ignored
(_, genno) = parser.nexttoken() # ignored
(_, kwd) = parser.nexttoken()
(_, stream) = parser.nextobject()
if not isinstance(stream,
PDFStream) or stream.dic['Type'] is not LITERAL_XREF:
raise PDFNoValidXRef('Invalid PDF stream spec.')
size = stream.dic['Size']
index_array = stream.dic.get('Index', (0, size))
if len(index_array) % 2 != 0:
raise PDFSyntaxError('Invalid index number')
self.objid_ranges = [
XRefObjRange(start, nobjs)
for (start, nobjs) in choplist(2, index_array)
]
(self.fl1, self.fl2, self.fl3) = stream.dic['W']
self.data = stream.get_data()
self.entlen = self.fl1 + self.fl2 + self.fl3
self.trailer = stream.dic
if debug:
print >> stderr, ('xref stream: objid=%s, fields=%d,%d,%d' %
(', '.join(map(repr, self.objid_ranges),
self.fl1, self.fl2, self.fl3)))
return
def lookup(d):
if 'Limits' in d:
(k1,k2) = list_value(d['Limits'])
if key < k1 or k2 < key: return None
if 'Names' in d:
objs = list_value(d['Names'])
names = dict(choplist(2, objs))
return names[key]
if 'Kids' in d:
for c in list_value(d['Kids']):
v = lookup(dict_value(c))
if v: return v
raise KeyError((cat,key))
def lookup(d):
if 'Limits' in d:
(k1, k2) = list_value(d['Limits'])
if key < k1 or k2 < key: return None
if 'Names' in d:
objs = list_value(d['Names'])
names = dict(choplist(2, objs))
return names[key]
if 'Kids' in d:
for c in list_value(d['Kids']):
v = lookup(dict_value(c))
if v: return v
raise KeyError((cat, key))
def nextobject(self):
'''
Yields a list of objects: keywords, literals, strings,
numbers, arrays and dictionaries. Arrays and dictionaries
are represented as Python sequence and dictionaries.
'''
while not self.results:
(pos, token) = self.nexttoken()
#print (pos,token), (self.curtype, self.curstack)
if (isinstance(token, int) or isinstance(token, float)
or isinstance(token, bool) or isinstance(token, str)
or isinstance(token, PSLiteral)):
# normal token
self.push((pos, token))
elif token == KEYWORD_ARRAY_BEGIN:
# begin array
self.start_type(pos, 'a')
elif token == KEYWORD_ARRAY_END:
# end array
try:
self.push(self.end_type('a'))
except PSTypeError:
if STRICT: raise
elif token == KEYWORD_DICT_BEGIN:
# begin dictionary
self.start_type(pos, 'd')
elif token == KEYWORD_DICT_END:
# end dictionary
try:
(pos, objs) = self.end_type('d')
if len(objs) % 2 != 0:
raise PSSyntaxError(
'Invalid dictionary construct: %r' % objs)
d = dict(
(literal_name(k), v) for (k, v) in choplist(2, objs))
self.push((pos, d))
except PSTypeError:
if STRICT: raise
else:
if 2 <= self.debug:
print >>stderr, 'do_keyword: pos=%r, token=%r, stack=%r' % \
(pos, token, self.curstack)
self.do_keyword(pos, token)
if self.context:
continue
else:
self.flush()
obj = self.results.pop(0)
if 2 <= self.debug:
print >> stderr, 'nextobject: %r' % (obj, )
return obj
def do_keyword(self, pos, token):
if token is self.KEYWORD_BI:
# inline image within a content stream
self.start_type(pos, 'inline')
elif token is self.KEYWORD_ID:
try:
(_, objs) = self.end_type('inline')
if len(objs) % 2 != 0:
raise PSTypeError('Invalid dictionary construct: %r' % objs)
d = dict( (literal_name(k), v) for (k,v) in choplist(2, objs) )
(pos, data) = self.get_inline_data(pos+len('ID '))
obj = PDFStream(d, data)
self.push((pos, obj))
self.push((pos, self.KEYWORD_EI))
except PSTypeError:
if STRICT: raise
else:
self.push((pos, token))
return
def load(self, parser, debug=0):
(_,objid) = parser.nexttoken() # ignored
(_,genno) = parser.nexttoken() # ignored
(_,kwd) = parser.nexttoken()
(_,stream) = parser.nextobject()
if not isinstance(stream, PDFStream) or stream.dic['Type'] is not LITERAL_XREF:
raise PDFNoValidXRef('Invalid PDF stream spec.')
size = stream.dic['Size']
index_array = stream.dic.get('Index', (0,size))
if len(index_array) % 2 != 0:
raise PDFSyntaxError('Invalid index number')
self.objid_ranges = [ XRefObjRange(start,nobjs) for (start,nobjs) in choplist(2, index_array) ]
(self.fl1, self.fl2, self.fl3) = stream.dic['W']
self.data = stream.get_data()
self.entlen = self.fl1+self.fl2+self.fl3
self.trailer = stream.dic
if debug:
print >>stderr, ('xref stream: objid=%s, fields=%d,%d,%d' %
(', '.join(map(repr, self.objid_ranges), self.fl1, self.fl2, self.fl3)))
return
def do_keyword(self, pos, token):
name = token.name
if name == 'begincmap':
self.in_cmap = True
self.popall()
return
elif name == 'endcmap':
self.in_cmap = False
return
if not self.in_cmap: return
#
if name == 'def':
try:
((_,k),(_,v)) = self.pop(2)
self.cmap.attrs[literal_name(k)] = v
except PSSyntaxError:
pass
return
if name == 'usecmap':
try:
((_,cmapname),) = self.pop(1)
self.cmap.copycmap(CMapDB.get_cmap(literal_name(cmapname)))
except PSSyntaxError:
pass
return
if name == 'begincodespacerange':
self.popall()
return
if name == 'endcodespacerange':
self.popall()
return
if name == 'begincidrange':
self.popall()
return
if name == 'endcidrange':
objs = [ obj for (_,obj) in self.popall() ]
for (s,e,cid) in choplist(3, objs):
if (not isinstance(s, str) or not isinstance(e, str) or
not isinstance(cid, int) or len(s) != len(e)): continue
sprefix = s[:-4]
eprefix = e[:-4]
if sprefix != eprefix: continue
svar = s[-4:]
evar = e[-4:]
s1 = nunpack(svar)
e1 = nunpack(evar)
vlen = len(svar)
#assert s1 <= e1
for i in xrange(e1-s1+1):
x = sprefix+pack('>L',s1+i)[-vlen:]
self.cmap.register_code2cid(x, cid+i)
return
if name == 'begincidchar':
self.popall()
return
if name == 'endcidchar':
objs = [ obj for (_,obj) in self.popall() ]
for (cid,code) in choplist(2, objs):
if isinstance(code, str) and isinstance(cid, str):
self.cmap.register_code2cid(code, nunpack(cid))
return
if name == 'beginbfrange':
self.popall()
return
if name == 'endbfrange':
objs = [ obj for (_,obj) in self.popall() ]
for (s,e,code) in choplist(3, objs):
if (not isinstance(s, str) or not isinstance(e, str) or
len(s) != len(e)): continue
s1 = nunpack(s)
e1 = nunpack(e)
#assert s1 <= e1
if isinstance(code, list):
for i in xrange(e1-s1+1):
self.cmap.register_cid2code(s1+i, code[i])
else:
var = code[-4:]
base = nunpack(var)
prefix = code[:-4]
vlen = len(var)
for i in xrange(e1-s1+1):
x = prefix+pack('>L',base+i)[-vlen:]
self.cmap.register_cid2code(s1+i, x)
return
if name == 'beginbfchar':
self.popall()
return
if name == 'endbfchar':
objs = [ obj for (_,obj) in self.popall() ]
for (cid,code) in choplist(2, objs):
if isinstance(cid, str) and isinstance(code, str):
self.cmap.register_cid2code(nunpack(cid), code)
return
if name == 'beginnotdefrange':
self.popall()
return
if name == 'endnotdefrange':
self.popall()
return
self.push((pos, token))
return
def do_keyword(self, pos, token):
name = token.name
if name == 'begincmap':
self.in_cmap = True
self.popall()
return
elif name == 'endcmap':
self.in_cmap = False
return
if not self.in_cmap: return
#
if name == 'def':
try:
((_, k), (_, v)) = self.pop(2)
self.cmap.attrs[literal_name(k)] = v
except PSSyntaxError:
pass
return
if name == 'usecmap':
try:
((_, cmapname), ) = self.pop(1)
self.cmap.copycmap(CMapDB.get_cmap(literal_name(cmapname)))
except PSSyntaxError:
pass
return
if name == 'begincodespacerange':
self.popall()
return
if name == 'endcodespacerange':
self.popall()
return
if name == 'begincidrange':
self.popall()
return
if name == 'endcidrange':
objs = [obj for (_, obj) in self.popall()]
for (s, e, cid) in choplist(3, objs):
if (not isinstance(s, str) or not isinstance(e, str)
or not isinstance(cid, int) or len(s) != len(e)):
continue
sprefix = s[:-4]
eprefix = e[:-4]
if sprefix != eprefix: continue
svar = s[-4:]
evar = e[-4:]
s1 = nunpack(svar)
e1 = nunpack(evar)
vlen = len(svar)
#assert s1 <= e1
for i in xrange(e1 - s1 + 1):
x = sprefix + pack('>L', s1 + i)[-vlen:]
self.cmap.register_code2cid(x, cid + i)
return
if name == 'begincidchar':
self.popall()
return
if name == 'endcidchar':
objs = [obj for (_, obj) in self.popall()]
for (cid, code) in choplist(2, objs):
if isinstance(code, str) and isinstance(cid, str):
self.cmap.register_code2cid(code, nunpack(cid))
return
if name == 'beginbfrange':
self.popall()
return
if name == 'endbfrange':
objs = [obj for (_, obj) in self.popall()]
for (s, e, code) in choplist(3, objs):
if (not isinstance(s, str) or not isinstance(e, str)
or len(s) != len(e)):
continue
s1 = nunpack(s)
e1 = nunpack(e)
#assert s1 <= e1
if isinstance(code, list):
for i in xrange(e1 - s1 + 1):
self.cmap.register_cid2code(s1 + i, code[i])
else:
var = code[-4:]
base = nunpack(var)
prefix = code[:-4]
vlen = len(var)
for i in xrange(e1 - s1 + 1):
x = prefix + pack('>L', base + i)[-vlen:]
self.cmap.register_cid2code(s1 + i, x)
return
if name == 'beginbfchar':
self.popall()
return
if name == 'endbfchar':
objs = [obj for (_, obj) in self.popall()]
for (cid, code) in choplist(2, objs):
if isinstance(cid, str) and isinstance(code, str):
self.cmap.register_cid2code(nunpack(cid), code)
return
if name == 'beginnotdefrange':
self.popall()
return
if name == 'endnotdefrange':
self.popall()
return
self.push((pos, token))
return