def getobj(self, objid):
assert objid != 0
if not self.xrefs:
raise PDFException('PDFDocument is not initialized')
if 2 <= self.debug:
print >> sys.stderr, 'getobj: objid=%r' % (objid)
if objid in self._cached_objs:
(obj, genno) = self._cached_objs[objid]
else:
for xref in self.xrefs:
try:
(strmid, index, genno) = xref.get_pos(objid)
except KeyError:
continue
try:
if strmid is not None:
stream = stream_value(self.getobj(strmid))
obj = self._getobj_objstm(stream, index, objid)
else:
obj = self._getobj_parse(index, objid)
if isinstance(obj, PDFStream):
obj.set_objid(objid, genno)
break
except (PSEOF, PDFSyntaxError):
continue
else:
raise PDFObjectNotFound(objid)
if 2 <= self.debug:
print >> sys.stderr, 'register: objid=%r: %r' % (objid, obj)
if self.caching:
self._cached_objs[objid] = (obj, genno)
if self.decipher:
obj = decipher_all(self.decipher, objid, genno, obj)
return obj
def get_pages(self):
if not self.xrefs:
raise PDFException('PDFDocument is not initialized')
def search(obj, parent):
if isinstance(obj, int):
objid = obj
tree = dict_value(self.getobj(objid)).copy()
else:
objid = obj.objid
tree = dict_value(obj).copy()
for (k,v) in parent.iteritems():
if k in self.INHERITABLE_ATTRS and k not in tree:
tree[k] = v
if tree.get('Type') is LITERAL_PAGES and 'Kids' in tree:
if 1 <= self.debug:
print >>sys.stderr, 'Pages: Kids=%r' % tree['Kids']
for c in list_value(tree['Kids']):
for x in search(c, tree):
yield x
elif tree.get('Type') is LITERAL_PAGE:
if 1 <= self.debug:
print >>sys.stderr, 'Page: %r' % tree
yield (objid, tree)
if 'Pages' not in self.catalog: return
for (pageid,tree) in search(self.catalog['Pages'], self.catalog):
yield PDFPage(self, pageid, tree)
return
def get_outlines(self):
if 'Outlines' not in self.catalog:
raise PDFException('No /Outlines defined!')
def search(entry, level):
entry = dict_value(entry)
if 'Title' in entry:
if 'A' in entry or 'Dest' in entry:
title = decode_text(str_value(entry['Title']))
dest = entry.get('Dest')
action = entry.get('A')
se = entry.get('SE')
yield (level, title, dest, action, se)
if 'First' in entry and 'Last' in entry:
for x in search(entry['First'], level + 1):
yield x
if 'Next' in entry:
for x in search(entry['Next'], level):
yield x
return
return search(self.catalog['Outlines'], 0)
def getobj(self, objid):
if not self.xrefs:
raise PDFException('PDFDocument is not initialized')
if 2 <= self.debug:
print >>sys.stderr, 'getobj: objid=%r' % (objid)
if objid in self._cached_objs:
genno = 0
obj = self._cached_objs[objid]
else:
for xref in self.xrefs:
try:
(strmid, index) = xref.get_pos(objid)
break
except KeyError:
pass
else:
if STRICT:
raise PDFSyntaxError('Cannot locate objid=%r' % objid)
# return null for a nonexistent reference.
return None
if strmid:
stream = stream_value(self.getobj(strmid))
if stream.get('Type') is not LITERAL_OBJSTM:
if STRICT:
raise PDFSyntaxError('Not a stream object: %r' % stream)
try:
n = stream['N']
except KeyError:
if STRICT:
raise PDFSyntaxError('N is not defined: %r' % stream)
n = 0
if strmid in self._parsed_objs:
objs = self._parsed_objs[strmid]
else:
parser = PDFStreamParser(stream.get_data())
parser.set_document(self)
objs = []
try:
while 1:
(_,obj) = parser.nextobject()
objs.append(obj)
except PSEOF:
pass
if self.caching:
self._parsed_objs[strmid] = objs
genno = 0
i = n*2+index
try:
obj = objs[i]
except IndexError:
if STRICT:
raise PDFSyntaxError('Invalid object number: objid=%r' % (objid))
# return None for an invalid object number
return None
if isinstance(obj, PDFStream):
obj.set_objid(objid, 0)
else:
self._parser.seek(index)
(_,objid1) = self._parser.nexttoken() # objid
(_,genno) = self._parser.nexttoken() # genno
(_,kwd) = self._parser.nexttoken()
# #### hack around malformed pdf files
#assert objid1 == objid, (objid, objid1)
if objid1 != objid:
x = []
while kwd is not self.KEYWORD_OBJ:
(_,kwd) = self._parser.nexttoken()
x.append(kwd)
if x:
objid1 = x[-2]
genno = x[-1]
# #### end hack around malformed pdf files
if kwd is not self.KEYWORD_OBJ:
raise PDFSyntaxError('Invalid object spec: offset=%r' % index)
try:
(_,obj) = self._parser.nextobject()
if isinstance(obj, PDFStream):
obj.set_objid(objid, genno)
except PSEOF:
return None
if 2 <= self.debug:
print >>sys.stderr, 'register: objid=%r: %r' % (objid, obj)
if self.caching:
self._cached_objs[objid] = obj
if self.decipher:
obj = decipher_all(self.decipher, objid, genno, obj)
return obj
