コード例 #1
0
def uncompress(mylist,
               leave_raw=False,
               warnings=set(),
               flate=PdfName.FlateDecode,
               decompress=decompressobj,
               isinstance=isinstance,
               list=list,
               len=len):
    ok = True
    for obj in streamobjects(mylist):
        ftype = obj.Filter
        if ftype is None:
            continue
        if isinstance(ftype, list) and len(ftype) == 1:
            # todo: multiple filters
            ftype = ftype[0]
        parms = obj.DecodeParms or obj.DP
        if ftype != flate:
            msg = ('Not decompressing: cannot use filter %s'
                   ' with parameters %s') % (repr(ftype), repr(parms))
            if msg not in warnings:
                warnings.add(msg)
                log.warning(msg)
            ok = False
        else:
            dco = decompress()
            try:
                data = dco.decompress(convert_store(obj.stream))
            except Exception as s:
                error = str(s)
            else:
                error = None
                if isinstance(parms, PdfArray):
                    oldparms = parms
                    parms = PdfDict()
                    for x in oldparms:
                        parms.update(x)
                if parms:
                    predictor = int(parms.Predictor or 1)
                    columns = int(parms.Columns or 1)
                    colors = int(parms.Colors or 1)
                    bpc = int(parms.BitsPerComponent or 8)
                    if 10 <= predictor <= 15:
                        data, error = flate_png(data, predictor, columns,
                                                colors, bpc)
                    elif predictor != 1:
                        error = ('Unsupported flatedecode predictor %s' %
                                 repr(predictor))
            if error is None:
                assert not dco.unconsumed_tail
                if dco.unused_data.strip():
                    error = ('Unconsumed compression data: %s' %
                             repr(dco.unused_data[:20]))
            if error is None:
                obj.Filter = None
                obj.stream = data if leave_raw else convert_load(data)
            else:
                log.error('%s %s' % (error, repr(obj.indirect)))
                ok = False
    return ok
コード例 #2
0
    def readpages(self, node):
        pagename = PdfName.Page
        pagesname = PdfName.Pages
        catalogname = PdfName.Catalog
        typename = PdfName.Type
        kidname = PdfName.Kids

        # PDFs can have arbitrarily nested Pages/Page
        # dictionary structures.
        def readnode(node):
            nodetype = node[typename]
            if nodetype == pagename:
                yield node
            elif nodetype == pagesname:
                if type(node[kidname]) == PdfArray:
                    for node in node[kidname]:
                        for node in readnode(node):
                            yield node
                # else it's one PdfDict
                else:
                    for node in readnode(node[kidname]):
                        yield node
            elif nodetype == catalogname:
                for node in readnode(node[pagesname]):
                    yield node
            else:
                log.error('Expected /Page or /Pages dictionary, got %s' %
                          repr(node))

        try:
            return list(readnode(node))
        except (AttributeError, TypeError), s:
            log.error('Invalid page tree: %s' % s)
            return []
コード例 #3
0
def uncompress(mylist, warnings=set(), flate = PdfName.FlateDecode,
                    decompress=zlib.decompressobj, isinstance=isinstance, list=list, len=len):
    ok = True
    for obj in streamobjects(mylist):
        ftype = obj.Filter
        if ftype is None:
            continue
        if isinstance(ftype, list) and len(ftype) == 1:
            # todo: multiple filters
            ftype = ftype[0]
        parms = obj.DecodeParms
        if ftype != flate or parms is not None:
            msg = 'Not decompressing: cannot use filter %s with parameters %s' % (repr(ftype), repr(parms))
            if msg not in warnings:
                warnings.add(msg)
                log.warning(msg)
            ok = False
        else:
            dco = decompress()
            error = None
            try:
                data = dco.decompress(obj.stream)
            except Exception, s:
                error = str(s)
            if error is None:
                assert not dco.unconsumed_tail
                if dco.unused_data.strip():
                    error = 'Unconsumed compression data: %s' % repr(dco.unused_data[:20])
            if error is None:
                obj.Filter = None
                obj.stream = data
            else:
                log.error('%s %s' % (error, repr(obj.indirect)))
コード例 #4
0
ファイル: pdfreader.py プロジェクト: deepsign/novela-workflow
    def readpages(self, node):
        pagename=PdfName.Page
        pagesname=PdfName.Pages
        catalogname = PdfName.Catalog
        typename = PdfName.Type
        kidname = PdfName.Kids

        # PDFs can have arbitrarily nested Pages/Page
        # dictionary structures.
        def readnode(node):
            nodetype = node[typename]
            if nodetype == pagename:
                yield node
            elif nodetype == pagesname:
                for node in node[kidname]:
                    for node in readnode(node):
                        yield node
            elif nodetype == catalogname:
                for node in readnode(node[pagesname]):
                    yield node
            else:
                log.error('Expected /Page or /Pages dictionary, got %s' % repr(node))
        try:
            return list(readnode(node))
        except (AttributeError, TypeError), s:
            log.error('Invalid page tree: %s' % s)
            return []
コード例 #5
0
ファイル: pdfreader.py プロジェクト: chubbymaggie/pdfrw
 def parsexref(self, source, int=int, range=range):
     ''' Parse (one of) the cross-reference file section(s)
     '''
     fdata = source.fdata
     setdefault = source.obj_offsets.setdefault
     add_offset = source.all_offsets.append
     next = source.next
     tok = next()
     if tok != 'xref':
         source.exception('Expected "xref" keyword')
     start = source.floc
     try:
         while 1:
             tok = next()
             if tok == 'trailer':
                 return
             startobj = int(tok)
             for objnum in range(startobj, startobj + int(next())):
                 offset = int(next())
                 generation = int(next())
                 inuse = next()
                 if inuse == 'n':
                     if offset != 0:
                         setdefault((objnum, generation), offset)
                         add_offset(offset)
                 elif inuse != 'f':
                     raise ValueError
     except:
         pass
     try:
         # Table formatted incorrectly.  See if
         # we can figure it out anyway.
         end = source.fdata.rindex('trailer', start)
         table = source.fdata[start:end].splitlines()
         for line in table:
             tokens = line.split()
             if len(tokens) == 2:
                 objnum = int(tokens[0])
             elif len(tokens) == 3:
                 offset, generation, inuse = (int(tokens[0]),
                                              int(tokens[1]), tokens[2])
                 if offset != 0 and inuse == 'n':
                     setdefault((objnum, generation), offset)
                     add_offset(offset)
                 objnum += 1
             elif tokens:
                 log.error('Invalid line in xref table: %s' % repr(line))
                 raise ValueError
         log.warning('Badly formatted xref table')
         source.floc = end
         source.next()
     except:
         source.floc = start
         source.exception('Invalid table format')
コード例 #6
0
 def old_parsexref(self, source, int=int, range=range):
     ''' Parse (one of) the cross-reference file section(s)
     '''
     fdata = source.fdata
     setdefault = source.obj_offsets.setdefault
     add_offset = source.all_offsets.append
     next = source.next
     tok = next()
     if tok != 'xref':
         source.exception('Expected "xref" keyword')
     start = source.floc
     try:
         while 1:
             tok = next()
             if tok == 'trailer':
                 return
             startobj = int(tok)
             for objnum in range(startobj, startobj + int(next())):
                 offset = int(next())
                 generation = int(next())
                 inuse = next()
                 if inuse == 'n':
                     if offset != 0:
                         setdefault((objnum, generation), offset)
                         add_offset(offset)
                 elif inuse != 'f':
                     raise ValueError
     except:
         pass
     try:
         # Table formatted incorrectly.  See if
         # we can figure it out anyway.
         end = source.fdata.rindex('trailer', start)
         table = source.fdata[start:end].splitlines()
         for line in table:
             tokens = line.split()
             if len(tokens) == 2:
                 objnum = int(tokens[0])
             elif len(tokens) == 3:
                 offset, generation, inuse = (int(tokens[0]),
                                              int(tokens[1]), tokens[2])
                 if offset != 0 and inuse == 'n':
                     setdefault((objnum, generation), offset)
                     add_offset(offset)
                 objnum += 1
             elif tokens:
                 log.error('Invalid line in xref table: %s' % repr(line))
                 raise ValueError
         log.warning('Badly formatted xref table')
         source.floc = end
         source.next()
     except:
         source.floc = start
         source.exception('Invalid table format')
コード例 #7
0
ファイル: pdfreader.py プロジェクト: deepsign/novela-workflow
 def readnode(node):
     nodetype = node[typename]
     if nodetype == pagename:
         yield node
     elif nodetype == pagesname:
         for node in node[kidname]:
             for node in readnode(node):
                 yield node
     elif nodetype == catalogname:
         for node in readnode(node[pagesname]):
             yield node
     else:
         log.error('Expected /Page or /Pages dictionary, got %s' % repr(node))
コード例 #8
0
 def readnode(node):
     nodetype = node[typename]
     if nodetype == pagename:
         yield node
     elif nodetype == pagesname:
         for node in node[kidname]:
             for node in readnode(node):
                 yield node
     elif nodetype == catalogname:
         for node in readnode(node[pagesname]):
             yield node
     else:
         log.error('Expected /Page or /Pages dictionary, got %s' %
                   repr(node))
コード例 #9
0
def uncompress(mylist,
               warnings=set(),
               flate=PdfName.FlateDecode,
               decompress=zlib.decompressobj,
               isinstance=isinstance,
               list=list,
               len=len):
    ok = True
    for obj in streamobjects(mylist):
        ftype = obj.Filter
        if ftype is None:
            continue
        if isinstance(ftype, list) and len(ftype) == 1:
            # todo: multiple filters
            ftype = ftype[0]
        parms = obj.DecodeParms
        if ftype != flate or parms is not None:
            msg = ('Not decompressing: cannot use filter %s '
                   'with parameters %s' % (repr(ftype), repr(parms)))
            if msg not in warnings:
                warnings.add(msg)
                log.warning(msg)
            ok = False
        else:
            dco = decompress()
            error = None
            try:
                data = dco.decompress(obj.stream)
            except Exception, s:
                error = str(s)
            if error is None:
                assert not dco.unconsumed_tail
                if dco.unused_data.strip():
                    error = ('Unconsumed compression data: %s' %
                             repr(dco.unused_data[:20]))
            if error is None:
                obj.Filter = None
                obj.stream = data
            else:
                log.error('%s %s' % (error, repr(obj.indirect)))
コード例 #10
0
ファイル: buildxobj.py プロジェクト: chubbymaggie/pdfrw
 def __init__(self, pageinfo='', **kw):
     pageinfo = pageinfo.split('#', 1)
     if len(pageinfo) == 2:
         pageinfo[1:] = pageinfo[1].replace('&', '#').split('#')
     for key in 'page viewrect'.split():
         if pageinfo[0].startswith(key + '='):
             break
     else:
         self.docname = pageinfo.pop(0)
     for item in pageinfo:
         key, value = item.split('=')
         key = key.strip()
         value = value.replace(',', ' ').split()
         if key in ('page', 'rotate'):
             assert len(value) == 1
             setattr(self, key, int(value[0]))
         elif key == 'viewrect':
             assert len(value) == 4
             setattr(self, key, [float(x) for x in value])
         else:
             log.error('Unknown option: %s', key)
     for key, value in kw.iteritems():
         assert hasattr(self, key), key
         setattr(self, key, value)
コード例 #11
0
 def __init__(self, pageinfo='', **kw):
     pageinfo = pageinfo.split('#', 1)
     if len(pageinfo) == 2:
         pageinfo[1:] = pageinfo[1].replace('&', '#').split('#')
     for key in 'page viewrect'.split():
         if pageinfo[0].startswith(key + '='):
             break
     else:
         self.docname = pageinfo.pop(0)
     for item in pageinfo:
         key, value = item.split('=')
         key = key.strip()
         value = value.replace(',', ' ').split()
         if key in ('page', 'rotate'):
             assert len(value) == 1
             setattr(self, key, int(value[0]))
         elif key == 'viewrect':
             assert len(value) == 4
             setattr(self, key, [float(x) for x in value])
         else:
             log.error('Unknown option: %s', key)
     for key, value in iteritems(kw):
         assert hasattr(self, key), key
         setattr(self, key, value)
コード例 #12
0
class PdfReader(PdfDict):

    warned_bad_stream_start = False  # Use to keep from spewing warnings
    warned_bad_stream_end = False  # Use to keep from spewing warnings

    def findindirect(self, objnum, gennum, PdfIndirect=PdfIndirect, int=int):
        ''' Return a previously loaded indirect object, or create
            a placeholder for it.
        '''
        key = int(objnum), int(gennum)
        result = self.indirect_objects.get(key)
        if result is None:
            self.indirect_objects[key] = result = PdfIndirect(key)
            self.deferred_objects.add(key)
            result._loader = self.loadindirect
        return result

    def readarray(self, source, PdfArray=PdfArray):
        ''' Found a [ token.  Parse the tokens after that.
        '''
        specialget = self.special.get
        result = []
        pop = result.pop
        append = result.append

        for value in source:
            if value in ']R':
                if value == ']':
                    break
                generation = pop()
                value = self.findindirect(pop(), generation)
            else:
                func = specialget(value)
                if func is not None:
                    value = func(source)
            append(value)
        return PdfArray(result)

    def readdict(self, source, PdfDict=PdfDict):
        ''' Found a << token.  Parse the tokens after that.
        '''
        specialget = self.special.get
        result = PdfDict()
        next = source.next

        tok = next()
        while tok != '>>':
            if not tok.startswith('/'):
                # Just skip the incorrect /name object.
                source.warning('Expected PDF /name object')
                tok = next()
                continue
            key = tok
            value = next()
            func = specialget(value)
            if func is not None:
                # Just keep working when bad token occurs.
                if func == self.badtoken:
                    tok = value
                    continue
                value = func(source)
                tok = next()
            else:
                tok = next()
                if value.isdigit() and tok.isdigit():
                    if next() != 'R':
                        source.exception('Expected "R" following two integers')
                    value = self.findindirect(value, tok)
                    tok = next()
            result[key] = value
        return result

    def empty_obj(self, source, PdfObject=PdfObject):
        ''' Some silly git put an empty object in the
            file.  Back up so the caller sees the endobj.
        '''
        source.floc = source.tokstart
        return PdfObject()

    def badtoken(self, source):
        ''' Didn't see that coming.
        '''
        source.exception('Unexpected delimiter')

    def findstream(self,
                   obj,
                   tok,
                   source,
                   PdfDict=PdfDict,
                   isinstance=isinstance,
                   len=len):
        ''' Figure out if there is a content stream
            following an object, and return the start
            pointer to the content stream if so.

            (We can't read it yet, because we might not
            know how long it is, because Length might
            be an indirect object.)
        '''

        isdict = isinstance(obj, PdfDict)
        if not isdict or tok != 'stream':
            source.exception("Expected 'endobj'%s token",
                             isdict and " or 'stream'" or '')
        fdata = source.fdata
        startstream = source.tokstart + len(tok)

        # Skip the possible delimiters.
        possible_delimiters = ('\r', '\n', ' ')
        gotcr = gotlf = False
        while fdata[startstream] in possible_delimiters:
            if fdata[startstream] == '\r':
                gotcr = True
            if fdata[startstream] == '\n':
                gotlf = True
            startstream += 1
        if not gotlf:
            if not gotcr:
                source.warning(r'stream keyword not followed by \n')
                self.private.warned_bad_stream_start = True
            if not self.warned_bad_stream_start:
                source.warning(r"stream keyword terminated by \r without \n")
                self.private.warned_bad_stream_start = True
        return startstream

    def readstream(self,
                   obj,
                   startstream,
                   source,
                   streamending='endstream endobj'.split(),
                   int=int):
        fdata = source.fdata

        # Get a length by looking 'endstream'
        end_loc = fdata.find('endstream', startstream)
        possible_delimiters = ('\r', '\n', ' ')
        while fdata[end_loc - 1] in possible_delimiters:
            end_loc -= 1
        observed_length = end_loc - startstream

        if obj.Length == None:
            length = observed_length
            source.warning(
                'Lacking the stream length declaration, using the observed value %d.'
                % (observed_length))
        else:
            try:
                length = int(obj.Length)
            except:
                source.warning(
                    'Incorrect representation of stream length: %s. Use observed value %d instead.'
                    % (obj.Length, observed_length))
                length = observed_length
            if length != observed_length:
                source.warning(
                    'Inconsistent stream length: %d declared, %d observed.' %
                    (length, observed_length))
                length = observed_length

        source.floc = target_endstream = startstream + length
        endit = source.multiple(2)
        obj._stream = fdata[startstream:target_endstream]
        if endit == streamending:
            return

        # The length attribute does not match the distance between the
        # stream and endstream keywords.

        do_warn, self.private.warned_bad_stream_end = (
            self.warned_bad_stream_end, False)

        # TODO:  Extract maxstream from dictionary of object offsets
        # and use rfind instead of find.
        maxstream = len(fdata) - 20
        endstream = fdata.find('endstream', startstream, maxstream)
        source.floc = startstream
        room = endstream - startstream
        if endstream < 0:
            source.error('Could not find endstream')
            return
        if (length == room + 1
                and fdata[startstream - 2:startstream] == '\r\n'):
            source.warning(r"stream keyword terminated by \r without \n")
            obj._stream = fdata[startstream - 1:target_endstream - 1]
            return
        source.floc = endstream
        if length > room:
            source.error(
                'stream /Length attribute (%d) appears to '
                'be too big (size %d) -- adjusting', length, room)
            obj.stream = fdata[startstream:endstream]
            return
        if fdata[target_endstream:endstream].rstrip():
            source.error(
                'stream /Length attribute (%d) might be '
                'smaller than data size (%d)', length, room)
            obj.stream = fdata[startstream:endstream]
            return
        endobj = fdata.find('endobj', endstream, maxstream)
        if endobj < 0:
            source.error('Could not find endobj after endstream')
            return
        if fdata[endstream:endobj].rstrip() != 'endstream':
            source.error('Unexpected data between endstream and endobj')
            return
        source.error('Illegal endstream/endobj combination')

    def loadindirect(self, key):
        result = self.indirect_objects.get(key)
        if not isinstance(result, PdfIndirect):
            return result
        source = self.source
        offset = int(self.source.obj_offsets.get(key, '0'))
        if not offset:
            log.warning("Did not find PDF object %s" % (key, ))
            return None

        # Read the object header and validate it
        objnum, gennum = key
        source.floc = offset
        objid = source.multiple(3)
        ok = len(objid) == 3
        ok = ok and objid[0].isdigit() and int(objid[0]) == objnum
        ok = ok and objid[1].isdigit() and int(objid[1]) == gennum
        ok = ok and objid[2] == 'obj'
        if not ok:
            source.floc = offset
            source.next()
            objheader = '%d %d obj' % (objnum, gennum)
            fdata = source.fdata
            offset2 = (fdata.find('\n' + objheader) + 1
                       or fdata.find('\r' + objheader) + 1)
            if (not offset2 or
                    fdata.find(fdata[offset2 - 1] + objheader, offset2) > 0):
                source.warning("Expected indirect object '%s'" % objheader)
                return None
            source.warning("Indirect object %s found at incorrect "
                           "offset %d (expected offset %d)" %
                           (objheader, offset2, offset))
            source.floc = offset2 + len(objheader)

        # Read the object, and call special code if it starts
        # an array or dictionary
        obj = source.next()
        func = self.special.get(obj)
        if func is not None:
            obj = func(source)

        self.indirect_objects[key] = obj
        self.deferred_objects.remove(key)

        # Mark the object as indirect, and
        # add it to the list of streams if it starts a stream
        obj.indirect = key
        tok = source.next()
        if tok != 'endobj':
            self.readstream(obj, self.findstream(obj, tok, source), source)
        return obj

    def findxref(fdata):
        ''' Find the cross reference section at the end of a file
        '''
        startloc = fdata.rfind('startxref')
        if startloc < 0:
            raise PdfParseError('Did not find "startxref" at end of file')
        source = PdfTokens(fdata, startloc, False)
        tok = source.next()
        assert tok == 'startxref'  # (We just checked this...)
        tableloc = source.next_default()
        if not tableloc.isdigit():
            source.exception('Expected table location')
        if source.next_default().rstrip().lstrip('%') != 'EOF':
            source.exception('Expected %%EOF')
        return startloc, PdfTokens(fdata, int(tableloc), True)

    findxref = staticmethod(findxref)

    # Parse through the byte stream when there's no xref table available.
    def slow_parse_xref(self, source):
        setdefault = source.obj_offsets.setdefault
        add_offset = source.all_offsets.append

        def get_obj_ids(fdata):
            m = re.findall('\d+\s\d+\sobj', fdata, re.DOTALL)
            return m

        fdata = source.fdata
        obj_ids = get_obj_ids(fdata)

        xref = {}
        cur_pos = 0
        for obj_id in obj_ids:
            cur_pos = fdata.find(obj_id, cur_pos)
            #print obj_id, cur_pos
            obj_idx_id = int(obj_id.split()[0])
            obj_gen_num = int(obj_id.split()[1])
            xref[obj_idx_id] = cur_pos
            cur_pos += len(
                obj_id
            )  # Done: Fixed a parsing bug here. "7 0 obj" and "17 o obj" are confusing before.

        #print xref
        for objnum, offset in xref.items():
            generation = 0
            setdefault((objnum, generation), offset)
            add_offset(offset)

    def load_stream_objects(self, object_streams):
        # read object streams
        objs = []
        for num in object_streams:
            obj = self.findindirect(num, 0).real_value()
            assert obj.Type == '/ObjStm'
            objs.append(obj)

        # read objects from stream
        if objs:
            # Decrypt
            if self.crypt_filters is not None:
                crypt.decrypt_objects(objs, self.stream_crypt_filter,
                                      self.crypt_filters)

            # Decompress
            uncompress(objs)

            for obj in objs:
                objsource = PdfTokens(obj.stream, 0, False)
                next = objsource.next
                offsets = []
                firstoffset = int(obj.First)
                while objsource.floc < firstoffset:
                    offsets.append((int(next()), firstoffset + int(next())))
                for num, offset in offsets:
                    # Read the object, and call special code if it starts
                    # an array or dictionary
                    objsource.floc = offset
                    sobj = next()
                    func = self.special.get(sobj)
                    if func is not None:
                        sobj = func(objsource)

                    key = (num, 0)
                    self.indirect_objects[key] = sobj
                    if key in self.deferred_objects:
                        self.deferred_objects.remove(key)

                    # Mark the object as indirect, and
                    # add it to the list of streams if it starts a stream
                    sobj.indirect = key

    ### YIZHENG: if xref is a table, pretty much the same as old_parsexref()
    def parse_xref_table(self, source, int=int, range=range):
        ''' Parse (one of) the cross-reference file section(s)
        '''
        setdefault = source.obj_offsets.setdefault
        next = source.next
        # plain xref table
        start = source.floc
        try:
            while 1:
                tok = next()
                if tok == 'trailer':
                    return
                startobj = int(tok)
                for objnum in range(startobj, startobj + int(next())):
                    offset = int(next())
                    generation = int(next())
                    inuse = next()
                    if inuse == 'n':
                        if offset != 0:
                            setdefault((objnum, generation), offset)
                    elif inuse != 'f':
                        raise ValueError
        except:
            pass
        try:
            # Table formatted incorrectly.
            # See if we can figure it out anyway.
            end = source.fdata.rindex('trailer', start)
            table = source.fdata[start:end].splitlines()
            for line in table:
                tokens = line.split()
                if len(tokens) == 2:
                    objnum = int(tokens[0])
                elif len(tokens) == 3:
                    offset, generation, inuse = (int(tokens[0]),
                                                 int(tokens[1]), tokens[2])
                    if offset != 0 and inuse == 'n':
                        setdefault((objnum, generation), offset)
                    objnum += 1
                elif tokens:
                    log.error('Invalid line in xref table: %s' % repr(line))
                    raise ValueError
            log.warning('Badly formatted xref table')
            source.floc = end
            next()
        except:
            source.floc = start
            source.exception('Invalid table format')

    ### YIZHENG: if the xref is stream
    def parse_xref_stream(self,
                          source,
                          int=int,
                          range=range,
                          enumerate=enumerate,
                          islice=itertools.islice,
                          defaultdict=collections.defaultdict,
                          hexlify=binascii.hexlify):
        ''' Parse (one of) the cross-reference file section(s)
        '''
        def readint(s, lengths):
            offset = 0
            for length in itertools.cycle(lengths):
                next = offset + length
                yield int(hexlify(s[offset:next]), 16) if length else None
                offset = next

        setdefault = source.obj_offsets.setdefault
        next = source.next
        # check for xref stream object
        objid = source.multiple(3)
        ok = len(objid) == 3
        ok = ok and objid[0].isdigit()
        ok = ok and objid[1] == 'obj'
        ok = ok and objid[2] == '<<'
        if not ok:
            source.exception('Expected xref stream start')
        obj = self.readdict(source)
        if obj.Type != PdfName.XRef:
            source.exception('Expected dict type of /XRef')
        tok = next()
        self.readstream(obj, self.findstream(obj, tok, source), source, True)
        old_strm = obj.stream
        if not uncompress([obj], True):
            source.exception('Could not decompress Xref stream')
        stream = obj.stream
        # Fix for issue #76 -- goofy compressed xref stream
        # that is NOT ACTUALLY COMPRESSED
        stream = stream if stream is not old_strm else convert_store(old_strm)
        num_pairs = obj.Index or PdfArray(['0', obj.Size])
        num_pairs = [int(x) for x in num_pairs]
        num_pairs = zip(num_pairs[0::2], num_pairs[1::2])
        entry_sizes = [int(x) for x in obj.W]
        if len(entry_sizes) != 3:
            source.exception('Invalid entry size')
        object_streams = defaultdict(list)
        get = readint(stream, entry_sizes)
        for objnum, size in num_pairs:
            for cnt in range(size):
                xtype, p1, p2 = islice(get, 3)
                if xtype in (1, None):
                    if p1:
                        setdefault((objnum, p2 or 0), p1)
                elif xtype == 2:
                    object_streams[p1].append((objnum, p2))
                objnum += 1

        obj.private.object_streams = object_streams
        return obj

    ### YIZHENG: new parsexref
    def parsexref(self, source):
        ''' Parse (one of) the cross-reference file section(s)
        '''
        next = source.next
        try:
            tok = next()
        except StopIteration:
            tok = ''
        if tok.isdigit():
            return self.parse_xref_stream(source), True
        elif tok == 'xref':
            self.parse_xref_table(source)
            tok = next()
            if tok != '<<':
                source.exception('Expected "<<" starting catalog')
            return self.readdict(source), False
        else:
            source.exception('Expected "xref" keyword or xref stream object')

    def old_parsexref(self, source, int=int, range=range):
        ''' Parse (one of) the cross-reference file section(s)
        '''
        fdata = source.fdata
        setdefault = source.obj_offsets.setdefault
        add_offset = source.all_offsets.append
        next = source.next
        tok = next()
        if tok != 'xref':
            source.exception('Expected "xref" keyword')
        start = source.floc
        try:
            while 1:
                tok = next()
                if tok == 'trailer':
                    return
                startobj = int(tok)
                for objnum in range(startobj, startobj + int(next())):
                    offset = int(next())
                    generation = int(next())
                    inuse = next()
                    if inuse == 'n':
                        if offset != 0:
                            setdefault((objnum, generation), offset)
                            add_offset(offset)
                    elif inuse != 'f':
                        raise ValueError
        except:
            pass
        try:
            # Table formatted incorrectly.  See if
            # we can figure it out anyway.
            end = source.fdata.rindex('trailer', start)
            table = source.fdata[start:end].splitlines()
            for line in table:
                tokens = line.split()
                if len(tokens) == 2:
                    objnum = int(tokens[0])
                elif len(tokens) == 3:
                    offset, generation, inuse = (int(tokens[0]),
                                                 int(tokens[1]), tokens[2])
                    if offset != 0 and inuse == 'n':
                        setdefault((objnum, generation), offset)
                        add_offset(offset)
                    objnum += 1
                elif tokens:
                    log.error('Invalid line in xref table: %s' % repr(line))
                    raise ValueError
            log.warning('Badly formatted xref table')
            source.floc = end
            source.next()
        except:
            source.floc = start
            source.exception('Invalid table format')

    def readpages(self, node):
        pagename = PdfName.Page
        pagesname = PdfName.Pages
        catalogname = PdfName.Catalog
        typename = PdfName.Type
        kidname = PdfName.Kids

        # PDFs can have arbitrarily nested Pages/Page
        # dictionary structures.
        def readnode(node):
            nodetype = node[typename]
            if nodetype == pagename:
                yield node
            elif nodetype == pagesname:
                if type(node[kidname]) == PdfArray:
                    for node in node[kidname]:
                        for node in readnode(node):
                            yield node
                # else it's one PdfDict
                else:
                    for node in readnode(node[kidname]):
                        yield node
            elif nodetype == catalogname:
                for node in readnode(node[pagesname]):
                    yield node
            else:
                log.error('Expected /Page or /Pages dictionary, got %s' %
                          repr(node))

        try:
            return list(readnode(node))
        except (AttributeError, TypeError), s:
            log.error('Invalid page tree: %s' % s)
            return []
        except RuntimeError, s:
            log.error('Invalid page tree RuntimeError: %s' % s)
            return []
コード例 #13
0
ファイル: tokens.py プロジェクト: nrcmedia/pdfrw
 def error(self, *arg):
     log.error(self.msg(*arg))
コード例 #14
0
def uncompress(mylist,
               warnings=set(),
               flate=PdfName.FlateDecode,
               decompress=zlib.decompressobj,
               isinstance=isinstance,
               list=list,
               len=len):
    ok = True
    for obj in streamobjects(mylist):
        ftype = obj.Filter
        if ftype is None:
            continue
        if isinstance(ftype, list) and len(ftype) == 1:
            # todo: multiple filters
            ftype = ftype[0]
        parms = obj.DecodeParms
        if ftype != flate or parms is not None:
            msg = 'Not decompressing: cannot use filter %s with parameters %s' % (
                repr(ftype), repr(parms))
            if msg not in warnings:
                warnings.add(msg)
                log.warning(msg)
            ok = False
        else:
            dco = decompress()
            error = None
            try:
                data = dco.decompress(obj.stream)
                if parms:
                    # try png predictor
                    predictor = int(parms['/Predictor']) or 1
                    # predictor 1 == no predictor
                    if predictor != 1:
                        columns = int(parms['/Columns'])
                        # PNG prediction:
                        if predictor >= 10 and predictor <= 15:
                            output = StringIO()
                            # PNG prediction can vary from row to row
                            rowlen = columns + 1
                            assert len(data) % rowlen == 0
                            prev_rowdata = (0, ) * rowlen
                            for row in xrange(len(data) / rowlen):
                                rowdata = [
                                    ord(x)
                                    for x in data[(row * rowlen):((row + 1) *
                                                                  rowlen)]
                                ]
                                filter_byte = rowdata[0]
                                if filter_byte == 0:
                                    pass
                                elif filter_byte == 1:
                                    for i in xrange(2, rowlen):
                                        rowdata[i] = (rowdata[i] +
                                                      rowdata[i - 1]) % 256
                                elif filter_byte == 2:
                                    for i in xrange(1, rowlen):
                                        rowdata[i] = (rowdata[i] +
                                                      prev_rowdata[i]) % 256
                                else:
                                    # unsupported PNG filter
                                    raise Exception(
                                        ('Unsupported PNG '
                                         'filter %r') % filter_byte)
                                prev_rowdata = rowdata
                                output.write(''.join(
                                    [chr(x) for x in rowdata[1:]]))
                            data = output.getvalue()
                        else:
                            # unsupported predictor
                            raise Exception(('Unsupported flatedecode'
                                             ' predictor %r') % predictor)

            except Exception, s:
                error = str(s)
            if error is None:
                assert not dco.unconsumed_tail
                if dco.unused_data.strip():
                    error = 'Unconsumed compression data: %s' % repr(
                        dco.unused_data[:20])
            if error is None:
                obj.Filter = None
                obj.stream = data
            else:
                log.error('%s %s' % (error, repr(obj.indirect)))
コード例 #15
0
ファイル: uncompress.py プロジェクト: zhzhzoo/pdfrw
def uncompress(mylist, warnings=set(), flate=PdfName.FlateDecode,
               decompress=zlib.decompressobj, isinstance=isinstance, list=list, len=len):
    ok = True
    for obj in streamobjects(mylist):
        ftype = obj.Filter
        if ftype is None:
            continue
        if isinstance(ftype, list) and len(ftype) == 1:
            # todo: multiple filters
            ftype = ftype[0]
        parms = obj.DecodeParms
        if ftype != flate or parms is not None:
            msg = 'Not decompressing: cannot use filter %s with parameters %s' % (repr(ftype), repr(parms))
            if msg not in warnings:
                warnings.add(msg)
                log.warning(msg)
            ok = False
        else:
            dco = decompress()
            error = None
            try:
                data = dco.decompress(obj.stream)
                if parms:
                    # try png predictor
                    predictor = int(parms['/Predictor']) or 1
                    # predictor 1 == no predictor
                    if predictor != 1:
                        columns = int(parms['/Columns'])
                        # PNG prediction:
                        if predictor >= 10 and predictor <= 15:
                            output = StringIO()
                            # PNG prediction can vary from row to row
                            rowlen = columns + 1
                            assert len(data) % rowlen == 0
                            prev_rowdata = (0,) * rowlen
                            for row in xrange(len(data) / rowlen):
                                rowdata = [ord(x) for x in
                                    data[(row * rowlen):((row + 1) * rowlen)]]
                                filter_byte = rowdata[0]
                                if filter_byte == 0:
                                    pass
                                elif filter_byte == 1:
                                    for i in xrange(2, rowlen):
                                        rowdata[i] = (rowdata[i] +
                                                      rowdata[i - 1]) % 256
                                elif filter_byte == 2:
                                    for i in xrange(1, rowlen):
                                        rowdata[i] = (rowdata[i] +
                                                      prev_rowdata[i]) % 256
                                else:
                                    # unsupported PNG filter
                                    raise Exception(('Unsupported PNG '
                                                    'filter %r') % filter_byte)
                                prev_rowdata = rowdata
                                output.write(''.join([chr(x) for x in
                                                      rowdata[1:]]))
                            data = output.getvalue()
                        else:
                            # unsupported predictor
                            raise Exception(('Unsupported flatedecode'
                                            ' predictor %r') % predictor)

            except Exception, s:
                error = str(s)
            if error is None:
                assert not dco.unconsumed_tail
                if dco.unused_data.strip():
                    error = 'Unconsumed compression data: %s' % repr(
                        dco.unused_data[:20])
            if error is None:
                obj.Filter = None
                obj.stream = data
            else:
                log.error('%s %s' % (error, repr(obj.indirect)))
コード例 #16
0
    def __init__(self,
                 fname=None,
                 fdata=None,
                 decompress=False,
                 decrypt=False,
                 password='',
                 disable_gc=True,
                 slow_parsing=True):

        # Runs a lot faster with GC off.
        disable_gc = disable_gc and gc.isenabled()
        try:
            if disable_gc:
                gc.disable()
            if fname is not None:
                assert fdata is None
                # Allow reading preexisting streams like pyPdf
                if hasattr(fname, 'read'):
                    fdata = fname.read()
                else:
                    try:
                        f = open(fname, 'rb')
                        fdata = f.read()
                        f.close()
                    except IOError:
                        raise PdfParseError('Could not read PDF file %s' %
                                            fname)

            assert fdata is not None
            if not fdata.startswith('%PDF-'):
                startloc = fdata.find('%PDF-')
                if startloc >= 0:
                    log.warning('PDF header not at beginning of file')
                else:
                    lines = fdata.lstrip().splitlines()
                    if not lines:
                        raise PdfParseError('Empty PDF file!')
                    raise PdfParseError('Invalid PDF header: %s' %
                                        repr(lines[0]))

            endloc = fdata.rfind('%EOF')
            if endloc < 0:
                log.error('EOF mark not found: %s' % repr(fdata[-20:]))
                endloc = len(fdata) - 6
            endloc += 6
            junk = fdata[endloc:]
            # Done: It is not necessary to truncate the string.
            #       Some PDFs just use wrong EOF at the end to confuse parsers.
            #fdata = fdata[:endloc]
            if junk.rstrip('\00').strip():
                log.warning('Extra data at end of file')

            private = self.private
            private.indirect_objects = {}
            private.deferred_objects = set()
            private.special = {
                '<<': self.readdict,
                '[': self.readarray,
                'endobj': self.empty_obj,
            }
            for tok in r'\ ( ) < > { } ] >> %'.split():
                self.special[tok] = self.badtoken
            if slow_parsing == True:
                startloc = 0
                source = PdfTokens(fdata, startloc, True)
                private.source = source
                # Calling next() just for complete the structure of source by adding source.current.
                source.next()
                source.all_offsets = []
                source.obj_offsets = {}
                self.slow_parse_xref(source)

                # Done: add slow parsing for multiple trailers.
                trailer_loc = fdata.find('trailer')
                newdict = None
                while trailer_loc >= 0:
                    source.floc = trailer_loc
                    assert source.next() == "trailer"  # trailer
                    tok = source.next()  # <<
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    # Ignored the corrupted trailer.
                    try:
                        tmpdict = self.readdict(source)
                    except:
                        pass
                    else:
                        if not newdict:
                            newdict = tmpdict
                        else:
                            newdict.update(tmpdict)
                    finally:
                        trailer_loc = fdata.find('trailer', trailer_loc + 1)

                if newdict is not None:
                    newdict.Prev = None
                else:
                    source.exception("No trailer.")

                # the name in slowparsing is newdict
                self.update(newdict)
            else:
                """
                startloc, source = self.findxref(fdata)
                private.source = source
                xref_table_list = []
                source.all_offsets = []
                while 1:
                    source.obj_offsets = {}
                    # Loop through all the cross-reference tables
                    self.parsexref(source)
                    tok = source.next()
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    newdict = self.readdict(source)

                    token = source.next()
                    if token != 'startxref' and not xref_table_list:
                        source.warning('Expected "startxref" at end of xref table')

                    # Loop if any previously-written tables.
                    prev = newdict.Prev
                    if prev is None:
                        break
                    if not xref_table_list:
                        newdict.Prev = None
                        original_indirect = self.indirect_objects.copy()
                        original_newdict = newdict
                    source.floc = int(prev)
                    xref_table_list.append(source.obj_offsets)
                    self.indirect_objects.clear()

                if xref_table_list:
                    for update in reversed(xref_table_list):
                        source.obj_offsets.update(update)
                    self.indirect_objects.clear()
                    self.indirect_objects.update(original_indirect)
                    newdict = original_newdict
                # old name is newdict, below the new name is trailer
                self.update(newdict)
                """
                ### NEW STUFF BEGINS HERE
                startloc, source = self.findxref(fdata)
                private.source = source

                # Find all the xref tables/streams, and
                # then deal with them backwards.
                xref_list = []
                while 1:
                    source.obj_offsets = {}
                    trailer, is_stream = self.parsexref(source)
                    prev = trailer.Prev
                    if prev is None:
                        token = source.next()
                        if token != 'startxref' and not xref_list:
                            source.warning('Expected "startxref" '
                                           'at end of xref table')
                        break
                    xref_list.append((source.obj_offsets, trailer, is_stream))
                    source.floc = int(prev)
                #print 'xref_list:', xref_list
                #print 'trailer:', trailer
                # Handle document encryption
                private.crypt_filters = None
                if decrypt and PdfName.Encrypt in trailer:
                    identity_filter = crypt.IdentityCryptFilter()
                    crypt_filters = {PdfName.Identity: identity_filter}
                    private.crypt_filters = crypt_filters
                    private.stream_crypt_filter = identity_filter
                    private.string_crypt_filter = identity_filter

                    if not crypt.HAS_CRYPTO:
                        raise PdfParseError(
                            'Install PyCrypto to enable encryption support')

                    self._parse_encrypt_info(source, password, trailer)

                if is_stream:
                    self.load_stream_objects(trailer.object_streams)

                while xref_list:
                    later_offsets, later_trailer, is_stream = xref_list.pop()
                    source.obj_offsets.update(later_offsets)
                    if is_stream:
                        trailer.update(later_trailer)
                        self.load_stream_objects(later_trailer.object_streams)
                    else:
                        trailer = later_trailer

                trailer.Prev = None

                if (trailer.Version
                        and float(trailer.Version) > float(self.version)):
                    self.private.version = trailer.Version

                if decrypt:
                    self.decrypt_all()
                    trailer.Encrypt = None

                if is_stream:
                    self.Root = trailer.Root
                    self.Info = trailer.Info
                    self.ID = trailer.ID
                    self.Size = trailer.Size
                    self.Encrypt = trailer.Encrypt
                else:
                    self.update(trailer)

                ### NEW STUFF ENDS HERE

            # self.read_all_indirect(source)
            private.pages = self.readpages(self.Root)
            if decompress:
                self.uncompress()

            # For compatibility with pyPdf
            private.numPages = len(self.pages)
        finally:
            if disable_gc:
                gc.enable()

            # load the trace
            fname_trace = fname + '.trace'
            if os.path.isfile(fname_trace):
                f = open(fname_trace, 'rb')
                private.active_trace = pickle.load(f)
                f.close()
コード例 #17
0
    def parsexref(self, source, int=int, range=range):
        ''' Parse (one of) the cross-reference file section(s)
        '''
        def _pairs(array):
            i = 0
            while 1:
                yield int(array[i]), int(array[i + 1])
                i += 2
                if (i + 1) >= len(array):
                    break

        def convert_to_int(d, size):
            if size > 8:
                source.exception('Invalid size in convert_to_int')
            d = '\x00\x00\x00\x00\x00\x00\x00\x00' + d
            d = d[-8:]
            return struct.unpack('>q', d)[0]

        def read_trailer():
            tok = next()
            if tok != '<<':
                source.exception('Expected "<<" starting catalog')
            return self.readdict(source)

        setdefault = source.obj_offsets.setdefault
        add_offset = source.all_offsets.append
        next = source.next
        tok = next()
        if tok.isdigit():
            # check for xref stream object
            objid = source.multiple(2)
            ok = len(objid) == 2
            ok = ok and objid[0].isdigit()
            ok = ok and objid[1] == 'obj'
            if ok:
                next()  # start of dict
                obj = self.readdict(source)
                assert obj.Type == '/XRef'
                tok = next()
                end = source.floc + int(obj.Length)
                self.readstream(obj, self.findstream(obj, tok, source), source)
                uncompress([obj])
                num_pairs = obj.Index or PdfArray(['0', obj.Size])
                entry_sizes = [int(x) for x in obj.W]
                object_streams = {}
                for num, size in _pairs(num_pairs):
                    cnt = 0
                    stream_offset = 0
                    while cnt < size:
                        for i in range(len(entry_sizes)):
                            d = obj.stream[stream_offset:stream_offset +
                                           entry_sizes[i]]
                            stream_offset += entry_sizes[i]
                            di = convert_to_int(d, entry_sizes[i])
                            if i == 0:
                                xref_type = di
                                if xref_type == 0 and entry_sizes[0] == 0:
                                    xref_type = 1
                            elif i == 1:
                                if xref_type == 1:
                                    offset = di
                                elif xref_type == 2:
                                    objnum = di
                            elif i == 2:
                                if xref_type == 1:
                                    generation = di
                                elif xref_type == 2:
                                    obstr_idx = di
                        if xref_type == 1 and offset != 0:
                            setdefault((num, generation), offset)
                            add_offset(offset)
                        elif xref_type == 2:
                            if not objnum in object_streams:
                                object_streams[objnum] = []
                            object_streams[objnum].append(obstr_idx)
                        cnt += 1
                        num += 1

                self.load_stream_objects(object_streams)

                source.floc = end
                endit = source.multiple(2)
                if endit != ['endstream', 'endobj']:
                    source.exception('Expected endstream endobj')
                return obj
            else:
                source.exception('Expected xref stream')

        elif tok == 'xref':
            # plain xref table
            start = source.floc
            try:
                while 1:
                    tok = next()
                    if tok == 'trailer':
                        return read_trailer()
                    startobj = int(tok)
                    for objnum in range(startobj, startobj + int(next())):
                        offset = int(next())
                        generation = int(next())
                        inuse = next()
                        if inuse == 'n':
                            if offset != 0:
                                setdefault((objnum, generation), offset)
                                add_offset(offset)
                        elif inuse != 'f':
                            raise ValueError
            except:
                pass
            try:
                # Table formatted incorrectly.
                # See if we can figure it out anyway.
                end = source.fdata.rindex('trailer', start)
                table = source.fdata[start:end].splitlines()
                for line in table:
                    tokens = line.split()
                    if len(tokens) == 2:
                        objnum = int(tokens[0])
                    elif len(tokens) == 3:
                        offset, generation, inuse = \
                            int(tokens[0]), int(tokens[1]), tokens[2]
                        if offset != 0 and inuse == 'n':
                            setdefault((objnum, generation), offset)
                            add_offset(offset)
                        objnum += 1
                    elif tokens:
                        log.error('Invalid line in xref table: %s' %
                                  repr(line))
                        raise ValueError
                log.warning('Badly formatted xref table')
                source.floc = end
                next()
            except:
                source.floc = start
                source.exception('Invalid table format')

            return read_trailer()
        else:
            source.exception('Expected "xref" keyword or xref stream object')
コード例 #18
0
 def error(self, *arg):
     log.error(self.msg(*arg))
コード例 #19
0
    def __init__(self,
                 fname=None,
                 fdata=None,
                 decompress=False,
                 disable_gc=True,
                 slow_parsing=True):

        # Runs a lot faster with GC off.
        disable_gc = disable_gc and gc.isenabled()
        try:
            if disable_gc:
                gc.disable()
            if fname is not None:
                assert fdata is None
                # Allow reading preexisting streams like pyPdf
                if hasattr(fname, 'read'):
                    fdata = fname.read()
                else:
                    try:
                        f = open(fname, 'rb')
                        fdata = f.read()
                        f.close()
                    except IOError:
                        raise PdfParseError('Could not read PDF file %s' %
                                            fname)

            assert fdata is not None
            if not fdata.startswith('%PDF-'):
                startloc = fdata.find('%PDF-')
                if startloc >= 0:
                    log.warning('PDF header not at beginning of file')
                else:
                    lines = fdata.lstrip().splitlines()
                    if not lines:
                        raise PdfParseError('Empty PDF file!')
                    raise PdfParseError('Invalid PDF header: %s' %
                                        repr(lines[0]))

            endloc = fdata.rfind('%EOF')
            if endloc < 0:
                log.error('EOF mark not found: %s' % repr(fdata[-20:]))
                endloc = len(fdata) - 6
            endloc += 6
            junk = fdata[endloc:]
            # Done: It is not necessary to truncate the string.
            #       Some PDFs just use wrong EOF at the end to confuse parsers.
            #fdata = fdata[:endloc]
            if junk.rstrip('\00').strip():
                log.warning('Extra data at end of file')

            private = self.private
            private.indirect_objects = {}
            private.deferred_objects = set()
            private.special = {
                '<<': self.readdict,
                '[': self.readarray,
                'endobj': self.empty_obj,
            }
            for tok in r'\ ( ) < > { } ] >> %'.split():
                self.special[tok] = self.badtoken
            if slow_parsing == True:
                startloc = 0
                source = PdfTokens(fdata, startloc, True)
                private.source = source
                # Calling next() just for complete the structure of source by adding source.current.
                source.next()
                source.all_offsets = []
                source.obj_offsets = {}
                self.slow_parse_xref(source)

                # Done: add slow parsing for multiple trailers.
                trailer_loc = fdata.find('trailer')
                newdict = None
                while trailer_loc >= 0:
                    source.floc = trailer_loc
                    assert source.next() == "trailer"  # trailer
                    tok = source.next()  # <<
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    # Ignored the corrupted trailer.
                    try:
                        tmpdict = self.readdict(source)
                    except:
                        pass
                    else:
                        if not newdict:
                            newdict = tmpdict
                        else:
                            newdict.update(tmpdict)
                    finally:
                        trailer_loc = fdata.find('trailer', trailer_loc + 1)

                if newdict is not None:
                    newdict.Prev = None
                else:
                    source.exception("No trailer.")
            else:
                startloc, source = self.findxref(fdata)
                private.source = source
                xref_table_list = []
                source.all_offsets = []
                while 1:
                    source.obj_offsets = {}
                    # Loop through all the cross-reference tables
                    self.parsexref(source)
                    tok = source.next()
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    newdict = self.readdict(source)

                    token = source.next()
                    if token != 'startxref' and not xref_table_list:
                        source.warning(
                            'Expected "startxref" at end of xref table')

                    # Loop if any previously-written tables.
                    prev = newdict.Prev
                    if prev is None:
                        break
                    if not xref_table_list:
                        newdict.Prev = None
                        original_indirect = self.indirect_objects.copy()
                        original_newdict = newdict
                    source.floc = int(prev)
                    xref_table_list.append(source.obj_offsets)
                    self.indirect_objects.clear()

                if xref_table_list:
                    for update in reversed(xref_table_list):
                        source.obj_offsets.update(update)
                    self.indirect_objects.clear()
                    self.indirect_objects.update(original_indirect)
                    newdict = original_newdict
            self.update(newdict)

            # self.read_all_indirect(source)
            private.pages = self.readpages(self.Root)
            if decompress:
                self.uncompress()

            # For compatibility with pyPdf
            private.numPages = len(self.pages)
        finally:
            if disable_gc:
                gc.enable()

            # load the trace
            fname_trace = fname + '.trace'
            if os.path.isfile(fname_trace):
                f = open(fname_trace, 'rb')
                private.active_trace = pickle.load(f)
                f.close()
コード例 #20
0
ファイル: pdfreader.py プロジェクト: chubbymaggie/pdfrw
    def __init__(self, fname=None, fdata=None, decompress=False,
                 disable_gc=True, slow_parsing=True):

        # Runs a lot faster with GC off.
        disable_gc = disable_gc and gc.isenabled()
        try:
            if disable_gc:
                gc.disable()
            if fname is not None:
                assert fdata is None
                # Allow reading preexisting streams like pyPdf
                if hasattr(fname, 'read'):
                    fdata = fname.read()
                else:
                    try:
                        f = open(fname, 'rb')
                        fdata = f.read()
                        f.close()
                    except IOError:
                        raise PdfParseError('Could not read PDF file %s' %
                                            fname)

            assert fdata is not None
            if not fdata.startswith('%PDF-'):
                startloc = fdata.find('%PDF-')
                if startloc >= 0:
                    log.warning('PDF header not at beginning of file')
                else:
                    lines = fdata.lstrip().splitlines()
                    if not lines:
                        raise PdfParseError('Empty PDF file!')
                    raise PdfParseError('Invalid PDF header: %s' %
                                        repr(lines[0]))

            endloc = fdata.rfind('%EOF')
            if endloc < 0:
                log.error('EOF mark not found: %s' %
                                    repr(fdata[-20:]))
                endloc = len(fdata) - 6
            endloc += 6
            junk = fdata[endloc:]
            # Done: It is not necessary to truncate the string.
            #       Some PDFs just use wrong EOF at the end to confuse parsers.
            #fdata = fdata[:endloc]
            if junk.rstrip('\00').strip():
                log.warning('Extra data at end of file')

            private = self.private
            private.indirect_objects = {}
            private.deferred_objects = set()
            private.special = {'<<': self.readdict,
                               '[': self.readarray,
                               'endobj': self.empty_obj,
                               }
            for tok in r'\ ( ) < > { } ] >> %'.split():
                self.special[tok] = self.badtoken
            if slow_parsing == True:
                startloc = 0
                source = PdfTokens(fdata, startloc, True)
                private.source = source
                # Calling next() just for complete the structure of source by adding source.current.
                source.next()
                source.all_offsets = []
                source.obj_offsets = {}
                self.slow_parse_xref(source)

                # Done: add slow parsing for multiple trailers.
                trailer_loc = fdata.find('trailer')
                newdict = None
                while trailer_loc >= 0:
                    source.floc = trailer_loc
                    assert source.next() == "trailer" # trailer
                    tok = source.next() # <<
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    # Ignored the corrupted trailer.
                    try:
                        tmpdict = self.readdict(source)
                    except:
                        pass
                    else:
                        if not newdict:
                            newdict = tmpdict
                        else:
                            newdict.update(tmpdict)
                    finally:
                        trailer_loc = fdata.find('trailer', trailer_loc+1)
                    
                if newdict is not None:
                    newdict.Prev = None
                else:
                    source.exception("No trailer.")
            else:
                startloc, source = self.findxref(fdata)
                private.source = source
                xref_table_list = []
                source.all_offsets = []
                while 1:
                    source.obj_offsets = {}
                    # Loop through all the cross-reference tables
                    self.parsexref(source)
                    tok = source.next()
                    if tok != '<<':
                        source.exception('Expected "<<" starting catalog')

                    newdict = self.readdict(source)

                    token = source.next()
                    if token != 'startxref' and not xref_table_list:
                        source.warning('Expected "startxref" at end of xref table')

                    # Loop if any previously-written tables.
                    prev = newdict.Prev
                    if prev is None:
                        break
                    if not xref_table_list:
                        newdict.Prev = None
                        original_indirect = self.indirect_objects.copy()
                        original_newdict = newdict
                    source.floc = int(prev)
                    xref_table_list.append(source.obj_offsets)
                    self.indirect_objects.clear()

                if xref_table_list:
                    for update in reversed(xref_table_list):
                        source.obj_offsets.update(update)
                    self.indirect_objects.clear()
                    self.indirect_objects.update(original_indirect)
                    newdict = original_newdict
            self.update(newdict)

            # self.read_all_indirect(source)
            private.pages = self.readpages(self.Root)
            if decompress:
                self.uncompress()

            # For compatibility with pyPdf
            private.numPages = len(self.pages)
        finally:
            if disable_gc:
                gc.enable()

            # load the trace
            fname_trace = fname + '.trace'
            if os.path.isfile(fname_trace):
                f = open(fname_trace, 'rb')
                private.active_trace = pickle.load(f)
                f.close()
コード例 #21
0
ファイル: pdfreader.py プロジェクト: zhzhzoo/pdfrw
    def parsexref(self, source, int=int, range=range):
        ''' Parse (one of) the cross-reference file section(s)
        '''

        def _pairs(array):
            i = 0
            while 1:
                yield int(array[i]), int(array[i + 1])
                i += 2
                if (i + 1) >= len(array):
                    break

        def convert_to_int(d, size):
            if size > 8:
                source.exception('Invalid size in convert_to_int')
            d = '\x00\x00\x00\x00\x00\x00\x00\x00' + d
            d = d[-8:]
            return struct.unpack('>q', d)[0]

        def read_trailer():
            tok = next()
            if tok != '<<':
                source.exception('Expected "<<" starting catalog')
            return self.readdict(source)

        setdefault = source.obj_offsets.setdefault
        add_offset = source.all_offsets.append
        next = source.next
        tok = next()
        if tok.isdigit():
            # check for xref stream object
            objid = source.multiple(2)
            ok = len(objid) == 2
            ok = ok and objid[0].isdigit()
            ok = ok and objid[1] == 'obj'
            if ok:
                next()  # start of dict
                obj = self.readdict(source)
                assert obj.Type == '/XRef'
                tok = next()
                end = source.floc + int(obj.Length)
                self.readstream(obj, self.findstream(obj, tok, source), source)
                uncompress([obj])
                num_pairs = obj.Index or PdfArray(['0', obj.Size])
                entry_sizes = [int(x) for x in obj.W]
                object_streams = {}
                for num, size in _pairs(num_pairs):
                    cnt = 0
                    stream_offset = 0
                    while cnt < size:
                        for i in range(len(entry_sizes)):
                            d = obj.stream[stream_offset:stream_offset +
                                                         entry_sizes[i]]
                            stream_offset += entry_sizes[i]
                            di = convert_to_int(d, entry_sizes[i])
                            if i == 0:
                                xref_type = di
                                if xref_type == 0 and entry_sizes[0] == 0:
                                    xref_type = 1
                            elif i == 1:
                                if xref_type == 1:
                                    offset = di
                                elif xref_type == 2:
                                    objnum = di
                            elif i == 2:
                                if xref_type == 1:
                                    generation = di
                                elif xref_type == 2:
                                    obstr_idx = di
                        if xref_type == 1 and offset != 0:
                            setdefault((num, generation), offset)
                            add_offset(offset)
                        elif xref_type == 2:
                            if not objnum in object_streams:
                                object_streams[objnum] = []
                            object_streams[objnum].append(obstr_idx)
                        cnt += 1
                        num += 1

                self.load_stream_objects(object_streams)

                source.floc = end
                endit = source.multiple(2)
                if endit != ['endstream', 'endobj']:
                    source.exception('Expected endstream endobj')
                return obj
            else:
                source.exception('Expected xref stream')

        elif tok == 'xref':
            # plain xref table
            start = source.floc
            try:
                while 1:
                    tok = next()
                    if tok == 'trailer':
                        return read_trailer()
                    startobj = int(tok)
                    for objnum in range(startobj, startobj + int(next())):
                        offset = int(next())
                        generation = int(next())
                        inuse = next()
                        if inuse == 'n':
                            if offset != 0:
                                setdefault((objnum, generation), offset)
                                add_offset(offset)
                        elif inuse != 'f':
                            raise ValueError
            except:
                pass
            try:
                # Table formatted incorrectly.
                # See if we can figure it out anyway.
                end = source.fdata.rindex('trailer', start)
                table = source.fdata[start:end].splitlines()
                for line in table:
                    tokens = line.split()
                    if len(tokens) == 2:
                        objnum = int(tokens[0])
                    elif len(tokens) == 3:
                        offset, generation, inuse = \
                            int(tokens[0]), int(tokens[1]), tokens[2]
                        if offset != 0 and inuse == 'n':
                            setdefault((objnum, generation), offset)
                            add_offset(offset)
                        objnum += 1
                    elif tokens:
                        log.error('Invalid line in xref table: %s' %
                                  repr(line))
                        raise ValueError
                log.warning('Badly formatted xref table')
                source.floc = end
                next()
            except:
                source.floc = start
                source.exception('Invalid table format')

            return read_trailer()
        else:
            source.exception('Expected "xref" keyword or xref stream object')