- moved parser files from old interpeter into parser folder

- renamed interpreter_v2 to just interpreter
- removed legacy code
This commit is contained in:
ari committed 2018-12-20 10:41:10 -05:00
1 parent 487f9b6c08
commit 3ba0b95001
81 files changed
+138 -2565

No files matched your search

-44
View File
@@ -1,44 +0,0 @@
from ..utils import LATIN_ENCODING, name2unicode
from .PSStackParser import PSLiteral
class EncodingDB(object):
std2unicode = {}
mac2unicode = {}
win2unicode = {}
pdf2unicode = {}
for (name, std, mac, win, pdf) in LATIN_ENCODING:
c = name2unicode(name)
if std:
std2unicode[std] = c
if mac:
mac2unicode[mac] = c
if win:
win2unicode[win] = c
if pdf:
pdf2unicode[pdf] = c
encodings = {
'StandardEncoding': std2unicode,
'MacRomanEncoding': mac2unicode,
'WinAnsiEncoding': win2unicode,
'PDFDocEncoding': pdf2unicode,
}
@classmethod
def get_encoding(klass, name, diff=None):
cid2unicode = klass.encodings.get(name, klass.std2unicode)
if diff:
cid2unicode = cid2unicode.copy()
cid = 0
for x in diff:
if isinstance(x, int):
cid = x
elif isinstance(x, PSLiteral):
try:
cid2unicode[cid] = name2unicode(x.name)
except KeyError:
pass
cid += 1
return cid2unicode
-109
View File
@@ -1,109 +0,0 @@
import re
import operator
from typing import List
from io import BytesIO
from ..utils import int2byte, choplist, settings
from .PSStackParser import literal_name
from .PSStackParser import PSStackParser
from .PSStackParser import PSEOF, PSTypeError
from .PSStackParser import KWD
from .PDFStream import PDFStream
class PDFContentParser(PSStackParser):
def __init__(self, streams: List[PDFStream]):
self.streams = streams
self.istream = 0
PSStackParser.__init__(self, None)
return
def fillfp(self):
if not self.fp:
if self.istream < len(self.streams):
strm = PDFStream.validated_stream(self.streams[self.istream])
self.istream += 1
else:
raise PSEOF('Unexpected EOF, file truncated?')
self.fp = BytesIO(strm.get_data())
return
def seek(self, pos):
self.fillfp()
PSStackParser.seek(self, pos)
return
def fillbuf(self):
if self.charpos < len(self.buf):
return
while 1:
self.fillfp()
self.bufpos = self.fp.tell()
self.buf = self.fp.read(self.BUFSIZ)
if self.buf:
break
self.fp = None
self.charpos = 0
return
def get_inline_data(self, pos, target=b'EI'):
self.seek(pos)
i = 0
data = b''
while i <= len(target):
self.fillbuf()
if i:
c = operator.getitem(self.buf,self.charpos)
c = int2byte(c)
data += c
self.charpos += 1
if len(target) <= i and c.isspace():
i += 1
elif i < len(target) and c == (int2byte(target[i])):
i += 1
else:
i = 0
else:
try:
j = self.buf.index(target[0], self.charpos)
data += self.buf[self.charpos:j+1]
self.charpos = j+1
i = 1
except ValueError:
data += self.buf[self.charpos:]
self.charpos = len(self.buf)
data = data[:-(len(target)+1)] # strip the last part
data = re.sub(br'(\x0d\x0a|[\x0d\x0a])$', b'', data)
return (pos, data)
def flush(self):
self.add_results(*self.popall())
return
KEYWORD_BI = KWD(b'BI')
KEYWORD_ID = KWD(b'ID')
KEYWORD_EI = KWD(b'EI')
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(b'ID '))
obj = PDFStream(d, data)
self.push((pos, obj))
self.push((pos, self.KEYWORD_EI))
except PSTypeError:
if settings.STRICT:
raise
else:
self.push((pos, token))
return
@@ -1,235 +0,0 @@
import hashlib as md5
from Crypto.Cipher import ARC4, AES
from Crypto.Hash import SHA256
from .exceptions import * # pylint: disable=W0614
from ..PDFStream import int_value, str_value, dict_value
from ..PSStackParser import literal_name
from ...utils import int2byte, struct
## PDFSecurityHandler
##
class PDFStandardSecurityHandler(object):
PASSWORD_PADDING = (b'(\xbfN^Nu\x8aAd\x00NV\xff\xfa\x01\x08'
b'..\x00\xb6\xd0h>\x80/\x0c\xa9\xfedSiz')
supported_revisions = (2, 3)
def __init__(self, docid, param, password=''):
self.docid = docid
self.param = param
self.password = password
self.encrypt_metadata = False
self.init()
return
def init(self):
self.init_params()
if self.r not in self.supported_revisions:
raise PDFEncryptionError('Unsupported revision: param=%r' % self.param)
self.init_key()
return
def init_params(self):
self.v = int_value(self.param.get('V', 0))
self.r = int_value(self.param['R'])
self.p = int_value(self.param['P'])
self.o = str_value(self.param['O'])
self.u = str_value(self.param['U'])
self.length = int_value(self.param.get('Length', 40))
return
def init_key(self):
self.key = self.authenticate(self.password)
if self.key is None:
raise PDFPasswordIncorrect
return
def is_printable(self):
return bool(self.p & 4)
def is_modifiable(self):
return bool(self.p & 8)
def is_extractable(self):
return bool(self.p & 16)
def compute_u(self, key):
if self.r == 2:
# Algorithm 3.4
return ARC4.new(key).encrypt(self.PASSWORD_PADDING) # 2
else:
# Algorithm 3.5
hash = md5.md5(self.PASSWORD_PADDING) # 2
hash.update(self.docid[0]) # 3
result = ARC4.new(key).encrypt(hash.digest()) # 4
for i in range(1, 20): # 5
k = b''.join(int2byte(c ^ i) for c in iter(key))
result = ARC4.new(k).encrypt(result)
result += result # 6
return result
def compute_encryption_key(self, password):
# Algorithm 3.2
password = (password + self.PASSWORD_PADDING)[:32] # 1
hash = md5.md5(password) # 2
hash.update(self.o) # 3
hash.update(struct.pack('<l', self.p)) # 4
hash.update(self.docid[0]) # 5
if self.r >= 4:
if not self.encrypt_metadata:
hash.update(b'\xff\xff\xff\xff')
result = hash.digest()
n = 5
if self.r >= 3:
n = self.length // 8
for _ in range(50):
result = md5.md5(result[:n]).digest()
return result[:n]
def authenticate(self, password):
password = password.encode("latin1")
key = self.authenticate_user_password(password)
if key is None:
key = self.authenticate_owner_password(password)
return key
def authenticate_user_password(self, password):
key = self.compute_encryption_key(password)
if self.verify_encryption_key(key):
return key
else:
return None
def verify_encryption_key(self, key):
# Algorithm 3.6
u = self.compute_u(key)
if self.r == 2:
return u == self.u
return u[:16] == self.u[:16]
def authenticate_owner_password(self, password):
# Algorithm 3.7
password = (password + self.PASSWORD_PADDING)[:32]
hash = md5.md5(password)
if self.r >= 3:
for _ in range(50):
hash = md5.md5(hash.digest())
n = 5
if self.r >= 3:
n = self.length // 8
key = hash.digest()[:n]
if self.r == 2:
user_password = ARC4.new(key).decrypt(self.o)
else:
user_password = self.o
for i in range(19, -1, -1):
k = b''.join(int2byte(c ^ i) for c in iter(key))
user_password = ARC4.new(k).decrypt(user_password)
return self.authenticate_user_password(user_password)
def decrypt(self, objid, genno, data, attrs=None):
return self.decrypt_rc4(objid, genno, data)
def decrypt_rc4(self, objid, genno, data):
key = self.key + struct.pack('<L', objid)[:3] + struct.pack('<L', genno)[:2]
hash = md5.md5(key)
key = hash.digest()[:min(len(key), 16)]
return ARC4.new(key).decrypt(data)
class PDFStandardSecurityHandlerV4(PDFStandardSecurityHandler):
supported_revisions = (4,)
def init_params(self):
super(PDFStandardSecurityHandlerV4, self).init_params()
self.length = 128
self.cf = dict_value(self.param.get('CF'))
self.stmf = literal_name(self.param['StmF'])
self.strf = literal_name(self.param['StrF'])
self.encrypt_metadata = bool(self.param.get('EncryptMetadata', True))
if self.stmf != self.strf:
raise PDFEncryptionError('Unsupported crypt filter: param=%r' % self.param)
self.cfm = {}
for k, v in self.cf.items():
f = self.get_cfm(literal_name(v['CFM']))
if f is None:
raise PDFEncryptionError('Unknown crypt filter method: param=%r' % self.param)
self.cfm[k] = f
self.cfm['Identity'] = self.decrypt_identity
if self.strf not in self.cfm:
raise PDFEncryptionError('Undefined crypt filter: param=%r' % self.param)
return
def get_cfm(self, name):
if name == 'V2':
return self.decrypt_rc4
elif name == 'AESV2':
return self.decrypt_aes128
else:
return None
def decrypt(self, objid, genno, data, attrs=None, name=None):
if not self.encrypt_metadata and attrs is not None:
t = attrs.get('Type')
if t is not None and literal_name(t) == 'Metadata':
return data
if name is None:
name = self.strf
return self.cfm[name](objid, genno, data)
def decrypt_identity(self, objid, genno, data):
return data
def decrypt_aes128(self, objid, genno, data):
key = self.key + struct.pack('<L', objid)[:3] + struct.pack('<L', genno)[:2] + b'sAlT'
hash = md5.md5(key)
key = hash.digest()[:min(len(key), 16)]
return AES.new(key, mode=AES.MODE_CBC, IV=data[:16]).decrypt(data[16:])
class PDFStandardSecurityHandlerV5(PDFStandardSecurityHandlerV4):
supported_revisions = (5,)
def init_params(self):
super(PDFStandardSecurityHandlerV5, self).init_params()
self.length = 256
self.oe = str_value(self.param['OE'])
self.ue = str_value(self.param['UE'])
self.o_hash = self.o[:32]
self.o_validation_salt = self.o[32:40]
self.o_key_salt = self.o[40:]
self.u_hash = self.u[:32]
self.u_validation_salt = self.u[32:40]
self.u_key_salt = self.u[40:]
return
def get_cfm(self, name):
if name == 'AESV3':
return self.decrypt_aes256
else:
return None
def authenticate(self, password):
password = password.encode('utf-8')[:127]
hash = SHA256.new(password)
hash.update(self.o_validation_salt)
hash.update(self.u)
if hash.digest() == self.o_hash:
hash = SHA256.new(password)
hash.update(self.o_key_salt)
hash.update(self.u)
return AES.new(hash.digest(), mode=AES.MODE_CBC, IV=b'\x00' * 16).decrypt(self.oe)
hash = SHA256.new(password)
hash.update(self.u_validation_salt)
if hash.digest() == self.u_hash:
hash = SHA256.new(password)
hash.update(self.u_key_salt)
return AES.new(hash.digest(), mode=AES.MODE_CBC, IV=b'\x00' * 16).decrypt(self.ue)
return None
def decrypt_aes256(self, objid, genno, data):
return AES.new(self.key, mode=AES.MODE_CBC, IV=data[:16]).decrypt(data[16:])
-232
View File
@@ -1,232 +0,0 @@
import re
from logging import getLogger
from ...utils import settings, choplist, nunpack
from ..PSStackParser import PSEOF
from ..PSStackParser import KWD
from ..PDFParser import PDFStreamParser
from ..PDFStream import PDFStream
from ..PDFStream import dict_value
from .constants import LITERAL_OBJSTM, LITERAL_XREF
from .exceptions import PDFNoValidXRef, PDFSyntaxError
log = getLogger(__name__)
## XRefs
##
class PDFBaseXRef(object):
def get_trailer(self):
raise NotImplementedError
def get_objids(self):
return []
# Must return
# (strmid, index, genno)
# or (None, pos, genno)
def get_pos(self, objid):
raise KeyError(objid)
## PDFXRef
##
class PDFXRef(PDFBaseXRef):
def __init__(self):
self.offsets = {}
self.trailer = {}
return
def __repr__(self):
return '<PDFXRef: offsets=%r>' % (self.offsets.keys())
def load(self, parser):
while True:
try:
(pos, line) = parser.nextline()
if not line.strip():
continue
except PSEOF:
raise PDFNoValidXRef('Unexpected EOF - file corrupted?')
if not line:
raise PDFNoValidXRef('Premature eof: %r' % parser)
if line.startswith(b'trailer'):
parser.seek(pos)
break
f = line.strip().split(b' ')
if len(f) != 2:
raise PDFNoValidXRef('Trailer not found: %r: line=%r' % (parser, line))
try:
(start, nobjs) = map(int, f)
except ValueError:
raise PDFNoValidXRef('Invalid line: %r: line=%r' % (parser, line))
for objid in range(start, start+nobjs):
try:
(_, line) = parser.nextline()
except PSEOF:
raise PDFNoValidXRef('Unexpected EOF - file corrupted?')
f = line.strip().split(b' ')
if len(f) != 3:
raise PDFNoValidXRef('Invalid XRef format: %r, line=%r' % (parser, line))
(pos, genno, use) = f
if use != b'n':
continue
self.offsets[objid] = (None, int(pos), int(genno))
log.info('xref objects: %r', self.offsets)
self.load_trailer(parser)
return
def load_trailer(self, parser):
try:
(_, kwd) = parser.nexttoken()
assert kwd is KWD(b'trailer'), str(kwd)
(_, dic) = parser.nextobject()
except PSEOF:
x = parser.pop(1)
if not x:
raise PDFNoValidXRef('Unexpected EOF - file corrupted')
(_, dic) = x[0]
self.trailer.update(dict_value(dic))
log.debug('trailer=%r', self.trailer)
return
def get_trailer(self):
return self.trailer
def get_objids(self):
return iter(self.offsets.keys())
def get_pos(self, objid):
try:
return self.offsets[objid]
except KeyError:
raise
## PDFXRefFallback
##
class PDFXRefFallback(PDFXRef):
def __repr__(self):
return '<PDFXRefFallback: offsets=%r>' % (self.offsets.keys())
PDFOBJ_CUE = re.compile(r'^(\d+)\s+(\d+)\s+obj\b')
def load(self, parser):
parser.seek(0)
while 1:
try:
(pos, line) = parser.nextline()
except PSEOF:
break
if line.startswith(b'trailer'):
parser.seek(pos)
self.load_trailer(parser)
log.info('trailer: %r', self.trailer)
break
line=line.decode('latin-1') #default pdf encoding
m = self.PDFOBJ_CUE.match(line)
if not m:
continue
(objid, genno) = m.groups()
objid = int(objid)
genno = int(genno)
self.offsets[objid] = (None, pos, genno)
# expand ObjStm.
parser.seek(pos)
(_, obj) = parser.nextobject()
if isinstance(obj, PDFStream) and obj.get('Type') is LITERAL_OBJSTM:
stream = PDFStream.validated_stream(obj)
try:
n = stream['N']
except KeyError:
if settings.STRICT:
raise PDFSyntaxError('N is not defined: %r' % stream)
n = 0
parser1 = PDFStreamParser(stream.get_data())
objs = []
try:
while 1:
(_, obj) = parser1.nextobject()
objs.append(obj)
except PSEOF:
pass
n = min(n, len(objs)//2)
for index in range(n):
objid1 = objs[index*2]
self.offsets[objid1] = (objid, index, 0)
return
## PDFXRefStream
##
class PDFXRefStream(PDFBaseXRef):
def __init__(self):
self.data = None
self.entlen = None
self.fl1 = self.fl2 = self.fl3 = None
self.ranges = []
return
def __repr__(self):
return '<PDFXRefStream: ranges=%r>' % (self.ranges)
def load(self, parser):
(_, objid) = parser.nexttoken() # ignored
(_, genno) = parser.nexttoken() # ignored
(_, kwd) = parser.nexttoken()
(_, stream) = parser.nextobject()
if not isinstance(stream, PDFStream) or stream['Type'] is not LITERAL_XREF:
raise PDFNoValidXRef('Invalid PDF stream spec.')
size = stream['Size']
index_array = stream.get('Index', (0, size))
if len(index_array) % 2 != 0:
raise PDFSyntaxError('Invalid index number')
self.ranges.extend(choplist(2, index_array))
(self.fl1, self.fl2, self.fl3) = stream['W']
self.data = stream.get_data()
self.entlen = self.fl1+self.fl2+self.fl3
self.trailer = stream.attrs
log.info('xref stream: objid=%s, fields=%d,%d,%d',
', '.join(map(repr, self.ranges)),
self.fl1, self.fl2, self.fl3)
return
def get_trailer(self):
return self.trailer
def get_objids(self):
for (start, nobjs) in self.ranges:
for i in range(nobjs):
offset = self.entlen * i
ent = self.data[offset:offset+self.entlen]
f1 = nunpack(ent[:self.fl1], 1)
if f1 == 1 or f1 == 2:
yield start+i
return
def get_pos(self, objid):
index = 0
for (start, nobjs) in self.ranges:
if start <= objid and objid < start+nobjs:
index += objid - start
break
else:
index += nobjs
else:
raise KeyError(objid)
offset = self.entlen * index
ent = self.data[offset:offset+self.entlen]
f1 = nunpack(ent[:self.fl1], 1)
f2 = nunpack(ent[self.fl1:self.fl1+self.fl2])
f3 = nunpack(ent[self.fl1+self.fl2:])
if f1 == 1:
return (None, f2, f3)
elif f1 == 2:
return (f2, f3, 0)
else:
# this is a free object
raise KeyError(objid)
@@ -1,324 +0,0 @@
import hashlib as md5
from Crypto.Cipher import ARC4, AES
from Crypto.Hash import SHA256
from logging import getLogger
from ...utils import settings, choplist, decode_text
from ..PSStackParser import KWD
from ..PSStackParser import literal_name
from ..PSStackParser import PSEOF
from ..PDFStream import int_value, str_value, dict_value, list_value, decipher_all
from ..PDFStream import PDFObjectNotFound, PDFTypeError
from ..PDFParser import PDFStreamParser, PDFStream, PDFSyntaxError
from .exceptions import *
from .PDFSecurityHandler import PDFStandardSecurityHandler
from .PDFSecurityHandler import PDFStandardSecurityHandlerV4
from .PDFSecurityHandler import PDFStandardSecurityHandlerV5
from .PDFXRef import PDFXRefFallback, PDFXRef, PDFXRefStream
from .constants import LITERAL_CATALOG, LITERAL_OBJSTM
log = getLogger(__name__)
## PDFDocument
##
class PDFDocument(object):
"""PDFDocument object represents a PDF document.
Since a PDF file can be very big, normally it is not loaded at
once. So PDF document has to cooperate with a PDF parser in order to
dynamically import the data as processing goes.
Typical usage:
doc = PDFDocument(parser, password)
obj = doc.getobj(objid)
"""
security_handler_registry = {
1: PDFStandardSecurityHandler,
2: PDFStandardSecurityHandler,
}
if AES is not None:
security_handler_registry[4] = PDFStandardSecurityHandlerV4
if SHA256 is not None:
security_handler_registry[5] = PDFStandardSecurityHandlerV5
def __init__(self, parser, password='', caching=True, fallback=True):
"Set the document to use a given PDFParser object."
self.caching = caching
self.xrefs = []
self.info = []
self.catalog = None
self.encryption = None
self.decipher = None
self._parser = None
self._cached_objs = {}
self._parsed_objs = {}
self._parser = parser
self._parser.set_document(self)
self.is_printable = self.is_modifiable = self.is_extractable = True
# Retrieve the information of each header that was appended
# (maybe multiple times) at the end of the document.
try:
pos = self.find_xref(parser)
self.read_xref_from(parser, pos, self.xrefs)
except PDFNoValidXRef:
pass # fallback = True
if fallback:
parser.fallback = True
xref = PDFXRefFallback()
xref.load(parser)
self.xrefs.append(xref)
for xref in self.xrefs:
trailer = xref.get_trailer()
if not trailer:
continue
# If there's an encryption info, remember it.
if 'Encrypt' in trailer:
#assert not self.encryption, str(self.encryption)
self.encryption = (list_value(trailer['ID']),
dict_value(trailer['Encrypt']))
self._initialize_password(password)
if 'Info' in trailer:
self.info.append(dict_value(trailer['Info']))
if 'Root' in trailer:
# Every PDF file must have exactly one /Root dictionary.
self.catalog = dict_value(trailer['Root'])
break
else:
raise PDFSyntaxError('No /Root object! - Is this really a PDF?')
if self.catalog.get('Type') is not LITERAL_CATALOG:
if settings.STRICT:
raise PDFSyntaxError('Catalog not found!')
return
KEYWORD_OBJ = KWD(b'obj')
# _initialize_password(password=b'')
# Perform the initialization with a given password.
def _initialize_password(self, password=''):
(docid, param) = self.encryption
if literal_name(param.get('Filter')) != 'Standard':
raise PDFEncryptionError('Unknown filter: param=%r' % param)
v = int_value(param.get('V', 0))
factory = self.security_handler_registry.get(v)
if factory is None:
raise PDFEncryptionError('Unknown algorithm: param=%r' % param)
handler = factory(docid, param, password)
self.decipher = handler.decrypt
self.is_printable = handler.is_printable()
self.is_modifiable = handler.is_modifiable()
self.is_extractable = handler.is_extractable()
self._parser.fallback = False # need to read streams with exact length
return
def _getobj_objstm(self, stream, index, objid):
if stream.objid in self._parsed_objs:
(objs, n) = self._parsed_objs[stream.objid]
else:
(objs, n) = self._get_objects(stream)
if self.caching:
self._parsed_objs[stream.objid] = (objs, n)
i = n*2+index
try:
obj = objs[i]
except IndexError:
raise PDFSyntaxError('index too big: %r' % index)
return obj
def _get_objects(self, stream):
if stream.get('Type') is not LITERAL_OBJSTM:
if settings.STRICT:
raise PDFSyntaxError('Not a stream object: %r' % stream)
try:
n = stream['N']
except KeyError:
if settings.STRICT:
raise PDFSyntaxError('N is not defined: %r' % stream)
n = 0
parser = PDFStreamParser(stream.get_data())
parser.set_document(self)
objs = []
try:
while 1:
(_, obj) = parser.nextobject()
objs.append(obj)
except PSEOF:
pass
return (objs, n)
def _getobj_parse(self, pos, objid):
self._parser.seek(pos)
(_, objid1) = self._parser.nexttoken() # objid
(_, genno) = self._parser.nexttoken() # genno
(_, kwd) = self._parser.nexttoken()
# #### hack around malformed pdf files
# copied from https://github.com/jaepil/pdfminer3k/blob/master/pdfminer/pdfparser.py#L399
#to solve https://github.com/pdfminer/pdfminer.six/issues/56
#assert objid1 == objid, str((objid1, objid))
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 objid1 != objid:
raise PDFSyntaxError('objid mismatch: %r=%r' % (objid1, objid))
if kwd != KWD(b'obj'):
raise PDFSyntaxError('Invalid object spec: offset=%r' % pos)
(_, obj) = self._parser.nextobject()
return obj
# can raise PDFObjectNotFound
def getobj(self, objid):
assert objid != 0
if not self.xrefs:
raise PDFException('PDFDocument is not initialized')
log.debug('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 = PDFStream.validated_stream(self.getobj(strmid))
obj = self._getobj_objstm(stream, index, objid)
else:
obj = self._getobj_parse(index, objid)
if self.decipher:
obj = decipher_all(self.decipher, objid, genno, obj)
if isinstance(obj, PDFStream):
obj.set_objid(objid, genno)
break
except (PSEOF, PDFSyntaxError):
continue
else:
raise PDFObjectNotFound(objid)
log.debug('register: objid=%r: %r', objid, obj)
if self.caching:
self._cached_objs[objid] = (obj, genno)
return obj
def get_outlines(self):
if 'Outlines' not in self.catalog:
raise PDFNoOutlines
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 lookup_name(self, cat, key):
try:
names = dict_value(self.catalog['Names'])
except (PDFTypeError, KeyError):
raise KeyError((cat, key))
# may raise KeyError
d0 = dict_value(names[cat])
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))
return lookup(d0)
def get_dest(self, name):
try:
# PDF-1.2 or later
obj = self.lookup_name('Dests', name)
except KeyError:
# PDF-1.1 or prior
if 'Dests' not in self.catalog:
raise PDFDestinationNotFound(name)
d0 = dict_value(self.catalog['Dests'])
if name not in d0:
raise PDFDestinationNotFound(name)
obj = d0[name]
return obj
# find_xref
def find_xref(self, parser):
"""Internal function used to locate the first XRef."""
# search the last xref table by scanning the file backwards.
prev = None
for line in parser.revreadlines():
line = line.strip()
log.debug('find_xref: %r', line)
if line == b'startxref':
break
if line:
prev = line
else:
raise PDFNoValidXRef('Unexpected EOF')
log.info('xref found: pos=%r', prev)
return int(prev)
# read xref table
def read_xref_from(self, parser, start, xrefs):
"""Reads XRefs from the given location."""
parser.seek(start)
parser.reset()
try:
(pos, token) = parser.nexttoken()
except PSEOF:
raise PDFNoValidXRef('Unexpected EOF')
log.info('read_xref_from: start=%d, token=%r', start, token)
if isinstance(token, int):
# XRefStream: PDF-1.5
parser.seek(pos)
parser.reset()
xref = PDFXRefStream()
xref.load(parser)
else:
if token is parser.KEYWORD_XREF:
parser.nextline()
xref = PDFXRef()
xref.load(parser)
xrefs.append(xref)
trailer = xref.get_trailer()
log.info('trailer: %r', trailer)
if 'XRefStm' in trailer:
pos = int_value(trailer['XRefStm'])
self.read_xref_from(parser, pos, xrefs)
if 'Prev' in trailer:
# find previous xref
pos = int_value(trailer['Prev'])
self.read_xref_from(parser, pos, xrefs)
return
@@ -1,6 +0,0 @@
from ..PSStackParser import LIT
# some predefined literals and keywords.
LITERAL_OBJSTM = LIT('ObjStm')
LITERAL_XREF = LIT('XRef')
LITERAL_CATALOG = LIT('Catalog')
@@ -1,22 +0,0 @@
from ..PDFParser import PDFSyntaxError
from ..PDFStream import PDFException
## Exceptions
##
class PDFNoValidXRef(PDFSyntaxError):
pass
class PDFNoOutlines(PDFException):
pass
class PDFDestinationNotFound(PDFException):
pass
class PDFEncryptionError(PDFException):
pass
class PDFPasswordIncorrect(PDFEncryptionError):
pass
class PDFTextExtractionNotAllowed(PDFEncryptionError):
pass
-755
View File
@@ -1,755 +0,0 @@
import sys
import struct
from io import BytesIO
from abc import abstractmethod
from .cmapdb import CMapDB
from .cmapdb import CMapParser
from .cmapdb import FileUnicodeMap
from .cmapdb import CMap
from .EncodingDB import EncodingDB, name2unicode
from .PSStackParser import PSStackParser
from .PSStackParser import PSEOF
from .PSStackParser import LIT
from .PSStackParser import KWD
from .PSStackParser import PSLiteral
from .PSStackParser import literal_name
from .PDFStream import PDFException
from .PDFStream import resolve1
from .PDFStream import int_value
from .PDFStream import num_value
from .PDFStream import list_value
from .PDFStream import dict_value
from .PDFStream import PDFStream
from ..utils import settings, FONT_METRICS, apply_matrix_norm, nunpack, choplist, isnumber
from .constants import COMMON_FONT_ATTRIBUTES
def get_widths(seq):
widths = {}
r = []
for v in seq:
if isinstance(v, list):
if r:
char1 = r[-1]
for (i, w) in enumerate(v):
widths[char1+i] = w
r = []
elif isnumber(v):
r.append(v)
if len(r) == 3:
(char1, char2, w) = r
for i in range(char1, char2+1):
widths[i] = w
r = []
return widths
#assert get_widths([1]) == {}
#assert get_widths([1,2,3]) == {1:3, 2:3}
#assert get_widths([1,[2,3],6,[7,8]]) == {1:2,2:3, 6:7,7:8}
def get_widths2(seq):
widths = {}
r = []
for v in seq:
if isinstance(v, list):
if r:
char1 = r[-1]
for (i, (w, vx, vy)) in enumerate(choplist(3, v)):
widths[char1+i] = (w, (vx, vy))
r = []
elif isnumber(v):
r.append(v)
if len(r) == 5:
(char1, char2, w, vx, vy) = r # pylint: disable=E0632
for i in range(char1, char2+1):
widths[i] = (w, (vx, vy))
r = []
return widths
#assert get_widths2([1]) == {}
#assert get_widths2([1,2,3,4,5]) == {1:(3, (4,5)), 2:(3, (4,5))}
#assert get_widths2([1,[2,3,4,5],6,[7,8,9]]) == {1:(2, (3,4)), 6:(7, (8,9))}
## FontMetricsDB
##
class FontMetricsDB(object):
@classmethod
def get_metrics(klass, fontname):
return FONT_METRICS[fontname]
## Type1FontHeaderParser
##
class Type1FontHeaderParser(PSStackParser):
KEYWORD_BEGIN = KWD(b'begin')
KEYWORD_END = KWD(b'end')
KEYWORD_DEF = KWD(b'def')
KEYWORD_PUT = KWD(b'put')
KEYWORD_DICT = KWD(b'dict')
KEYWORD_ARRAY = KWD(b'array')
KEYWORD_READONLY = KWD(b'readonly')
KEYWORD_FOR = KWD(b'for')
KEYWORD_FOR = KWD(b'for')
def __init__(self, data):
PSStackParser.__init__(self, data)
self._cid2unicode = {}
return
def get_encoding(self):
while 1:
try:
(cid, name) = self.nextobject()
except PSEOF:
break
try:
self._cid2unicode[cid] = name2unicode(name)
except KeyError:
pass
return self._cid2unicode
def do_keyword(self, pos, token):
if token is self.KEYWORD_PUT:
((_, key), (_, value)) = self.pop(2) # pylint: disable=E0632
if (isinstance(key, int) and
isinstance(value, PSLiteral)):
self.add_results((key, literal_name(value)))
return
NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-')
## CFFFont
## (Format specified in Adobe Technical Note: #5176
## "The Compact Font Format Specification")
##
def getdict(data):
d = {}
fp = BytesIO(data)
stack = []
while 1:
c = fp.read(1)
if not c:
break
b0 = ord(c)
if b0 <= 21:
d[b0] = stack
stack = []
continue
if b0 == 30:
s = ''
loop = True
while loop:
b = ord(fp.read(1))
for n in (b >> 4, b & 15):
if n == 15:
loop = False
else:
s += NIBBLES[n]
value = float(s)
elif 32 <= b0 and b0 <= 246:
value = b0-139
else:
b1 = ord(fp.read(1))
if 247 <= b0 and b0 <= 250:
value = ((b0-247) << 8)+b1+108
elif 251 <= b0 and b0 <= 254:
value = -((b0-251) << 8)-b1-108
else:
b2 = ord(fp.read(1))
if 128 <= b1:
b1 -= 256
if b0 == 28:
value = b1 << 8 | b2
else:
value = b1 << 24 | b2 << 16 | struct.unpack('>H', fp.read(2))[0]
stack.append(value)
return d
class CFFFont(object):
STANDARD_STRINGS = (
'.notdef', 'space', 'exclam', 'quotedbl', 'numbersign',
'dollar', 'percent', 'ampersand', 'quoteright', 'parenleft',
'parenright', 'asterisk', 'plus', 'comma', 'hyphen', 'period',
'slash', 'zero', 'one', 'two', 'three', 'four', 'five', 'six',
'seven', 'eight', 'nine', 'colon', 'semicolon', 'less', 'equal',
'greater', 'question', 'at', 'A', 'B', 'C', 'D', 'E', 'F', 'G',
'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T',
'U', 'V', 'W', 'X', 'Y', 'Z', 'bracketleft', 'backslash',
'bracketright', 'asciicircum', 'underscore', 'quoteleft', 'a',
'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n',
'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
'braceleft', 'bar', 'braceright', 'asciitilde', 'exclamdown',
'cent', 'sterling', 'fraction', 'yen', 'florin', 'section',
'currency', 'quotesingle', 'quotedblleft', 'guillemotleft',
'guilsinglleft', 'guilsinglright', 'fi', 'fl', 'endash',
'dagger', 'daggerdbl', 'periodcentered', 'paragraph', 'bullet',
'quotesinglbase', 'quotedblbase', 'quotedblright',
'guillemotright', 'ellipsis', 'perthousand', 'questiondown',
'grave', 'acute', 'circumflex', 'tilde', 'macron', 'breve',
'dotaccent', 'dieresis', 'ring', 'cedilla', 'hungarumlaut',
'ogonek', 'caron', 'emdash', 'AE', 'ordfeminine', 'Lslash',
'Oslash', 'OE', 'ordmasculine', 'ae', 'dotlessi', 'lslash',
'oslash', 'oe', 'germandbls', 'onesuperior', 'logicalnot', 'mu',
'trademark', 'Eth', 'onehalf', 'plusminus', 'Thorn',
'onequarter', 'divide', 'brokenbar', 'degree', 'thorn',
'threequarters', 'twosuperior', 'registered', 'minus', 'eth',
'multiply', 'threesuperior', 'copyright', 'Aacute',
'Acircumflex', 'Adieresis', 'Agrave', 'Aring', 'Atilde',
'Ccedilla', 'Eacute', 'Ecircumflex', 'Edieresis', 'Egrave',
'Iacute', 'Icircumflex', 'Idieresis', 'Igrave', 'Ntilde',
'Oacute', 'Ocircumflex', 'Odieresis', 'Ograve', 'Otilde',
'Scaron', 'Uacute', 'Ucircumflex', 'Udieresis', 'Ugrave',
'Yacute', 'Ydieresis', 'Zcaron', 'aacute', 'acircumflex',
'adieresis', 'agrave', 'aring', 'atilde', 'ccedilla', 'eacute',
'ecircumflex', 'edieresis', 'egrave', 'iacute', 'icircumflex',
'idieresis', 'igrave', 'ntilde', 'oacute', 'ocircumflex',
'odieresis', 'ograve', 'otilde', 'scaron', 'uacute',
'ucircumflex', 'udieresis', 'ugrave', 'yacute', 'ydieresis',
'zcaron', 'exclamsmall', 'Hungarumlautsmall', 'dollaroldstyle',
'dollarsuperior', 'ampersandsmall', 'Acutesmall',
'parenleftsuperior', 'parenrightsuperior', 'twodotenleader',
'onedotenleader', 'zerooldstyle', 'oneoldstyle', 'twooldstyle',
'threeoldstyle', 'fouroldstyle', 'fiveoldstyle', 'sixoldstyle',
'sevenoldstyle', 'eightoldstyle', 'nineoldstyle',
'commasuperior', 'threequartersemdash', 'periodsuperior',
'questionsmall', 'asuperior', 'bsuperior', 'centsuperior',
'dsuperior', 'esuperior', 'isuperior', 'lsuperior', 'msuperior',
'nsuperior', 'osuperior', 'rsuperior', 'ssuperior', 'tsuperior',
'ff', 'ffi', 'ffl', 'parenleftinferior', 'parenrightinferior',
'Circumflexsmall', 'hyphensuperior', 'Gravesmall', 'Asmall',
'Bsmall', 'Csmall', 'Dsmall', 'Esmall', 'Fsmall', 'Gsmall',
'Hsmall', 'Ismall', 'Jsmall', 'Ksmall', 'Lsmall', 'Msmall',
'Nsmall', 'Osmall', 'Psmall', 'Qsmall', 'Rsmall', 'Ssmall',
'Tsmall', 'Usmall', 'Vsmall', 'Wsmall', 'Xsmall', 'Ysmall',
'Zsmall', 'colonmonetary', 'onefitted', 'rupiah', 'Tildesmall',
'exclamdownsmall', 'centoldstyle', 'Lslashsmall', 'Scaronsmall',
'Zcaronsmall', 'Dieresissmall', 'Brevesmall', 'Caronsmall',
'Dotaccentsmall', 'Macronsmall', 'figuredash', 'hypheninferior',
'Ogoneksmall', 'Ringsmall', 'Cedillasmall', 'questiondownsmall',
'oneeighth', 'threeeighths', 'fiveeighths', 'seveneighths',
'onethird', 'twothirds', 'zerosuperior', 'foursuperior',
'fivesuperior', 'sixsuperior', 'sevensuperior', 'eightsuperior',
'ninesuperior', 'zeroinferior', 'oneinferior', 'twoinferior',
'threeinferior', 'fourinferior', 'fiveinferior', 'sixinferior',
'seveninferior', 'eightinferior', 'nineinferior',
'centinferior', 'dollarinferior', 'periodinferior',
'commainferior', 'Agravesmall', 'Aacutesmall',
'Acircumflexsmall', 'Atildesmall', 'Adieresissmall',
'Aringsmall', 'AEsmall', 'Ccedillasmall', 'Egravesmall',
'Eacutesmall', 'Ecircumflexsmall', 'Edieresissmall',
'Igravesmall', 'Iacutesmall', 'Icircumflexsmall',
'Idieresissmall', 'Ethsmall', 'Ntildesmall', 'Ogravesmall',
'Oacutesmall', 'Ocircumflexsmall', 'Otildesmall',
'Odieresissmall', 'OEsmall', 'Oslashsmall', 'Ugravesmall',
'Uacutesmall', 'Ucircumflexsmall', 'Udieresissmall',
'Yacutesmall', 'Thornsmall', 'Ydieresissmall', '001.000',
'001.001', '001.002', '001.003', 'Black', 'Bold', 'Book',
'Light', 'Medium', 'Regular', 'Roman', 'Semibold',
)
class INDEX(object):
def __init__(self, fp):
self.fp = fp
self.offsets = []
(count, offsize) = struct.unpack('>HB', self.fp.read(3))
for _ in range(count+1):
self.offsets.append(nunpack(self.fp.read(offsize)))
self.base = self.fp.tell()-1
self.fp.seek(self.base+self.offsets[-1])
return
def __repr__(self):
return '<INDEX: size=%d>' % len(self)
def __len__(self): # pylint: disable=E0303
return len(self.offsets)-1
def __getitem__(self, i):
self.fp.seek(self.base+self.offsets[i])
return self.fp.read(self.offsets[i+1]-self.offsets[i])
def __iter__(self):
return iter(self[i] for i in range(len(self)))
def __init__(self, name, fp):
self.name = name
self.fp = fp
# Header
(_major, _minor, hdrsize, _) = struct.unpack('BBBB', self.fp.read(4))
self.fp.read(hdrsize-4)
# Name INDEX
self.name_index = self.INDEX(self.fp)
# Top DICT INDEX
self.dict_index = self.INDEX(self.fp)
# String INDEX
self.string_index = self.INDEX(self.fp)
# Global Subr INDEX
self.subr_index = self.INDEX(self.fp)
# Top DICT DATA
self.top_dict = getdict(self.dict_index[0])
(charset_pos,) = self.top_dict.get(15, [0])
(encoding_pos,) = self.top_dict.get(16, [0])
(charstring_pos,) = self.top_dict.get(17, [0])
# CharStrings
self.fp.seek(charstring_pos)
self.charstring = self.INDEX(self.fp)
self.nglyphs = len(self.charstring)
# Encodings
self.code2gid = {}
self.gid2code = {}
self.fp.seek(encoding_pos)
format = self.fp.read(1)
if format == b'\x00':
# Format 0
(n,) = struct.unpack('B', self.fp.read(1))
for (code, gid) in enumerate(struct.unpack('B'*n, self.fp.read(n))):
self.code2gid[code] = gid
self.gid2code[gid] = code
elif format == b'\x01':
# Format 1
(n,) = struct.unpack('B', self.fp.read(1))
code = 0
for _ in range(n):
(first, nleft) = struct.unpack('BB', self.fp.read(2))
for gid in range(first, first+nleft+1):
self.code2gid[code] = gid
self.gid2code[gid] = code
code += 1
else:
raise ValueError('unsupported encoding format: %r' % format)
# Charsets
self.name2gid = {}
self.gid2name = {}
self.fp.seek(charset_pos)
format = self.fp.read(1)
if format == b'\x00':
# Format 0
n = self.nglyphs-1
for (gid, sid) in enumerate(struct.unpack('>'+'H'*n, self.fp.read(2*n))):
gid += 1
name = self.getstr(sid)
self.name2gid[name] = gid
self.gid2name[gid] = name
elif format == b'\x01':
# Format 1
(n,) = struct.unpack('B', self.fp.read(1))
sid = 0
for _ in range(n):
(first, nleft) = struct.unpack('BB', self.fp.read(2))
for gid in range(first, first+nleft+1):
name = self.getstr(sid)
self.name2gid[name] = gid
self.gid2name[gid] = name
sid += 1
elif format == b'\x02':
# Format 2
assert False, str(('Unhandled', format))
else:
raise ValueError('unsupported charset format: %r' % format)
#print self.code2gid
#print self.name2gid
#assert 0
return
def getstr(self, sid):
if sid < len(self.STANDARD_STRINGS):
return self.STANDARD_STRINGS[sid]
return self.string_index[sid-len(self.STANDARD_STRINGS)]
## TrueTypeFont
##
class TrueTypeFont(object):
class CMapNotFound(Exception):
pass
def __init__(self, name, fp):
self.name = name
self.fp = fp
self.tables = {}
self.fonttype = fp.read(4)
try:
(ntables, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8))
for _ in range(ntables):
(name, _, offset, length) = struct.unpack('>4sLLL', fp.read(16))
self.tables[name] = (offset, length)
except struct.error:
# Do not fail if there are not enough bytes to read. Even for
# corrupted PDFs we would like to get as much information as
# possible, so continue.
pass
return
def create_unicode_map(self):
if 'cmap' not in self.tables:
raise TrueTypeFont.CMapNotFound
(base_offset, _) = self.tables['cmap']
fp = self.fp
fp.seek(base_offset)
(_, nsubtables) = struct.unpack('>HH', fp.read(4))
subtables = []
for i in range(nsubtables):
subtables.append(struct.unpack('>HHL', fp.read(8)))
char2gid = {}
# Only supports subtable type 0, 2 and 4.
for (_1, _2, st_offset) in subtables:
fp.seek(base_offset+st_offset)
(fmttype, _, _) = struct.unpack('>HHH', fp.read(6))
if fmttype == 0:
char2gid.update(enumerate(struct.unpack('>256B', fp.read(256))))
elif fmttype == 2:
subheaderkeys = struct.unpack('>256H', fp.read(512))
firstbytes = [0]*8192
for (i, k) in enumerate(subheaderkeys):
firstbytes[k//8] = i
nhdrs = max(subheaderkeys)//8 + 1
hdrs = []
for i in range(nhdrs):
(firstcode, entcount, delta, offset) = struct.unpack('>HHhH', fp.read(8))
hdrs.append((i, firstcode, entcount, delta, fp.tell()-2+offset))
for (i, firstcode, entcount, delta, pos) in hdrs:
if not entcount:
continue
first = firstcode + (firstbytes[i] << 8)
fp.seek(pos)
for c in range(entcount):
gid = struct.unpack('>H', fp.read(2))
if gid:
gid += delta
char2gid[first+c] = gid
elif fmttype == 4:
(segcount, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8))
segcount //= 2
ecs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
fp.read(2)
scs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
idds = struct.unpack('>%dh' % segcount, fp.read(2*segcount))
pos = fp.tell()
idrs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
for (ec, sc, idd, idr) in zip(ecs, scs, idds, idrs):
if idr:
fp.seek(pos+idr)
for c in range(sc, ec+1):
char2gid[c] = (struct.unpack('>H', fp.read(2))[0] + idd) & 0xffff
else:
for c in range(sc, ec+1):
char2gid[c] = (c + idd) & 0xffff
else:
assert False, str(('Unhandled', fmttype))
# create unicode map
unicode_map = FileUnicodeMap()
for (char, gid) in char2gid.items():
unicode_map.add_cid2unichr(gid, char)
return unicode_map
## Fonts
##
class PDFFontError(PDFException):
pass
class PDFUnicodeNotDefined(PDFFontError):
pass
LITERAL_STANDARD_ENCODING = LIT('StandardEncoding')
LITERAL_TYPE1C = LIT('Type1C')
# PDFFont
class PDFFont(object):
def __init__(self, descriptor, widths, default_width=None):
self.descriptor = descriptor
self.widths = widths
self.fontname = resolve1(descriptor.get('FontName', 'unknown'))
if isinstance(self.fontname, PSLiteral):
self.fontname = literal_name(self.fontname)
self.flags = int_value(descriptor.get('Flags', 0))
self.ascent = num_value(descriptor.get('Ascent', 0))
self.descent = num_value(descriptor.get('Descent', 0))
self.italic_angle = num_value(descriptor.get('ItalicAngle', 0))
self.default_width = default_width or num_value(descriptor.get('MissingWidth', 0))
self.leading = num_value(descriptor.get('Leading', 0))
self.bbox = list_value(descriptor.get('FontBBox', (0, 0, 0, 0)))
self.hscale = self.vscale = .001
return
def __repr__(self):
return '<PDFFont>'
def is_vertical(self):
return False
def is_multibyte(self):
return False
def decode(self, bytes):
return bytearray(bytes) # map(ord, bytes)
def get_ascent(self):
return self.ascent * self.vscale
def get_descent(self):
return self.descent * self.vscale
def get_width(self):
w = self.bbox[2]-self.bbox[0]
if w == 0:
w = -self.default_width
return w * self.hscale
def get_height(self):
h = self.bbox[3]-self.bbox[1]
if h == 0:
h = self.ascent - self.descent
return h * self.vscale
def char_width(self, cid):
try:
return self.widths[cid] * self.hscale
except KeyError:
try:
return self.widths[self.to_unichr(cid)] * self.hscale
except (KeyError, PDFUnicodeNotDefined):
return self.default_width * self.hscale
def char_disp(self, cid):
return 0
def string_width(self, s):
return sum(self.char_width(cid) for cid in self.decode(s))
@abstractmethod
def to_unichr(self, cid):
raise NotImplementedError
@property
def font_weight(self):
if 'FontWeight' in self.descriptor.keys():
return self.descriptor['FontWeight']
elif 'bold' in self.fontname.lower():
return 'bold'
else:
return None
@property
def is_bold(self):
return self.font_weight == 'bold' or self.font_weight > 549
# PDFSimpleFont
class PDFSimpleFont(PDFFont):
def __init__(self, descriptor, widths, spec):
# Font encoding is specified either by a name of
# built-in encoding or a dictionary that describes
# the differences.
if 'Encoding' in spec:
encoding = resolve1(spec['Encoding'])
else:
encoding = LITERAL_STANDARD_ENCODING
if isinstance(encoding, dict):
name = literal_name(encoding.get('BaseEncoding', LITERAL_STANDARD_ENCODING))
diff = list_value(encoding.get('Differences', []))
self.cid2unicode = EncodingDB.get_encoding(name, diff)
else:
self.cid2unicode = EncodingDB.get_encoding(literal_name(encoding))
self.unicode_map = None
if 'ToUnicode' in spec:
strm = PDFStream.validated_stream(spec['ToUnicode'])
self.unicode_map = FileUnicodeMap()
CMapParser(self.unicode_map, BytesIO(strm.get_data())).run()
PDFFont.__init__(self, descriptor, widths)
return
def to_unichr(self, cid):
if self.unicode_map:
try:
return self.unicode_map.get_unichr(cid)
except KeyError:
pass
try:
return self.cid2unicode[cid]
except KeyError:
raise PDFUnicodeNotDefined(None, cid)
# PDFType1Font
class PDFType1Font(PDFSimpleFont):
def __init__(self, rsrcmgr, spec):
try:
self.basefont = literal_name(spec['BaseFont'])
except KeyError:
if settings.STRICT:
raise PDFFontError('BaseFont is missing')
self.basefont = 'unknown'
try:
(descriptor, widths) = FontMetricsDB.get_metrics(self.basefont)
except KeyError:
descriptor = dict_value(spec.get('FontDescriptor', {}))
firstchar = int_value(spec.get('FirstChar', 0))
#lastchar = int_value(spec.get('LastChar', 255))
widths = list_value(spec.get('Widths', [0]*256))
widths = dict((i+firstchar, w) for (i, w) in enumerate(widths))
PDFSimpleFont.__init__(self, descriptor, widths, spec)
if 'Encoding' not in spec and 'FontFile' in descriptor:
# try to recover the missing encoding info from the font file.
self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile'))
length1 = int_value(self.fontfile['Length1'])
data = self.fontfile.get_data()[:length1]
parser = Type1FontHeaderParser(BytesIO(data))
self.cid2unicode = parser.get_encoding()
return
def __repr__(self):
return '<PDFType1Font: basefont=%r>' % self.basefont
# PDFTrueTypeFont
class PDFTrueTypeFont(PDFType1Font):
def __repr__(self):
return '<PDFTrueTypeFont: basefont=%r>' % self.basefont
# PDFType3Font
class PDFType3Font(PDFSimpleFont):
def __init__(self, rsrcmgr, spec):
firstchar = int_value(spec.get('FirstChar', 0))
#lastchar = int_value(spec.get('LastChar', 0))
widths = list_value(spec.get('Widths', [0]*256))
widths = dict((i+firstchar, w) for (i, w) in enumerate(widths))
if 'FontDescriptor' in spec:
descriptor = dict_value(spec['FontDescriptor'])
else:
descriptor = {'Ascent': 0, 'Descent': 0,
'FontBBox': spec['FontBBox']}
PDFSimpleFont.__init__(self, descriptor, widths, spec)
self.matrix = tuple(list_value(spec.get('FontMatrix')))
(_, self.descent, _, self.ascent) = self.bbox
(self.hscale, self.vscale) = apply_matrix_norm(self.matrix, (1, 1))
return
def __repr__(self):
return '<PDFType3Font>'
# PDFCIDFont
class PDFCIDFont(PDFFont):
def __init__(self, rsrcmgr, spec, strict=settings.STRICT):
try:
self.basefont = literal_name(spec['BaseFont'])
except KeyError:
if strict:
raise PDFFontError('BaseFont is missing')
self.basefont = 'unknown'
self.cidsysteminfo = dict_value(spec.get('CIDSystemInfo', {}))
self.cidcoding = '%s-%s' % (resolve1(self.cidsysteminfo.get('Registry', b'unknown')).decode("latin1"),
resolve1(self.cidsysteminfo.get('Ordering', b'unknown')).decode("latin1"))
try:
name = literal_name(spec['Encoding'])
except KeyError:
if strict:
raise PDFFontError('Encoding is unspecified')
name = 'unknown'
try:
self.cmap = CMapDB.get_cmap(name)
except CMapDB.CMapNotFound as e:
if strict:
raise PDFFontError(e)
self.cmap = CMap()
try:
descriptor = dict_value(spec['FontDescriptor'])
except KeyError:
if strict:
raise PDFFontError('FontDescriptor is missing')
descriptor = {}
ttf = None
if 'FontFile2' in descriptor:
self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile2'))
ttf = TrueTypeFont(self.basefont,
BytesIO(self.fontfile.get_data()))
self.unicode_map = None
if 'ToUnicode' in spec:
strm = PDFStream.validated_stream(spec['ToUnicode'])
self.unicode_map = FileUnicodeMap()
CMapParser(self.unicode_map, BytesIO(strm.get_data())).run()
elif self.cidcoding in ('Adobe-Identity', 'Adobe-UCS'):
if ttf:
try:
self.unicode_map = ttf.create_unicode_map()
except TrueTypeFont.CMapNotFound:
pass
else:
try:
self.unicode_map = CMapDB.get_unicode_map(self.cidcoding, self.cmap.is_vertical())
except CMapDB.CMapNotFound as e:
pass
self.vertical = self.cmap.is_vertical()
if self.vertical:
# writing mode: vertical
widths = get_widths2(list_value(spec.get('W2', [])))
self.disps = dict((cid, (vx, vy)) for (cid, (_, (vx, vy))) in iter(widths.items()))
(vy, w) = spec.get('DW2', [880, -1000])
self.default_disp = (None, vy)
widths = dict((cid, w) for (cid, (w, _)) in iter(widths.items()))
default_width = w
else:
# writing mode: horizontal
self.disps = {}
self.default_disp = 0
widths = get_widths(list_value(spec.get('W', [])))
default_width = spec.get('DW', 1000)
PDFFont.__init__(self, descriptor, widths, default_width=default_width)
return
def __repr__(self):
return '<PDFCIDFont: basefont=%r, cidcoding=%r>' % (self.basefont, self.cidcoding)
def is_vertical(self):
return self.vertical
def is_multibyte(self):
return True
def decode(self, bytes):
return self.cmap.decode(bytes)
def char_disp(self, cid):
"Returns an integer for horizontal fonts, a tuple for vertical fonts."
return self.disps.get(cid, self.default_disp)
def to_unichr(self, cid):
try:
if not self.unicode_map:
raise KeyError(cid)
return self.unicode_map.get_unichr(cid)
except KeyError:
raise PDFUnicodeNotDefined(self.cidcoding, cid)
# main
def main(argv):
for fname in argv[1:]:
fp = open(fname, 'rb')
#font = TrueTypeFont(fname, fp)
font = CFFFont(fname, fp)
print (font)
fp.close()
return
if __name__ == '__main__':
sys.exit(main(sys.argv))
@@ -1,75 +0,0 @@
class PDFGraphicStateColor:
class NotSet(Exception): pass
def __init__(self):
self.__rgb = None
self.__gray = None
self.__cmyk = None
self.__custom = None
def get_color(self):
if self.__rgb is not None:
return 'rgb', self.rgb
elif self.__cmyk is not None:
return 'cmyk', self.cmyk
elif self.__gray is not None:
return 'gray', self.gray
elif self.__custom is not None:
return 'custom', self.custom
else: return None, None
@property
def is_set(self):
return self.__rgb is not None or self.__cmyk is not None or self.__gray is not None
@property
def rgb(self):
if self.__rgb is not None: return [*self.__rgb]
else: raise self.NotSet
@property
def cmyk(self):
if self.__cmyk is not None: return [*self.__cmyk]
else: raise self.NotSet
@property
def gray(self):
if self.__gray is not None: return [float(self.__gray)]
else: raise self.NotSet
@property
def custom(self):
if self.__custom is not None: return dict(self.__custom)
else: raise self.NotSet
@rgb.setter
def rgb(self, rgb):
self.__rgb = rgb
@cmyk.setter
def cmyk(self, cmyk):
self.__cmyk = cmyk
@gray.setter
def gray(self, color):
self.__gray = color
@custom.setter
def custom(self, color):
self.__custom = color
def copy(self):
cp = self.__class__()
cp.gray = self.__gray
cp.rgb = self.__rgb
cp.cmyk = self.__cmyk
cp.custom = self.__custom
return cp
def __repr__(self):
rep = "<PDFGraphicStateColor"
if self.__gray is not None: rep += f' gray={self.__gray}'
if self.__rgb is not None: rep += f' rgb=[{",".join(map(str, self.__rgb))}]'
if self.__cmyk is not None: rep += f' cmyk=[{",".join(map(str, self.__cmyk))}]'
if self.__custom is not None: rep += f' custom={self.__custom}'
return rep + '/>'
@@ -1,45 +0,0 @@
from .PDFGraphicStateColor import PDFGraphicStateColor
class PDFGraphicState(object):
def __init__(self):
self.linewidth = 0
self.linecap = None
self.linejoin = None
self.miterlimit = None
self.dash = None
self.intent = None
self.flatness = None
# stroking color
self.scolor = PDFGraphicStateColor()
# non stroking color
self.ncolor = PDFGraphicStateColor()
return
def copy(self):
obj = PDFGraphicState()
obj.linewidth = self.linewidth
obj.linecap = self.linecap
obj.linejoin = self.linejoin
obj.miterlimit = self.miterlimit
obj.dash = self.dash
obj.intent = self.intent
obj.flatness = self.flatness
obj.scolor = self.scolor.copy()
obj.ncolor = self.ncolor.copy()
return obj
def __repr__(self):
return ('<PDFGraphicState: linewidth=%r, linecap=%r, linejoin=%r, '
' miterlimit=%r, dash=%r, intent=%r, flatness=%r, '
' stroking color=%r, non stroking color=%r'
'>' %
(
self.linewidth, self.linecap, self.linejoin,
self.miterlimit, self.dash, self.intent, self.flatness,
str(self.scolor), str(self.ncolor)
)
)
-137
View File
@@ -1,137 +0,0 @@
import logging
from ..utils import settings
from .constants import LIT
from .PDFStream import PDFObjectNotFound
from .PDFStream import resolve1, int_value, list_value, dict_value
from .PDFParser import PDFParser
from .PDFDocument import PDFDocument, PDFTextExtractionNotAllowed
from ..utils import get_logger
log = get_logger(__name__)
# some predefined literals and keywords.
LITERAL_PAGE = LIT('Page')
LITERAL_PAGES = LIT('Pages')
## PDFPage
##
class PDFPage(object):
"""An object that holds the information about a page.
A PDFPage object is merely a convenience class that has a set
of keys and values, which describe the properties of a page
and point to its contents.
Attributes:
doc: a PDFDocument object.
pageid: any Python object that can uniquely identify the page.
attrs: a dictionary of page attributes.
contents: a list of PDFStream objects that represents the page content.
lastmod: the last modified time of the page.
resources: a list of resources used by the page.
mediabox: the physical size of the page.
cropbox: the crop rectangle of the page.
rotate: the page rotation (in degree).
annots: the page annotations.
beads: a chain that represents natural reading order.
"""
def __init__(self, doc: PDFDocument, pageid, attrs):
"""Initialize a page object.
doc: a PDFDocument object.
pageid: any Python object that can uniquely identify the page.
attrs: a dictionary of page attributes.
"""
self.doc = doc
self.pageid = pageid
self.attrs = dict_value(attrs)
self.lastmod = resolve1(self.attrs.get('LastModified'))
self.resources = resolve1(self.attrs.get('Resources', dict()))
self.mediabox = resolve1(self.attrs['MediaBox'])
if 'CropBox' in self.attrs:
self.cropbox = resolve1(self.attrs['CropBox'])
else:
self.cropbox = self.mediabox
self.rotate = (int_value(self.attrs.get('Rotate', 0))+360) % 360
self.annots = self.attrs.get('Annots')
self.beads = self.attrs.get('B')
if 'Contents' in self.attrs:
contents = resolve1(self.attrs['Contents'])
else:
contents = []
if not isinstance(contents, list):
contents = [contents]
self.contents = contents
return
def __repr__(self):
return '<PDFPage: Resources=%r, MediaBox=%r>' % (self.resources, self.mediabox)
INHERITABLE_ATTRS = set(['Resources', 'MediaBox', 'CropBox', 'Rotate'])
@classmethod
def create_pages(cls, document: PDFDocument):
def search(obj, parent):
if isinstance(obj, int):
objid = obj
tree = dict_value(document.getobj(objid)).copy()
else:
objid = obj.objid
tree = dict_value(obj).copy()
for (k, v) in parent.items():
if k in cls.INHERITABLE_ATTRS and k not in tree:
tree[k] = v
tree_type = tree.get('Type')
if tree_type is None and not settings.STRICT: # See #64
tree_type = tree.get('type')
if tree_type is LITERAL_PAGES and 'Kids' in tree:
log.debug('Pages: Kids=%r', tree['Kids'])
for c in list_value(tree['Kids']):
for x in search(c, tree):
yield x
elif tree_type is LITERAL_PAGE:
log.debug('Page: %r', tree)
yield (objid, tree)
pages = False
if 'Pages' in document.catalog:
for (objid, tree) in search(document.catalog['Pages'], document.catalog):
yield cls(document, objid, tree)
pages = True
if not pages:
# fallback when /Pages is missing.
for xref in document.xrefs:
for objid in xref.get_objids():
try:
obj = document.getobj(objid)
if isinstance(obj, dict) and obj.get('Type') is LITERAL_PAGE:
yield cls(document, objid, obj)
except PDFObjectNotFound:
pass
return
@classmethod
def get_pages(cls, fp,
pagenos=None, maxpages=0, password='',
caching=True, check_extractable=True):
# Create a PDF parser object associated with the file object.
parser = PDFParser(fp)
# Create a PDF document object that stores the document structure.
doc = PDFDocument(parser, password=password, caching=caching)
# Check if the document allows text extraction. If not, abort.
if check_extractable and not doc.is_extractable:
raise PDFTextExtractionNotAllowed('Text extraction is not allowed: %r' % fp)
# Process each page contained in the document.
for (pageno, page) in enumerate(cls.create_pages(doc)):
if pagenos and (pageno not in pagenos):
continue
yield page
if maxpages and maxpages <= pageno+1:
break
return
-668
View File
@@ -1,668 +0,0 @@
import logging
import operator
from collections import OrderedDict
from typing import List
from .PDFColorSpace import PDFColorSpace
from .PDFColorSpace import PREDEFINED_COLORSPACE
from .cmapdb import CMapDB
from .cmapdb import CMap
from .PDFStream import PDFException
from .PDFStream.PDFObjRef import PDFObjRef
from .PDFStream import resolve1, list_value, dict_value
from .PDFStream import PDFStream
from .PSStackParser import PSEOF
from .PSStackParser import PSKeyword
from .PSStackParser import literal_name
from .PSStackParser import keyword_name
from .PSStackParser import PSStackParser
from .PSStackParser import LIT
from .PDFGraphicState import PDFGraphicState
from .PDFTextState import PDFTextState
from .PDFContentParser import PDFContentParser
from ..utils import mult_matrix, MATRIX_IDENTITY
from .constants import LITERAL_FORM, LITERAL_IMAGE
from .types import CurvePoint, CurvePath
log = logging.getLogger(__name__)
## Exceptions
##
class PDFInterpreterError(PDFException):
pass
class PDFPageInterpreter(object):
def __init__(self, rsrcmgr, device):
self.rsrcmgr = rsrcmgr
self.device = device
self.curpath: List[CurvePath] = []
return
def dup(self):
return self.__class__(self.rsrcmgr, self.device)
# init_resources(resources):
# Prepare the fonts and XObjects listed in the Resource attribute.
def init_resources(self, resources):
self.resources = resources
self.fontmap = {}
self.xobjmap = {}
self.csmap:OrderedDict = PREDEFINED_COLORSPACE.copy()
if not resources:
return
def get_colorspace(spec):
if isinstance(spec, list):
name = literal_name(spec[0])
else:
name = literal_name(spec)
if name == 'ICCBased' and isinstance(spec, list) and 2 <= len(spec):
return PDFColorSpace(name, PDFStream.validated_stream(spec[1])['N'])
elif name == 'DeviceN' and isinstance(spec, list) and 2 <= len(spec):
return PDFColorSpace(name, len(list_value(spec[1])))
else:
return PREDEFINED_COLORSPACE.get(name)
for (k, v) in iter(dict_value(resources).items()):
log.debug('Resource: %r: %r', k, v)
if k == 'Font':
for (fontid, spec) in iter(dict_value(v).items()):
objid = None
if isinstance(spec, PDFObjRef):
objid = spec.objid
spec = dict_value(spec)
self.fontmap[fontid] = self.rsrcmgr.get_font(objid, spec)
elif k == 'ColorSpace':
for (csid, spec) in iter(dict_value(v).items()):
self.csmap[csid] = get_colorspace(resolve1(spec))
elif k == 'ProcSet':
self.rsrcmgr.get_procset(list_value(v))
elif k == 'XObject':
for (xobjid, xobjstrm) in iter(dict_value(v).items()):
self.xobjmap[xobjid] = xobjstrm
return
# init_state(ctm)
# Initialize the text and graphic states for rendering a page.
def init_state(self, ctm):
# gstack: stack for graphical states.
self.gstack = []
self.ctm = ctm
self.device.set_ctm(self.ctm)
self.textstate = PDFTextState()
self.graphicstate = PDFGraphicState()
self.curpath = []
# argstack: stack for command arguments.
self.argstack = []
# set some global states.
self.scs = self.ncs = None
if self.csmap:
self.scs = self.ncs = next(iter(self.csmap.values()))
return
def push(self, obj):
self.argstack.append(obj)
return
def pop(self, n):
if n == 0:
return []
x = self.argstack[-n:]
self.argstack = self.argstack[:-n]
return x
def get_current_state(self):
return (self.ctm, self.textstate.copy(), self.graphicstate.copy())
def set_current_state(self, state):
(self.ctm, self.textstate, self.graphicstate) = state
self.device.set_ctm(self.ctm)
return
# gsave
def do_q(self):
self.gstack.append(self.get_current_state())
return
# grestore
def do_Q(self):
if self.gstack:
self.set_current_state(self.gstack.pop())
return
# concat-matrix
def do_cm(self, a1, b1, c1, d1, e1, f1):
self.ctm = mult_matrix((a1, b1, c1, d1, e1, f1), self.ctm)
self.device.set_ctm(self.ctm)
return
# setlinewidth
def do_w(self, linewidth):
self.graphicstate.linewidth = linewidth
return
# setlinecap
def do_J(self, linecap):
self.graphicstate.linecap = linecap
return
# setlinejoin
def do_j(self, linejoin):
self.graphicstate.linejoin = linejoin
return
# setmiterlimit
def do_M(self, miterlimit):
self.graphicstate.miterlimit = miterlimit
return
# setdash
def do_d(self, dash, phase):
self.graphicstate.dash = (dash, phase)
return
# setintent
def do_ri(self, intent):
self.graphicstate.intent = intent
return
# setflatness
def do_i(self, flatness):
self.graphicstate.flatness = flatness
return
# load-gstate
def do_gs(self, name):
#XXX
return
# moveto
def do_m(self, x: float, y: float):
self.curpath.append(CurvePath(CurvePath.METHOD.MOVE_TO, CurvePoint(x, y)))
return
# lineto
def do_l(self, x: float, y: float):
self.curpath.append(CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x, y)))
return
# curveto
def do_c(self, x1: float, y1: float, x2: float, y2: float, x3: float, y3: float):
self.curpath.append(CurvePath(
CurvePath.METHOD.CURVE_BOTH_TO,
CurvePoint(x1, y1),
CurvePoint(x2, y2),
CurvePoint(x3, y3),
))
return
# urveto
def do_v(self, x2: float, y2: float, x3: float, y3: float):
self.curpath.append(CurvePath(
CurvePath.METHOD.CURVE_NEXT_TO,
CurvePoint(x2, y2),
CurvePoint(x3, y3),
))
return
# rveto
def do_y(self, x1: float, y1: float, x3: float, y3: float):
self.curpath.append(CurvePath(
CurvePath.METHOD.CURVE_FIRST_TO,
CurvePoint(x1, y1),
CurvePoint(x3, y3)
))
return
# closepath
def do_h(self):
self.curpath.append(CurvePath(CurvePath.METHOD.CLOSE_PATH))
return
# rectangle
def do_re(self, x, y, w, h):
self.curpath.append(CurvePath(CurvePath.METHOD.MOVE_TO, CurvePoint(x, y) ))
self.curpath.append(CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x+w, y) ))
self.curpath.append(CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x+w, y+h) ))
self.curpath.append(CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x, y+h) ))
self.curpath.append(CurvePath(CurvePath.METHOD.CLOSE_PATH))
return
# stroke
def do_S(self):
self.device.paint_path(self.graphicstate.copy(), False, self.curpath)
self.curpath = []
return
# close-and-stroke
def do_s(self):
self.do_h()
self.do_S()
return
# fill
def do_f(self):
self.device.paint_path(self.graphicstate.copy(), False, self.curpath)
self.curpath = []
return
# fill (obsolete)
do_F = do_f
# fill-even-odd
def do_f_a(self):
self.device.paint_path(self.graphicstate.copy(), True, self.curpath)
self.curpath = []
return
# fill-and-stroke
def do_B(self):
self.device.paint_path(self.graphicstate.copy(), False, self.curpath)
self.curpath = []
return
# fill-and-stroke-even-odd
def do_B_a(self):
self.device.paint_path(self.graphicstate.copy(), True, self.curpath)
self.curpath = []
return
# close-fill-and-stroke
def do_b(self):
self.do_h()
self.do_B()
return
# close-fill-and-stroke-even-odd
def do_b_a(self):
self.do_h()
self.do_B_a()
return
# close-only
def do_n(self):
self.curpath = []
return
# clip
def do_W(self):
return
# clip-even-odd
def do_W_a(self):
return
# setcolorspace-stroking
def do_CS(self, name):
try:
self.scs = self.csmap[literal_name(name)]
except KeyError:
raise PDFInterpreterError('Undefined ColorSpace: %r' % name)
return
# setcolorspace-non-strokine
def do_cs(self, name):
try:
self.ncs = self.csmap[literal_name(name)]
except KeyError:
raise PDFInterpreterError('Undefined ColorSpace: %r' % name)
return
# setgray-stroking
def do_G(self, gray):
self.graphicstate.scolor.gray = gray
#self.do_CS(LITERAL_DEVICE_GRAY)
return
# setgray-non-stroking
def do_g(self, gray):
self.graphicstate.ncolor.gray = gray
#self.do_cs(LITERAL_DEVICE_GRAY)
return
# setrgb-stroking
def do_RG(self, r, g, b):
self.graphicstate.scolor.rgb = (r*255, g*255, b*255)
#self.do_CS(LITERAL_DEVICE_RGB)
return
# setrgb-non-stroking
def do_rg(self, r, g, b):
self.graphicstate.ncolor.rgb = (r*255, g*255, b*255)
#self.do_cs(LITERAL_DEVICE_RGB)
return
# setcmyk-stroking
def do_K(self, c, m, y, k):
self.graphicstate.scolor.cmyk = (c, m, y, k)
#self.do_CS(LITERAL_DEVICE_CMYK)
return
# setcmyk-non-stroking
def do_k(self, c, m, y, k):
self.graphicstate.ncolor.cmyk = (c, m, y, k)
#self.do_cs(LITERAL_DEVICE_CMYK)
return
def __process_ICCBased(self, n: int, components: list):
if n == 1:
return {
"type": "gray",
"definition": components
}
elif n == 3:
return {
"type": "rgb",
"definition": components
}
elif n == 4:
return {
"type": "cmyk",
"definition": components
}
raise PDFInterpreterError("Unsupported ICCBased color")
# setcolor
def __process_scn(self):
if self.ncs:
n = self.ncs.ncomponents
else:
raise PDFInterpreterError('No colorspace specified!')
components = self.pop(n)
obj_id = components[0]
if self.ncs.name == "ICCBased":
return self.__process_ICCBased(n, components)
if self.ncs.name not in self.resources.keys():
log.warning("Color space not in resources")
return {
"type": self.ncs.name,
"defintion": f"{n}|({','.join(map(str, components))})"
}
if obj_id.name not in self.resources[self.ncs.name].keys():
raise PDFInterpreterError("Invalid Color reference")
color_data = self.resources[self.ncs.name][obj_id.name].resolve()
if isinstance(color_data, PDFStream):
color_data = color_data.attrs
return {
"type": self.ncs.name,
"defintion": color_data
}
def do_SCN(self):
processed_col = self.__process_scn()
if processed_col['type'] == 'rgb':
self.graphicstate.scolor.rgb = processed_col['definition']
elif processed_col['type'] == 'gray':
self.graphicstate.scolor.gray = processed_col['definition'][0]
elif processed_col['type'] == 'cmyk':
self.graphicstate.scolor.cmyk = processed_col['definition']
else:
self.graphicstate.scolor.custom = processed_col
return
def do_scn(self):
processed_col = self.__process_scn()
if processed_col['type'] == 'rgb':
self.graphicstate.ncolor.rgb = processed_col['definition']
elif processed_col['type'] == 'gray':
self.graphicstate.ncolor.gray = processed_col['definition'][0]
elif processed_col['type'] == 'cmyk':
self.graphicstate.ncolor.cmyk = processed_col['definition']
else:
self.graphicstate.ncolor.custom = processed_col
return
def do_SC(self):
self.do_SCN()
return
def do_sc(self):
self.do_scn()
return
# sharing-name
def do_sh(self, name):
return
# begin-text
def do_BT(self):
self.textstate.reset()
return
# end-text
def do_ET(self):
return
# begin-compat
def do_BX(self):
return
# end-compat
def do_EX(self):
return
# marked content operators
def do_MP(self, tag):
self.device.do_tag(tag)
return
def do_DP(self, tag, props):
self.device.do_tag(tag, props)
return
def do_BMC(self, tag):
self.device.begin_tag(tag)
return
def do_BDC(self, tag, props):
self.device.begin_tag(tag, props)
return
def do_EMC(self):
self.device.end_tag()
return
# setcharspace
def do_Tc(self, space):
self.textstate.charspace = space
return
# setwordspace
def do_Tw(self, space):
self.textstate.wordspace = space
return
# def do_Th(self, )
# textscale
def do_Tz(self, scale):
self.textstate.scaling = scale
return
# setleading
def do_TL(self, leading):
self.textstate.leading = -leading
return
# selectfont
def do_Tf(self, fontid, fontsize):
try:
self.textstate.font = self.fontmap[literal_name(fontid)]
except KeyError:
raise PDFInterpreterError('Undefined Font id: %r' % fontid)
# self.textstate.font = self.rsrcmgr.get_font(None, {})
self.textstate.fontsize = fontsize
return
# setrendering
def do_Tr(self, render):
self.textstate.render = render
return
# settextrise
def do_Ts(self, rise):
self.textstate.rise = rise
return
# text-move
def do_Td(self, tx, ty):
(a, b, c, d, e, f) = self.textstate.matrix
self.textstate.matrix = (a, b, c, d, tx*a+ty*c+e, tx*b+ty*d+f)
self.textstate.linematrix = [0, 0]
return
# text-move
def do_TD(self, tx, ty):
(a, b, c, d, e, f) = self.textstate.matrix
self.textstate.matrix = (a, b, c, d, tx*a+ty*c+e, tx*b+ty*d+f)
self.textstate.leading = ty
self.textstate.linematrix = [0, 0]
return
# textmatrix
def do_Tm(self, a, b, c, d, e, f):
self.textstate.matrix = (a, b, c, d, e, f)
self.textstate.linematrix = [0, 0]
return
# nextline
def do_T_a(self):
(a, b, c, d, e, f) = self.textstate.matrix
self.textstate.matrix = (a, b, c, d, self.textstate.leading*c+e, self.textstate.leading*d+f)
self.textstate.linematrix = [0, 0]
return
# show-pos
def do_TJ(self, seq: bytearray):
if self.textstate.font is None:
raise PDFInterpreterError('No font specified!')
self.device.render_string(self.textstate.copy(), seq, self.ncs, self.graphicstate.copy())
return
# show
def do_Tj(self, s):
self.do_TJ([s])
return
# quote
def do__q(self, s):
self.do_T_a()
self.do_TJ([s])
return
# doublequote
def do__w(self, aw, ac, s):
self.do_Tw(aw)
self.do_Tc(ac)
self.do_TJ([s])
return
# inline image
def do_BI(self): # never called
return
def do_ID(self): # never called
return
def do_EI(self, obj: PDFStream):
if 'W' in obj and 'H' in obj:
iobjid = str(id(obj))
self.device.begin_figure(iobjid, (0, 0, 1, 1), MATRIX_IDENTITY)
self.device.render_image(iobjid, obj)
self.device.end_figure(iobjid)
return
# invoke an XObject
def do_Do(self, xobjid):
xobjid = literal_name(xobjid)
try:
xobj = PDFStream.validated_stream(self.xobjmap[xobjid])
except KeyError:
raise PDFInterpreterError('Undefined xobject id: %r' % xobjid)
log.info('Processing xobj: %r', xobj)
subtype = xobj.get('Subtype')
if subtype is LITERAL_FORM and 'BBox' in xobj:
interpreter = self.dup()
bbox = list_value(xobj['BBox'])
matrix = list_value(xobj.get('Matrix', MATRIX_IDENTITY))
# According to PDF reference 1.7 section 4.9.1, XObjects in
# earlier PDFs (prior to v1.2) use the page's Resources entry
# instead of having their own Resources entry.
xobjres = xobj.get('Resources')
resources = dict_value(xobjres) if xobjres else self.resources.copy()
self.device.begin_figure(xobjid, bbox, matrix)
interpreter.render_contents(resources, [xobj], ctm=mult_matrix(matrix, self.ctm))
self.device.end_figure(xobjid)
elif subtype is LITERAL_IMAGE and 'Width' in xobj and 'Height' in xobj:
self.device.begin_figure(xobjid, (0, 0, 1, 1), MATRIX_IDENTITY)
self.device.render_image(xobjid, xobj)
self.device.end_figure(xobjid)
else:
# unsupported xobject type.
pass
return
def process_page(self, page):
log.info('Processing page: %r', page)
(x0, y0, x1, y1) = page.mediabox
if page.rotate == 90:
ctm = (0, -1, 1, 0, -y0, x1)
elif page.rotate == 180:
ctm = (-1, 0, 0, -1, x1, y1)
elif page.rotate == 270:
ctm = (0, 1, -1, 0, y1, -x0)
else:
ctm = (1, 0, 0, 1, -x0, -y0)
self.device.begin_page(page, ctm)
self.render_contents(page.resources, page.contents, ctm=ctm)
return self.device.end_page(page)
# render_contents(resources, streams, ctm)
# Render the content streams.
# This method may be called recursively.
def render_contents(self, resources, streams, ctm=MATRIX_IDENTITY):
log.info('render_contents: resources=%r, streams=%r, ctm=%r',
resources, streams, ctm)
self.init_resources(resources)
self.init_state(ctm)
self.execute(list_value(streams))
return
def execute(self, streams):
try:
parser = PDFContentParser(streams)
except PSEOF:
# empty page
return
while True:
try:
(_, obj) = parser.nextobject()
except PSEOF:
break
if isinstance(obj, PSKeyword):
name = keyword_name(obj)
method = 'do_%s' % name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
if hasattr(self, method):
func = getattr(self, method)
nargs = func.__code__.co_argcount-1
if nargs:
args = self.pop(nargs)
if len(args) == nargs:
func(*args)
else:
log.debug('exec: %s', name)
func()
else:
raise PDFInterpreterError('Unknown operator: %r' % name)
else:
self.push(obj)
return
-178
View File
@@ -1,178 +0,0 @@
import logging
from io import BytesIO, TextIOWrapper
from ..utils import settings
from .PSStackParser import PSStackParser
from .PSStackParser import PSSyntaxError
from .PSStackParser import PSEOF
from .PSStackParser import KWD
from .PDFStream import PDFException
from .PDFStream import PDFStream
from .PDFStream.PDFObjRef import PDFObjRef
from .PDFStream import int_value
from .PDFStream import dict_value
log = logging.getLogger(__name__)
## Exceptions
##
class PDFSyntaxError(PDFException):
pass
## PDFParser
##
class PDFParser(PSStackParser):
"""
PDFParser fetch PDF objects from a file stream.
It can handle indirect references by referring to
a PDF document set by set_document method.
It also reads XRefs at the end of every PDF file.
Typical usage:
parser = PDFParser(fp)
parser.read_xref()
parser.read_xref(fallback=True) # optional
parser.set_document(doc)
parser.seek(offset)
parser.nextobject()
"""
def __init__(self, fp: TextIOWrapper):
PSStackParser.__init__(self, fp)
self.doc = None
self.fallback = False
return
def set_document(self, doc):
"""Associates the parser with a PDFDocument object."""
self.doc = doc
return
KEYWORD_R = KWD(b'R')
KEYWORD_NULL = KWD(b'null')
KEYWORD_ENDOBJ = KWD(b'endobj')
KEYWORD_STREAM = KWD(b'stream')
KEYWORD_XREF = KWD(b'xref')
KEYWORD_STARTXREF = KWD(b'startxref')
def do_keyword(self, pos, token):
"""Handles PDF-related keywords."""
if token in (self.KEYWORD_XREF, self.KEYWORD_STARTXREF):
self.add_results(*self.pop(1))
elif token is self.KEYWORD_ENDOBJ:
self.add_results(*self.pop(4))
elif token is self.KEYWORD_NULL:
# null object
self.push((pos, None))
elif token is self.KEYWORD_R:
# reference to indirect object
try:
((_, objid), (_, genno)) = self.pop(2) # pylint: disable=E0632
(objid, genno) = (int(objid), int(genno)) # pylint: disable=E0632
obj = PDFObjRef(self.doc, objid, genno) # pylint: disable=E0632
self.push((pos, obj))
except PSSyntaxError:
pass
elif token is self.KEYWORD_STREAM:
# stream object
((_, dic),) = self.pop(1) # pylint: disable=E0632
dic = dict_value(dic)
objlen = 0
if not self.fallback:
try:
objlen = int_value(dic['Length'])
except KeyError:
if settings.STRICT:
raise PDFSyntaxError('/Length is undefined: %r' % dic)
self.seek(pos)
try:
(_, line) = self.nextline() # 'stream'
except PSEOF:
if settings.STRICT:
raise PDFSyntaxError('Unexpected EOF')
return
pos += len(line)
self.fp.seek(pos)
data = bytearray(self.fp.read(objlen))
self.seek(pos+objlen)
while 1:
try:
(_, line) = self.nextline()
except PSEOF:
if settings.STRICT:
raise PDFSyntaxError('Unexpected EOF')
break
if b'endstream' in line:
i = line.index(b'endstream')
objlen += i
if self.fallback:
data += line[:i]
break
objlen += len(line)
if self.fallback:
data += line
data = bytes(data)
self.seek(pos+objlen)
# XXX limit objlen not to exceed object boundary
log.debug('Stream: pos=%d, objlen=%d, dic=%r, data=%r...', pos, objlen, dic, data[:10])
obj = PDFStream(dic, data, self.doc.decipher)
self.push((pos, obj))
else:
# others
self.push((pos, token))
return
## PDFStreamParser
##
class PDFStreamParser(PDFParser):
"""
PDFStreamParser is used to parse PDF content streams
that is contained in each page and has instructions
for rendering the page. A reference to a PDF document is
needed because a PDF content stream can also have
indirect references to other objects in the same document.
"""
def __init__(self, data):
PDFParser.__init__(self, BytesIO(data))
return
def flush(self):
self.add_results(*self.popall())
return
KEYWORD_OBJ = KWD(b'obj')
def do_keyword(self, pos, token):
if token is self.KEYWORD_R:
# reference to indirect object
try:
((_, objid), (_, genno)) = self.pop(2) # pylint: disable=E0632
(objid, genno) = (int(objid), int(genno))
obj = PDFObjRef(self.doc, objid, genno)
self.push((pos, obj))
except PSSyntaxError:
pass
return
elif token in (self.KEYWORD_OBJ, self.KEYWORD_ENDOBJ):
if settings.STRICT:
# See PDF Spec 3.4.6: Only the object values are stored in the
# stream; the obj and endobj keywords are not used.
raise PDFSyntaxError('Keyword endobj found in stream')
return
# others
self.push((pos, token))
return
-103
View File
@@ -1,103 +0,0 @@
import logging
from .cmapdb import CMapDB, CMap
from .PDFFont import PDFFontError
from .PDFFont import PDFType1Font
from .PDFFont import PDFTrueTypeFont
from .PDFFont import PDFType3Font
from .PDFFont import PDFCIDFont
from .PDFStream import PDFException
from .PDFStream import resolve1, dict_value, list_value
from .constants import LITERAL_PDF, LITERAL_TEXT, LITERAL_FONT
from .PSStackParser import literal_name, settings
log = logging.getLogger(__name__)
## Exceptions
##
class PDFResourceError(PDFException):
pass
class PDFInterpreterError(PDFException):
pass
## Resource Manager
##
class PDFResourceManager(object):
"""Repository of shared resources.
ResourceManager facilitates reuse of shared resources
such as fonts and images so that large objects are not
allocated multiple times.
"""
def __init__(self, caching=True):
self.caching = caching
self._cached_fonts = {}
return
def get_procset(self, procs):
for proc in procs:
if proc is LITERAL_PDF:
pass
elif proc is LITERAL_TEXT:
pass
else:
#raise PDFResourceError('ProcSet %r is not supported.' % proc)
pass
return
def get_cmap(self, cmapname, strict=False):
try:
return CMapDB.get_cmap(cmapname)
except CMapDB.CMapNotFound:
if strict:
raise
return CMap()
def get_font(self, objid: int, spec: dict):
if objid and objid in self._cached_fonts:
font = self._cached_fonts[objid]
else:
log.info('get_font: create: objid=%r, spec=%r', objid, spec)
if settings.STRICT:
if spec['Type'] is not LITERAL_FONT:
raise PDFFontError('Type is not /Font')
# Create a Font object.
if 'Subtype' in spec:
subtype = literal_name(spec['Subtype'])
else:
if settings.STRICT:
raise PDFFontError('Font Subtype is not specified.')
subtype = 'Type1'
if subtype in ('Type1', 'MMType1'):
# Type1 Font
font = PDFType1Font(self, spec)
elif subtype == 'TrueType':
# TrueType Font
font = PDFTrueTypeFont(self, spec)
elif subtype == 'Type3':
# Type3 Font
font = PDFType3Font(self, spec)
elif subtype in ('CIDFontType0', 'CIDFontType2'):
# CID Font
font = PDFCIDFont(self, spec)
elif subtype == 'Type0':
# Type0 Font
dfonts = list_value(spec['DescendantFonts'])
assert dfonts
subspec = dict_value(dfonts[0]).copy()
for k in ('Encoding', 'ToUnicode'):
if k in spec:
subspec[k] = resolve1(spec[k])
font = self.get_font(None, subspec)
else:
if settings.STRICT:
raise PDFFontError('Invalid Font spec: %r' % spec)
font = PDFType1Font(self, spec) # this is so wrong!
if objid and self.caching:
self._cached_fonts[objid] = font
return font
@@ -1,32 +0,0 @@
import zlib
from ..PSStackParser import PSException
from ..PSStackParser import PSObject
from ...utils import lzwdecode, settings
from .constants import * # pylint: disable=W0614
from .types import * # pylint: disable=W0614
from .exceptions import * # pylint: disable=W0614
## PDFObjRef
##
class PDFObjRef(PDFObject):
def __init__(self, doc, objid, a):
if objid == 0:
if settings.STRICT:
raise PDFValueError('PDF object id cannot be 0.')
self.doc = doc
self.objid = objid
#self.genno = genno # Never used.
return
def __repr__(self):
return '<PDFObjRef:%d>' % (self.objid)
def resolve(self, default=None):
try:
return self.doc.getobj(self.objid)
except PDFObjectNotFound:
return default
-144
View File
@@ -1,144 +0,0 @@
import zlib
from ...utils import lzwdecode
from ...utils import ascii85decode, asciihexdecode, apply_png_predictor
from .constants import *
from .types import *
from .exceptions import *
from .util import *
from .ccitt import ccittfaxdecode
class PDFStream(PDFObject):
@classmethod
def validated_stream(cls, x):
x = resolve1(x)
if not isinstance(x, cls):
if settings.STRICT:
raise PDFTypeError('PDFStream required: %r' % x)
return PDFStream({}, b'')
return x
def __init__(self, attrs, rawdata, decipher=None):
assert isinstance(attrs, dict), str(type(attrs))
self.attrs = attrs
self.rawdata = rawdata
self.decipher = decipher
self.data = None
self.objid = None
self.genno = None
return
def set_objid(self, objid, genno):
self.objid = objid
self.genno = genno
return
def __repr__(self):
if self.data is None:
assert self.rawdata is not None
return '<PDFStream(%r): raw=%d, %r>' % (self.objid, len(self.rawdata), self.attrs)
else:
assert self.data is not None
return '<PDFStream(%r): len=%d, %r>' % (self.objid, len(self.data), self.attrs)
def __contains__(self, name):
return name in self.attrs
def __getitem__(self, name):
return self.attrs[name]
def get(self, name, default=None):
return self.attrs.get(name, default)
def get_any(self, names, default=None):
for name in names:
if name in self.attrs:
return self.attrs[name]
return default
def get_filters(self):
filters = self.get_any(('F', 'Filter'))
params = self.get_any(('DP', 'DecodeParms', 'FDecodeParms'), {})
if not filters:
return []
if not isinstance(filters, list):
filters = [filters]
if not isinstance(params, list):
# Make sure the parameters list is the same as filters.
params = [params] * len(filters)
if settings.STRICT and len(params) != len(filters):
raise PDFException("Parameters len filter mismatch")
# resolve filter if possible
_filters = []
for fltr in filters:
if hasattr(fltr, 'resolve'):
fltr = fltr.resolve()[0]
_filters.append(fltr)
return list(zip(_filters, params)) #solves https://github.com/pdfminer/pdfminer.six/issues/15
def decode(self):
assert self.data is None and self.rawdata is not None, str((self.data, self.rawdata))
data = self.rawdata
if self.decipher:
# Handle encryption
data = self.decipher(self.objid, self.genno, data, self.attrs)
filters = self.get_filters()
if not filters:
self.data = data
self.rawdata = None
return
for (f,params) in filters:
if f in LITERALS_FLATE_DECODE:
# will get errors if the document is encrypted.
try:
data = zlib.decompress(data)
except zlib.error as e:
if settings.STRICT:
raise PDFException('Invalid zlib bytes: %r, %r' % (e, data))
data = b''
elif f in LITERALS_LZW_DECODE:
data = lzwdecode(data)
elif f in LITERALS_ASCII85_DECODE:
data = ascii85decode(data)
elif f in LITERALS_ASCIIHEX_DECODE:
data = asciihexdecode(data)
elif f in LITERALS_RUNLENGTH_DECODE:
data = rldecode(data)
elif f in LITERALS_CCITTFAX_DECODE:
data = ccittfaxdecode(data, params)
elif f in LITERALS_DCT_DECODE:
# This is probably a JPG stream - it does not need to be decoded twice.
# Just return the stream to the user.
pass
elif f == LITERAL_CRYPT:
# not yet..
raise PDFNotImplementedError('/Crypt filter is unsupported')
else:
raise PDFNotImplementedError('Unsupported filter: %r' % f)
# apply predictors
if params and 'Predictor' in params:
pred = int_value(params['Predictor'])
if pred == 1:
# no predictor
pass
elif 10 <= pred:
# PNG predictor
colors = int_value(params.get('Colors', 1))
columns = int_value(params.get('Columns', 1))
bitspercomponent = int_value(params.get('BitsPerComponent', 8))
data = apply_png_predictor(pred, colors, columns, bitspercomponent, data)
else:
raise PDFNotImplementedError('Unsupported predictor: %r' % pred)
self.data = data
self.rawdata = None
return
def get_data(self):
if self.data is None:
self.decode()
return self.data
def get_rawdata(self):
return self.rawdata
-604
View File
@@ -1,604 +0,0 @@
# CCITT Fax decoder
#
# Bugs: uncompressed mode untested.
#
# cf.
# ITU-T Recommendation T.4
# "Standardization of Group 3 facsimile terminals for document transmission"
# ITU-T Recommendation T.6
# "FACSIMILE CODING SCHEMES AND CODING CONTROL FUNCTIONS FOR GROUP 4 FACSIMILE APPARATUS"
import sys
import array
from abc import abstractmethod
def get_bytes(data):
for byte in data:
yield byte
## BitParser
##
class BitParser(object):
def __init__(self):
self._pos = 0
self._state = None
return
@classmethod
def add(klass, root, v, bits):
p = root
b = None
for i in range(len(bits)):
if 0 < i:
if p[b] is None:
p[b] = [None, None]
p = p[b]
if bits[i] == '1':
b = 1
else:
b = 0
p[b] = v
return
def feedbytes(self, data):
for byte in get_bytes(data):
for m in (128, 64, 32, 16, 8, 4, 2, 1):
self._parse_bit(byte & m)
return
def _parse_bit(self, x):
if x:
v = self._state[1]
else:
v = self._state[0]
self._pos += 1
if isinstance(v, list):
self._state = v
else:
self._state = self._accept(v)
return
@abstractmethod
def _accept(self, v):
raise NotImplementedError
## CCITTG4Parser
##
class CCITTG4Parser(BitParser):
MODE = [None, None]
BitParser.add(MODE, 0, '1')
BitParser.add(MODE, +1, '011')
BitParser.add(MODE, -1, '010')
BitParser.add(MODE, 'h', '001')
BitParser.add(MODE, 'p', '0001')
BitParser.add(MODE, +2, '000011')
BitParser.add(MODE, -2, '000010')
BitParser.add(MODE, +3, '0000011')
BitParser.add(MODE, -3, '0000010')
BitParser.add(MODE, 'u', '0000001111')
BitParser.add(MODE, 'x1', '0000001000')
BitParser.add(MODE, 'x2', '0000001001')
BitParser.add(MODE, 'x3', '0000001010')
BitParser.add(MODE, 'x4', '0000001011')
BitParser.add(MODE, 'x5', '0000001100')
BitParser.add(MODE, 'x6', '0000001101')
BitParser.add(MODE, 'x7', '0000001110')
BitParser.add(MODE, 'e', '000000000001000000000001')
WHITE = [None, None]
BitParser.add(WHITE, 0 , '00110101')
BitParser.add(WHITE, 1 , '000111')
BitParser.add(WHITE, 2 , '0111')
BitParser.add(WHITE, 3 , '1000')
BitParser.add(WHITE, 4 , '1011')
BitParser.add(WHITE, 5 , '1100')
BitParser.add(WHITE, 6 , '1110')
BitParser.add(WHITE, 7 , '1111')
BitParser.add(WHITE, 8 , '10011')
BitParser.add(WHITE, 9 , '10100')
BitParser.add(WHITE, 10 , '00111')
BitParser.add(WHITE, 11 , '01000')
BitParser.add(WHITE, 12 , '001000')
BitParser.add(WHITE, 13 , '000011')
BitParser.add(WHITE, 14 , '110100')
BitParser.add(WHITE, 15 , '110101')
BitParser.add(WHITE, 16 , '101010')
BitParser.add(WHITE, 17 , '101011')
BitParser.add(WHITE, 18 , '0100111')
BitParser.add(WHITE, 19 , '0001100')
BitParser.add(WHITE, 20 , '0001000')
BitParser.add(WHITE, 21 , '0010111')
BitParser.add(WHITE, 22 , '0000011')
BitParser.add(WHITE, 23 , '0000100')
BitParser.add(WHITE, 24 , '0101000')
BitParser.add(WHITE, 25 , '0101011')
BitParser.add(WHITE, 26 , '0010011')
BitParser.add(WHITE, 27 , '0100100')
BitParser.add(WHITE, 28 , '0011000')
BitParser.add(WHITE, 29 , '00000010')
BitParser.add(WHITE, 30 , '00000011')
BitParser.add(WHITE, 31 , '00011010')
BitParser.add(WHITE, 32 , '00011011')
BitParser.add(WHITE, 33 , '00010010')
BitParser.add(WHITE, 34 , '00010011')
BitParser.add(WHITE, 35 , '00010100')
BitParser.add(WHITE, 36 , '00010101')
BitParser.add(WHITE, 37 , '00010110')
BitParser.add(WHITE, 38 , '00010111')
BitParser.add(WHITE, 39 , '00101000')
BitParser.add(WHITE, 40 , '00101001')
BitParser.add(WHITE, 41 , '00101010')
BitParser.add(WHITE, 42 , '00101011')
BitParser.add(WHITE, 43 , '00101100')
BitParser.add(WHITE, 44 , '00101101')
BitParser.add(WHITE, 45 , '00000100')
BitParser.add(WHITE, 46 , '00000101')
BitParser.add(WHITE, 47 , '00001010')
BitParser.add(WHITE, 48 , '00001011')
BitParser.add(WHITE, 49 , '01010010')
BitParser.add(WHITE, 50 , '01010011')
BitParser.add(WHITE, 51 , '01010100')
BitParser.add(WHITE, 52 , '01010101')
BitParser.add(WHITE, 53 , '00100100')
BitParser.add(WHITE, 54 , '00100101')
BitParser.add(WHITE, 55 , '01011000')
BitParser.add(WHITE, 56 , '01011001')
BitParser.add(WHITE, 57 , '01011010')
BitParser.add(WHITE, 58 , '01011011')
BitParser.add(WHITE, 59 , '01001010')
BitParser.add(WHITE, 60 , '01001011')
BitParser.add(WHITE, 61 , '00110010')
BitParser.add(WHITE, 62 , '00110011')
BitParser.add(WHITE, 63 , '00110100')
BitParser.add(WHITE, 64 , '11011')
BitParser.add(WHITE, 128 , '10010')
BitParser.add(WHITE, 192 , '010111')
BitParser.add(WHITE, 256 , '0110111')
BitParser.add(WHITE, 320 , '00110110')
BitParser.add(WHITE, 384 , '00110111')
BitParser.add(WHITE, 448 , '01100100')
BitParser.add(WHITE, 512 , '01100101')
BitParser.add(WHITE, 576 , '01101000')
BitParser.add(WHITE, 640 , '01100111')
BitParser.add(WHITE, 704 , '011001100')
BitParser.add(WHITE, 768 , '011001101')
BitParser.add(WHITE, 832 , '011010010')
BitParser.add(WHITE, 896 , '011010011')
BitParser.add(WHITE, 960 , '011010100')
BitParser.add(WHITE, 1024, '011010101')
BitParser.add(WHITE, 1088, '011010110')
BitParser.add(WHITE, 1152, '011010111')
BitParser.add(WHITE, 1216, '011011000')
BitParser.add(WHITE, 1280, '011011001')
BitParser.add(WHITE, 1344, '011011010')
BitParser.add(WHITE, 1408, '011011011')
BitParser.add(WHITE, 1472, '010011000')
BitParser.add(WHITE, 1536, '010011001')
BitParser.add(WHITE, 1600, '010011010')
BitParser.add(WHITE, 1664, '011000')
BitParser.add(WHITE, 1728, '010011011')
BitParser.add(WHITE, 1792, '00000001000')
BitParser.add(WHITE, 1856, '00000001100')
BitParser.add(WHITE, 1920, '00000001101')
BitParser.add(WHITE, 1984, '000000010010')
BitParser.add(WHITE, 2048, '000000010011')
BitParser.add(WHITE, 2112, '000000010100')
BitParser.add(WHITE, 2176, '000000010101')
BitParser.add(WHITE, 2240, '000000010110')
BitParser.add(WHITE, 2304, '000000010111')
BitParser.add(WHITE, 2368, '000000011100')
BitParser.add(WHITE, 2432, '000000011101')
BitParser.add(WHITE, 2496, '000000011110')
BitParser.add(WHITE, 2560, '000000011111')
BLACK = [None, None]
BitParser.add(BLACK, 0 , '0000110111')
BitParser.add(BLACK, 1 , '010')
BitParser.add(BLACK, 2 , '11')
BitParser.add(BLACK, 3 , '10')
BitParser.add(BLACK, 4 , '011')
BitParser.add(BLACK, 5 , '0011')
BitParser.add(BLACK, 6 , '0010')
BitParser.add(BLACK, 7 , '00011')
BitParser.add(BLACK, 8 , '000101')
BitParser.add(BLACK, 9 , '000100')
BitParser.add(BLACK, 10 , '0000100')
BitParser.add(BLACK, 11 , '0000101')
BitParser.add(BLACK, 12 , '0000111')
BitParser.add(BLACK, 13 , '00000100')
BitParser.add(BLACK, 14 , '00000111')
BitParser.add(BLACK, 15 , '000011000')
BitParser.add(BLACK, 16 , '0000010111')
BitParser.add(BLACK, 17 , '0000011000')
BitParser.add(BLACK, 18 , '0000001000')
BitParser.add(BLACK, 19 , '00001100111')
BitParser.add(BLACK, 20 , '00001101000')
BitParser.add(BLACK, 21 , '00001101100')
BitParser.add(BLACK, 22 , '00000110111')
BitParser.add(BLACK, 23 , '00000101000')
BitParser.add(BLACK, 24 , '00000010111')
BitParser.add(BLACK, 25 , '00000011000')
BitParser.add(BLACK, 26 , '000011001010')
BitParser.add(BLACK, 27 , '000011001011')
BitParser.add(BLACK, 28 , '000011001100')
BitParser.add(BLACK, 29 , '000011001101')
BitParser.add(BLACK, 30 , '000001101000')
BitParser.add(BLACK, 31 , '000001101001')
BitParser.add(BLACK, 32 , '000001101010')
BitParser.add(BLACK, 33 , '000001101011')
BitParser.add(BLACK, 34 , '000011010010')
BitParser.add(BLACK, 35 , '000011010011')
BitParser.add(BLACK, 36 , '000011010100')
BitParser.add(BLACK, 37 , '000011010101')
BitParser.add(BLACK, 38 , '000011010110')
BitParser.add(BLACK, 39 , '000011010111')
BitParser.add(BLACK, 40 , '000001101100')
BitParser.add(BLACK, 41 , '000001101101')
BitParser.add(BLACK, 42 , '000011011010')
BitParser.add(BLACK, 43 , '000011011011')
BitParser.add(BLACK, 44 , '000001010100')
BitParser.add(BLACK, 45 , '000001010101')
BitParser.add(BLACK, 46 , '000001010110')
BitParser.add(BLACK, 47 , '000001010111')
BitParser.add(BLACK, 48 , '000001100100')
BitParser.add(BLACK, 49 , '000001100101')
BitParser.add(BLACK, 50 , '000001010010')
BitParser.add(BLACK, 51 , '000001010011')
BitParser.add(BLACK, 52 , '000000100100')
BitParser.add(BLACK, 53 , '000000110111')
BitParser.add(BLACK, 54 , '000000111000')
BitParser.add(BLACK, 55 , '000000100111')
BitParser.add(BLACK, 56 , '000000101000')
BitParser.add(BLACK, 57 , '000001011000')
BitParser.add(BLACK, 58 , '000001011001')
BitParser.add(BLACK, 59 , '000000101011')
BitParser.add(BLACK, 60 , '000000101100')
BitParser.add(BLACK, 61 , '000001011010')
BitParser.add(BLACK, 62 , '000001100110')
BitParser.add(BLACK, 63 , '000001100111')
BitParser.add(BLACK, 64 , '0000001111')
BitParser.add(BLACK, 128 , '000011001000')
BitParser.add(BLACK, 192 , '000011001001')
BitParser.add(BLACK, 256 , '000001011011')
BitParser.add(BLACK, 320 , '000000110011')
BitParser.add(BLACK, 384 , '000000110100')
BitParser.add(BLACK, 448 , '000000110101')
BitParser.add(BLACK, 512 , '0000001101100')
BitParser.add(BLACK, 576 , '0000001101101')
BitParser.add(BLACK, 640 , '0000001001010')
BitParser.add(BLACK, 704 , '0000001001011')
BitParser.add(BLACK, 768 , '0000001001100')
BitParser.add(BLACK, 832 , '0000001001101')
BitParser.add(BLACK, 896 , '0000001110010')
BitParser.add(BLACK, 960 , '0000001110011')
BitParser.add(BLACK, 1024, '0000001110100')
BitParser.add(BLACK, 1088, '0000001110101')
BitParser.add(BLACK, 1152, '0000001110110')
BitParser.add(BLACK, 1216, '0000001110111')
BitParser.add(BLACK, 1280, '0000001010010')
BitParser.add(BLACK, 1344, '0000001010011')
BitParser.add(BLACK, 1408, '0000001010100')
BitParser.add(BLACK, 1472, '0000001010101')
BitParser.add(BLACK, 1536, '0000001011010')
BitParser.add(BLACK, 1600, '0000001011011')
BitParser.add(BLACK, 1664, '0000001100100')
BitParser.add(BLACK, 1728, '0000001100101')
BitParser.add(BLACK, 1792, '00000001000')
BitParser.add(BLACK, 1856, '00000001100')
BitParser.add(BLACK, 1920, '00000001101')
BitParser.add(BLACK, 1984, '000000010010')
BitParser.add(BLACK, 2048, '000000010011')
BitParser.add(BLACK, 2112, '000000010100')
BitParser.add(BLACK, 2176, '000000010101')
BitParser.add(BLACK, 2240, '000000010110')
BitParser.add(BLACK, 2304, '000000010111')
BitParser.add(BLACK, 2368, '000000011100')
BitParser.add(BLACK, 2432, '000000011101')
BitParser.add(BLACK, 2496, '000000011110')
BitParser.add(BLACK, 2560, '000000011111')
UNCOMPRESSED = [None, None]
BitParser.add(UNCOMPRESSED, '1', '1')
BitParser.add(UNCOMPRESSED, '01', '01')
BitParser.add(UNCOMPRESSED, '001', '001')
BitParser.add(UNCOMPRESSED, '0001', '0001')
BitParser.add(UNCOMPRESSED, '00001', '00001')
BitParser.add(UNCOMPRESSED, '00000', '000001')
BitParser.add(UNCOMPRESSED, 'T00', '00000011')
BitParser.add(UNCOMPRESSED, 'T10', '00000010')
BitParser.add(UNCOMPRESSED, 'T000', '000000011')
BitParser.add(UNCOMPRESSED, 'T100', '000000010')
BitParser.add(UNCOMPRESSED, 'T0000', '0000000011')
BitParser.add(UNCOMPRESSED, 'T1000', '0000000010')
BitParser.add(UNCOMPRESSED, 'T00000', '00000000011')
BitParser.add(UNCOMPRESSED, 'T10000', '00000000010')
class EOFB(Exception):
pass
class InvalidData(Exception):
pass
class ByteSkip(Exception):
pass
def __init__(self, width, bytealign=False):
BitParser.__init__(self)
self.width = width
self.bytealign = bytealign
self.reset()
return
def feedbytes(self, data):
for byte in get_bytes(data):
try:
for m in (128, 64, 32, 16, 8, 4, 2, 1):
self._parse_bit(byte & m)
except self.ByteSkip:
self._accept = self._parse_mode
self._state = self.MODE
except self.EOFB:
break
return
def _parse_mode(self, mode):
if mode == 'p':
self._do_pass()
self._flush_line()
return self.MODE
elif mode == 'h':
self._n1 = 0
self._accept = self._parse_horiz1
if self._color:
return self.WHITE
else:
return self.BLACK
elif mode == 'u':
self._accept = self._parse_uncompressed
return self.UNCOMPRESSED
elif mode == 'e':
raise self.EOFB
elif isinstance(mode, int):
self._do_vertical(mode)
self._flush_line()
return self.MODE
else:
raise self.InvalidData(mode)
def _parse_horiz1(self, n):
if n is None:
raise self.InvalidData
self._n1 += n
if n < 64:
self._n2 = 0
self._color = 1-self._color
self._accept = self._parse_horiz2
if self._color:
return self.WHITE
else:
return self.BLACK
def _parse_horiz2(self, n):
if n is None:
raise self.InvalidData
self._n2 += n
if n < 64:
self._color = 1-self._color
self._accept = self._parse_mode
self._do_horizontal(self._n1, self._n2)
self._flush_line()
return self.MODE
elif self._color:
return self.WHITE
else:
return self.BLACK
def _parse_uncompressed(self, bits):
if not bits:
raise self.InvalidData
if bits.startswith('T'):
self._accept = self._parse_mode
self._color = int(bits[1])
self._do_uncompressed(bits[2:])
return self.MODE
else:
self._do_uncompressed(bits)
return self.UNCOMPRESSED
def _get_bits(self):
return ''.join(str(b) for b in self._curline[:self._curpos])
def _get_refline(self, i):
if i < 0:
return '[]'+''.join(str(b) for b in self._refline)
elif len(self._refline) <= i:
return ''.join(str(b) for b in self._refline)+'[]'
else:
return (''.join(str(b) for b in self._refline[:i]) +
'['+str(self._refline[i])+']' +
''.join(str(b) for b in self._refline[i+1:]))
def reset(self):
self._y = 0
self._curline = array.array('b', [1]*self.width)
self._reset_line()
self._accept = self._parse_mode
self._state = self.MODE
return
def output_line(self, y, bits):
print (y, ''.join(str(b) for b in bits))
return
def _reset_line(self):
self._refline = self._curline
self._curline = array.array('b', [1]*self.width)
self._curpos = -1
self._color = 1
return
def _flush_line(self):
if self.width <= self._curpos:
self.output_line(self._y, self._curline)
self._y += 1
self._reset_line()
if self.bytealign:
raise self.ByteSkip
return
def _do_vertical(self, dx):
x1 = self._curpos+1
while 1:
if x1 == 0:
if (self._color == 1 and self._refline[x1] != self._color):
break
elif x1 == len(self._refline):
break
elif (self._refline[x1-1] == self._color and
self._refline[x1] != self._color):
break
x1 += 1
x1 += dx
x0 = max(0, self._curpos)
x1 = max(0, min(self.width, x1))
if x1 < x0:
for x in range(x1, x0):
self._curline[x] = self._color
elif x0 < x1:
for x in range(x0, x1):
self._curline[x] = self._color
self._curpos = x1
self._color = 1-self._color
return
def _do_pass(self):
x1 = self._curpos+1
while 1:
if x1 == 0:
if (self._color == 1 and self._refline[x1] != self._color):
break
elif x1 == len(self._refline):
break
elif (self._refline[x1-1] == self._color and
self._refline[x1] != self._color):
break
x1 += 1
while 1:
if x1 == 0:
if (self._color == 0 and self._refline[x1] == self._color):
break
elif x1 == len(self._refline):
break
elif (self._refline[x1-1] != self._color and
self._refline[x1] == self._color):
break
x1 += 1
for x in range(self._curpos, x1):
self._curline[x] = self._color
self._curpos = x1
return
def _do_horizontal(self, n1, n2):
if self._curpos < 0:
self._curpos = 0
x = self._curpos
for _ in range(n1):
if len(self._curline) <= x:
break
self._curline[x] = self._color
x += 1
for _ in range(n2):
if len(self._curline) <= x:
break
self._curline[x] = 1-self._color
x += 1
self._curpos = x
return
def _do_uncompressed(self, bits):
for c in bits:
self._curline[self._curpos] = int(c)
self._curpos += 1
self._flush_line()
return
class CCITTFaxDecoder(CCITTG4Parser):
def __init__(self, width, bytealign=False, reversed=False):
CCITTG4Parser.__init__(self, width, bytealign=bytealign)
self.reversed = reversed
self._buf = b''
return
def close(self):
return self._buf
def output_line(self, y, bits):
bytes = array.array('B', [0]*((len(bits)+7)//8))
if self.reversed:
bits = [1-b for b in bits]
for (i, b) in enumerate(bits):
if b:
bytes[i//8] += (128, 64, 32, 16, 8, 4, 2, 1)[i % 8]
self._buf += bytes.tostring()
return
def ccittfaxdecode(data, params):
K = params.get('K')
cols = params.get('Columns')
bytealign = params.get('EncodedByteAlign')
reversed = params.get('BlackIs1')
if K == -1:
parser = CCITTFaxDecoder(cols, bytealign=bytealign, reversed=reversed)
else:
raise ValueError(K)
parser.feedbytes(data)
return parser.close()
# test
def main(argv):
if not argv[1:]:
import unittest
return unittest.main()
class Parser(CCITTG4Parser):
def __init__(self, width, bytealign=False):
import pygame
CCITTG4Parser.__init__(self, width, bytealign=bytealign)
self.img = pygame.Surface((self.width, 1000))
return
def output_line(self, y, bits):
for (x, b) in enumerate(bits):
if b:
self.img.set_at((x, y), (255, 255, 255))
else:
self.img.set_at((x, y), (0, 0, 0))
return
def close(self):
import pygame
pygame.image.save(self.img, 'out.bmp')
return
for path in argv[1:]:
fp = open(path, 'rb')
(_, _, _, w, _, _) = path.split('.')
parser = Parser(int(w))
parser.feedbytes(fp.read())
parser.close()
fp.close()
return
if __name__ == '__main__':
sys.exit(main(sys.argv))
@@ -1,13 +0,0 @@
from ...utils import settings
from ..PSStackParser import LIT
LITERAL_CRYPT = LIT('Crypt')
# Abbreviation of Filter names in PDF 4.8.6. "Inline Images"
LITERALS_FLATE_DECODE = (LIT('FlateDecode'), LIT('Fl'))
LITERALS_LZW_DECODE = (LIT('LZWDecode'), LIT('LZW'))
LITERALS_ASCII85_DECODE = (LIT('ASCII85Decode'), LIT('A85'))
LITERALS_ASCIIHEX_DECODE = (LIT('ASCIIHexDecode'), LIT('AHx'))
LITERALS_RUNLENGTH_DECODE = (LIT('RunLengthDecode'), LIT('RL'))
LITERALS_CCITTFAX_DECODE = (LIT('CCITTFaxDecode'), LIT('CCF'))
LITERALS_DCT_DECODE = (LIT('DCTDecode'), LIT('DCT'))
@@ -1,16 +0,0 @@
from ..PSStackParser import PSException
class PDFException(PSException):
pass
class PDFTypeError(PDFException):
pass
class PDFValueError(PDFException):
pass
class PDFObjectNotFound(PDFException):
pass
class PDFNotImplementedError(PDFException):
pass
-6
View File
@@ -1,6 +0,0 @@
from ..PSStackParser import PSObject
## PDF Objects
##
class PDFObject(PSObject):
pass
-139
View File
@@ -1,139 +0,0 @@
from .PDFObjRef import PDFObjRef
from ...utils import settings, isnumber, int2byte
from .exceptions import PDFTypeError
from logging import getLogger
log = getLogger(__name__)
# resolve
def resolve1(x, default=None):
"""Resolves an object.
If this is an array or dictionary, it may still contains
some indirect objects inside.
"""
while isinstance(x, PDFObjRef):
x = x.resolve(default=default)
return x
def resolve_all(x, default=None):
"""Recursively resolves the given object and all the internals.
Make sure there is no indirect reference within the nested object.
This procedure might be slow.
"""
while isinstance(x, PDFObjRef):
x = x.resolve(default=default)
if isinstance(x, list):
x = [resolve_all(v, default=default) for v in x]
elif isinstance(x, dict):
for (k, v) in x.iteritems():
x[k] = resolve_all(v, default=default)
return x
def decipher_all(decipher, objid, genno, x):
"""Recursively deciphers the given object.
"""
if isinstance(x, bytes):
return decipher(objid, genno, x)
if isinstance(x, list):
x = [decipher_all(decipher, objid, genno, v) for v in x]
elif isinstance(x, dict):
for (k, v) in x.items():
x[k] = decipher_all(decipher, objid, genno, v)
return x
# Type cheking
def int_value(x):
x = resolve1(x)
if not isinstance(x, int):
if settings.STRICT:
raise PDFTypeError('Integer required: %r' % x)
return 0
return x
def float_value(x):
x = resolve1(x)
if not isinstance(x, float):
if settings.STRICT:
raise PDFTypeError('Float required: %r' % x)
return 0.0
return x
def num_value(x):
x = resolve1(x)
if not isnumber(x):
if settings.STRICT:
raise PDFTypeError('Int or Float required: %r' % x)
return 0
return x
def str_value(x):
x = resolve1(x)
if not isinstance(x, bytes):
if settings.STRICT:
raise PDFTypeError('String required: %r' % x)
return ''
return x
def list_value(x):
x = resolve1(x)
if not isinstance(x, (list, tuple)):
if settings.STRICT:
raise PDFTypeError('List required: %r' % x)
return []
return x
def dict_value(x):
x = resolve1(x)
if not isinstance(x, dict):
if settings.STRICT:
log.error('PDFTypeError : Dict required: %r', x)
raise PDFTypeError('Dict required: %r' % x)
return {}
return x
#
# RunLength decoder (Adobe version) implementation based on PDF Reference
# version 1.4 section 3.3.4.
#
# * public domain *
#
def rldecode(data):
"""
RunLength decoder (Adobe version) implementation based on PDF Reference
version 1.4 section 3.3.4:
The RunLengthDecode filter decodes data that has been encoded in a
simple byte-oriented format based on run length. The encoded data
is a sequence of runs, where each run consists of a length byte
followed by 1 to 128 bytes of data. If the length byte is in the
range 0 to 127, the following length + 1 (1 to 128) bytes are
copied literally during decompression. If length is in the range
129 to 255, the following single byte is to be copied 257 - length
(2 to 128) times during decompression. A length value of 128
denotes EOD.
"""
decoded = b''
i = 0
while i < len(data):
length = data[i]
if length == 128:
break
if length >= 0 and length < 128:
for j in range(i+1,(i+1)+(length+1)):
decoded+=int2byte(data[j])
i = (i+1) + (length+1)
if length > 128:
run = int2byte(data[i+1])*(257-length)
decoded+=run
i = (i+1) + 1
return decoded
@@ -1,345 +0,0 @@
import logging
log = logging.getLogger(__name__)
from io import TextIOWrapper
from typing import List, Tuple, Union
from ...utils import int2byte
from .exceptions import PSEOF
from .constants import EOL, NONSPC, KWD, END_LITERAL, LIT, HEX
from .constants import SPC, HEX_PAIR
from .constants import END_NUMBER, END_STRING, END_KEYWORD, END_HEX_STRING
from .constants import OCT_STRING, ESC_STRING, KEYWORD_DICT_BEGIN, KEYWORD_DICT_END
from .types import PSLiteral
class PSBaseParser(object):
"""Most basic PostScript parser that performs only tokenization.
"""
BUFSIZ = 4096
def __init__(self, fp: TextIOWrapper):
self.fp = fp
self.seek(0)
return
def __repr__(self):
return '<%s: %r, bufpos=%d>' % (self.__class__.__name__, self.fp, self.bufpos)
def flush(self):
return
def close(self):
self.flush()
return
def tell(self):
return self.bufpos+self.charpos
def poll(self, pos=None, n=80):
pos0 = self.fp.tell()
if not pos:
pos = self.bufpos+self.charpos
self.fp.seek(pos)
log.info('poll(%d): %r', pos, self.fp.read(n))
self.fp.seek(pos0)
return
def seek(self, pos: int):
"""Seeks the parser to the given position.
"""
log.debug('seek: %r', pos)
self.fp.seek(pos)
# reset the status for nextline()
self.bufpos: int = pos
self.buf: bytes = b''
self.charpos: int = 0
# reset the status for nexttoken()
self._current_parse_func = self._parse_main
self._curtoken: bytes = b''
self._curtokenpos: int = 0
self._tokens: List[Tuple[int, Union[int, float, bool, str, bytes, PSLiteral]]] = []
return
def fillbuf(self):
if self.charpos < len(self.buf):
return
# fetch next chunk.
self.bufpos = self.fp.tell()
self.buf = self.fp.read(self.BUFSIZ)
if not self.buf:
raise PSEOF('Unexpected EOF')
self.charpos = 0
return
def nextline(self):
"""Fetches a next line that ends either with \\r or \\n.
"""
linebuf = b''
linepos = self.bufpos + self.charpos
eol = False
while 1:
self.fillbuf()
if eol:
c = self.buf[self.charpos:self.charpos+1]
# handle b'\r\n'
if c == b'\n':
linebuf += c
self.charpos += 1
break
m = EOL.search(self.buf, self.charpos)
if m:
linebuf += self.buf[self.charpos:m.end(0)]
self.charpos = m.end(0)
if linebuf[-1:] == b'\r':
eol = True
else:
break
else:
linebuf += self.buf[self.charpos:]
self.charpos = len(self.buf)
log.debug('nextline: %r, %r', linepos, linebuf)
return (linepos, linebuf)
def revreadlines(self):
"""Fetches a next line backword.
This is used to locate the trailers at the end of a file.
"""
self.fp.seek(0, 2)
pos = self.fp.tell()
buf = b''
while 0 < pos:
prevpos = pos
pos = max(0, pos-self.BUFSIZ)
self.fp.seek(pos)
s = self.fp.read(prevpos-pos)
if not s:
break
while 1:
n = max(s.rfind(b'\r'), s.rfind(b'\n'))
if n == -1:
buf = s + buf
break
yield s[n:] + buf
s = s[:n]
buf = b''
return
def _parse_main(self, s, i):
m = NONSPC.search(s, i)
if not m:
return len(s)
j = m.start(0)
c = s[j:j+1]
self._curtokenpos = self.bufpos+j
if c == b'%':
self._curtoken = b'%'
self._current_parse_func = self._parse_comment
return j+1
elif c == b'/':
self._curtoken = b''
self._current_parse_func = self._parse_literal
return j+1
elif c in b'-+' or c.isdigit():
self._curtoken = c
self._current_parse_func = self._parse_number
return j+1
elif c == b'.':
self._curtoken = c
self._current_parse_func = self._parse_float
return j+1
elif c.isalpha():
self._curtoken = c
self._current_parse_func = self._parse_keyword
return j+1
elif c == b'(':
self._curtoken = b''
self.paren = 1
self._current_parse_func = self._parse_string
return j+1
elif c == b'<':
self._curtoken = b''
self._current_parse_func = self._parse_wopen
return j+1
elif c == b'>':
self._curtoken = b''
self._current_parse_func = self._parse_wclose
return j+1
else:
self._add_token(KWD(c))
return j+1
def _add_token(self, obj):
self._tokens.append((self._curtokenpos, obj))
return
def _parse_comment(self, s, i):
m = EOL.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
self._current_parse_func = self._parse_main
# We ignore comments.
#self._tokens.append(self._curtoken)
return j
def _parse_literal(self, s, i):
m = END_LITERAL.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
c = s[j:j+1]
if c == b'#':
self.hex = b''
self._current_parse_func = self._parse_literal_hex
return j+1
try:
self._curtoken=str(self._curtoken,'utf-8')
except:
pass
self._add_token(LIT(self._curtoken))
self._current_parse_func = self._parse_main
return j
def _parse_literal_hex(self, s, i):
c = s[i:i+1]
if HEX.match(c) and len(self.hex) < 2:
self.hex += c
return i+1
if self.hex:
self._curtoken += int2byte(int(self.hex, 16))
self._current_parse_func = self._parse_literal
return i
def _parse_number(self, s, i):
m = END_NUMBER.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
c = s[j:j+1]
if c == b'.':
self._curtoken += c
self._current_parse_func = self._parse_float
return j+1
try:
self._add_token(int(self._curtoken))
except ValueError:
pass
self._current_parse_func = self._parse_main
return j
def _parse_float(self, s, i):
m = END_NUMBER.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
try:
self._add_token(float(self._curtoken))
except ValueError:
pass
self._current_parse_func = self._parse_main
return j
def _parse_keyword(self, s, i):
m = END_KEYWORD.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
if self._curtoken == b'true':
token = True
elif self._curtoken == b'false':
token = False
else:
token = KWD(self._curtoken)
self._add_token(token)
self._current_parse_func = self._parse_main
return j
def _parse_string(self, s, i):
m = END_STRING.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
c = s[j:j+1]
if c == b'\\':
self.oct = b''
self._current_parse_func = self._parse_string_1
return j+1
if c == b'(':
self.paren += 1
self._curtoken += c
return j+1
if c == b')':
self.paren -= 1
if self.paren: # WTF, they said balanced parens need no special treatment.
self._curtoken += c
return j+1
self._add_token(self._curtoken)
self._current_parse_func = self._parse_main
return j+1
def _parse_string_1(self, s, i):
c = s[i:i+1]
if OCT_STRING.match(c) and len(self.oct) < 3:
self.oct += c
return i+1
if self.oct:
self._curtoken += int2byte(int(self.oct, 8))
self._current_parse_func = self._parse_string
return i
if c in ESC_STRING:
self._curtoken += int2byte(ESC_STRING[c])
self._current_parse_func = self._parse_string
return i+1
def _parse_wopen(self, s, i):
c = s[i:i+1]
if c == b'<':
self._add_token(KEYWORD_DICT_BEGIN)
self._current_parse_func = self._parse_main
i += 1
else:
self._current_parse_func = self._parse_hexstring
return i
def _parse_wclose(self, s, i):
c = s[i:i+1]
if c == b'>':
self._add_token(KEYWORD_DICT_END)
i += 1
self._current_parse_func = self._parse_main
return i
def _parse_hexstring(self, s, i):
m = END_HEX_STRING.search(s, i)
if not m:
self._curtoken += s[i:]
return len(s)
j = m.start(0)
self._curtoken += s[i:j]
token = HEX_PAIR.sub(lambda m: int2byte(int(m.group(0), 16)),SPC.sub(b'', self._curtoken))
self._add_token(token)
self._current_parse_func = self._parse_main
return j
def nexttoken(self):
while not self._tokens:
self.fillbuf()
self.charpos = self._current_parse_func(self.buf, self.charpos)
token = self._tokens.pop(0)
log.debug('nexttoken: %r', token)
return token
@@ -1,168 +0,0 @@
import logging
from io import TextIOWrapper
from ...utils import choplist, settings
from .PSBaseParser import PSBaseParser
from .types import *
from .exceptions import *
from .constants import *
log = logging.getLogger(__name__)
def literal_name(x):
if not isinstance(x, PSLiteral):
if settings.STRICT:
raise PSTypeError('Literal required: %r' % (x,))
else:
name=x
else:
name=x.name
try:
name = str(name,'utf-8')
except:
pass
return name
def keyword_name(x):
if not isinstance(x, PSKeyword):
if settings.STRICT:
raise PSTypeError('Keyword required: %r' % x)
else:
name=x
else:
name=x.name
name = str(name,'utf-8','ignore')
return name
class PSStackParser(PSBaseParser):
def __init__(self, fp: TextIOWrapper):
PSBaseParser.__init__(self, fp)
self.reset()
return
def reset(self):
self.context = []
self.curtype = None
self.curstack = []
self.results = []
return
def seek(self, pos):
PSBaseParser.seek(self, pos)
self.reset()
return
def push(self, *objs):
self.curstack.extend(objs)
return
def pop(self, n):
objs = self.curstack[-n:]
self.curstack[-n:] = []
return objs
def popall(self):
objs = self.curstack
self.curstack = []
return objs
def add_results(self, *objs):
try:
log.debug('add_results: %r', objs)
except:
log.debug('add_results: (unprintable object)')
self.results.extend(objs)
return
def start_type(self, pos, type):
self.context.append((pos, self.curtype, self.curstack))
(self.curtype, self.curstack) = (type, [])
log.debug('start_type: pos=%r, type=%r', pos, type)
return
def end_type(self, type):
if self.curtype != type:
raise PSTypeError('Type mismatch: %r != %r' % (self.curtype, type))
objs = [obj for (_, obj) in self.curstack]
(pos, self.curtype, self.curstack) = self.context.pop()
log.debug('end_type: pos=%r, type=%r, objs=%r', pos, type, objs)
return (pos, objs)
def do_keyword(self, pos, token):
return
def __iter__(self):
while True:
try:
(_, obj) = self.nextobject()
yield obj
except PSEOF:
break
def nextobject(self):
"""Yields a list of objects.
Returns keywords, literals, strings, numbers, arrays and dictionaries.
Arrays and dictionaries are represented as Python lists and dictionaries.
"""
while not self.results:
(pos, token) = self.nexttoken()
if isinstance(token, (int, float, bool, str, bytes, 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 settings.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)
# construct a Python dictionary.
d = dict((literal_name(k), v) for (k, v) in choplist(2, objs) if v is not None)
self.push((pos, d))
except PSTypeError:
if settings.STRICT:
raise
elif token == KEYWORD_PROC_BEGIN:
# begin proc
self.start_type(pos, 'p')
elif token == KEYWORD_PROC_END:
# end proc
try:
self.push(self.end_type('p'))
except PSTypeError:
if settings.STRICT:
raise
elif isinstance(token,PSKeyword):
log.debug('do_keyword: pos=%r, token=%r, stack=%r', pos, token, self.curstack)
self.do_keyword(pos, token)
else:
log.error('unknown token: pos=%r, token=%r, stack=%r', pos, token, self.curstack)
self.do_keyword(pos, token)
raise Exception("unknown tokens")
if self.context:
continue
else:
self.flush()
obj = self.results.pop(0)
try:
log.debug('nextobject: %r', obj)
except:
log.debug('nextobject: (unprintable object)')
return obj
@@ -1,28 +0,0 @@
import re
from .types import PSLiteral, PSSymbolTable, PSKeyword
PSLiteralTable = PSSymbolTable(PSLiteral)
PSKeywordTable = PSSymbolTable(PSKeyword)
LIT = PSLiteralTable.intern
KWD = PSKeywordTable.intern
KEYWORD_PROC_BEGIN = KWD(b'{')
KEYWORD_PROC_END = KWD(b'}')
KEYWORD_ARRAY_BEGIN = KWD(b'[')
KEYWORD_ARRAY_END = KWD(b']')
KEYWORD_DICT_BEGIN = KWD(b'<<')
KEYWORD_DICT_END = KWD(b'>>')
## PSBaseParser
##
EOL = re.compile(br'[\r\n]')
SPC = re.compile(br'\s')
NONSPC = re.compile(br'\S')
HEX = re.compile(br'[0-9a-fA-F]')
END_LITERAL = re.compile(br'[#/%\[\]()<>{}\s]')
END_HEX_STRING = re.compile(br'[^\s0-9a-fA-F]')
HEX_PAIR = re.compile(br'[0-9a-fA-F]{2}|.')
END_NUMBER = re.compile(br'[^0-9]')
END_KEYWORD = re.compile(br'[#/%\[\]()<>{}\s]')
END_STRING = re.compile(br'[()\134]')
OCT_STRING = re.compile(br'[0-7]')
ESC_STRING = {b'b': 8, b't': 9, b'n': 10, b'f': 12, b'r': 13, b'(': 40, b')': 41, b'\\': 92}
@@ -1,20 +0,0 @@
## PS Exceptions
##
class PSException(Exception):
pass
class PSEOF(PSException):
pass
class PSSyntaxError(PSException):
pass
class PSTypeError(PSException):
pass
class PSValueError(PSException):
pass
@@ -1,80 +0,0 @@
## Basic PostScript Types
##
from typing import Union
## PSObject
##
class PSObject(object):
"""Base class for all PS or PDF-related data types."""
pass
## PSLiteral
##
class PSLiteral(PSObject):
"""A class that represents a PostScript literal.
Postscript literals are used as identifiers, such as
variable names, property names and dictionary keys.
Literals are case sensitive and denoted by a preceding
slash sign (e.g. "/Name")
Note: Do not create an instance of PSLiteral directly.
Always use PSLiteralTable.intern().
"""
def __init__(self, name):
self.name = name
def __repr__(self):
name=self.name
return '/%r' % name
## PSKeyword
##
class PSKeyword(PSObject):
"""A class that represents a PostScript keyword.
PostScript keywords are a dozen of predefined words.
Commands and directives in PostScript are expressed by keywords.
They are also used to denote the content boundaries.
Note: Do not create an instance of PSKeyword directly.
Always use PSKeywordTable.intern().
"""
def __init__(self, name):
self.name = name
return
def __repr__(self):
name=self.name
return '/%r' % name
## PSSymbolTable
##
class PSSymbolTable(object):
"""A utility class for storing PSLiteral/PSKeyword objects.
Interned objects can be checked its identity with "is" operator.
"""
def __init__(self, klass):
self.dict = {}
self.klass = klass
return
def intern(self, name):
if name in self.dict:
lit = self.dict[name]
else:
lit = self.klass(name)
self.dict[name] = lit
return lit
+78
View File
@@ -0,0 +1,78 @@
from pdfmajor.parser.PDFPage import PDFPage
from pdfmajor.parser.PDFStream import list_value
from pdfmajor.parser.PDFContentParser import PDFContentParser
from pdfmajor.parser.PSStackParser import PSKeyword
from pdfmajor.parser.PSStackParser import keyword_name
from .commands import PDFStateStack, PDFCommands
from .commands import LTXObject
from .utils import init_resources
from .XObjectInterpreter import XObjectInterpreter
class PageInterpreter:
class PDFInterpreterError(Exception): pass
def __init__(self, page: PDFPage, page_num: int, font_cache: dict = {}):
(x0, y0, x1, y1) = page.mediabox
if page.rotate == 90:
ctm = [0, -1, 1, 0, -y0, x1]
elif page.rotate == 180:
ctm = [-1, 0, 0, -1, x1, y1]
elif page.rotate == 270:
ctm = [0, 1, -1, 0, y1, -x0]
else:
ctm = [1, 0, 0, 1, -x0, -y0]
# Locals
self.font_cache = font_cache
self.page = page
self.page_num = page_num
# Init State
self.state: PDFStateStack = PDFStateStack()
self.state.t_matrix = ctm
self.state.resources = page.resources
# set some global states.
self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = None
if self.state.colorspace_map:
col_space = next(iter(self.state.colorspace_map.values()))
self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = col_space
# Init Resources
init_resources(self.state, self.font_cache)
def __iter__(self):
parser = PDFContentParser(list_value(self.page.contents))
for obj in parser:
if isinstance(obj, PSKeyword):
name = keyword_name(obj)
method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
if method in PDFCommands.commands.keys():
func = PDFCommands.commands.get(method)
nargs = func.__code__.co_argcount-1
if nargs:
args = self.state.pop(nargs)
if len(args) == nargs:
func(self.state, *args)
else:
raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}")
else:
# log.debug('exec: %s', name)
func(self.state)
else:
raise self.PDFInterpreterError('Unknown operator: %r' % name)
else:
self.state.argstack.append(obj)
for complete_item in self.state.complete_layout_items:
if isinstance(complete_item, LTXObject):
interpeter = XObjectInterpreter(
streams=[complete_item.stream],
resources=complete_item.resources,
ctm=complete_item.t_matrix
)
for item in interpeter:
complete_item.add(item)
yield complete_item
self.state.complete_layout_items = []
def __repr__(self) -> str:
return f"<Page:{self.page_num}/>"
@@ -0,0 +1,70 @@
from pdfmajor.utils import MATRIX_IDENTITY
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFParser import PDFStream
from pdfmajor.parser.PSStackParser import PSEOF
from pdfmajor.parser.PSStackParser import PSKeyword
from pdfmajor.parser.PSStackParser import keyword_name
from pdfmajor.parser.PSStackParser import PSStackParser
from pdfmajor.parser.constants import LITERAL_PDF, LITERAL_TEXT, LITERAL_FONT
from pdfmajor.parser.PDFStream import resolve1, list_value, dict_value
from pdfmajor.parser.PDFStream.PDFObjRef import PDFObjRef
from pdfmajor.parser.PDFContentParser import PDFContentParser
from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE
from .commands import PDFCommands, LTXObject
from .utils import init_resources
class XObjectInterpreter:
class PDFInterpreterError(Exception): pass
def __init__(self, streams, resources, ctm):
self.streams = streams
self.font_cache = {}
self.state = PDFStateStack()
self.state.t_matrix = ctm
self.state.resources = resources
# set some global states.
self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = None
if self.state.colorspace_map:
col_space = next(iter(self.state.colorspace_map.values()))
self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = col_space
init_resources(self.state, self.font_cache)
def __iter__(self):
parser = PDFContentParser(self.streams)
for obj in parser:
if isinstance(obj, PSKeyword):
name = keyword_name(obj)
method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
if method in PDFCommands.commands.keys():
func = PDFCommands.commands.get(method)
nargs = func.__code__.co_argcount-1
if nargs:
args = self.state.pop(nargs)
if len(args) == nargs:
func(self.state, *args)
else:
raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}")
else:
# log.debug('exec: %s', name)
func(self.state)
else:
raise self.PDFInterpreterError('Unknown operator: %r' % name)
else:
self.state.argstack.append(obj)
for complete_item in self.state.complete_layout_items:
if isinstance(complete_item, LTXObject):
interpeter = self.__class__(
streams=[complete_item.stream],
resources=complete_item.resources,
ctm=complete_item.t_matrix
)
for item in interpeter:
complete_item.add(item)
yield complete_item
self.state.complete_layout_items = []
+72
View File
@@ -0,0 +1,72 @@
from typing import List
from tqdm import tqdm
from ..parser.PDFPage import PDFPage
from ..utils import set_log_level, get_logger, logging
from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE
from .commands import PDFCommands
from .commands import LTCharBlock, LTChar
from .commands import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
from .commands import LTImage
from .commands import LTXObject
from .PageInterpreter import PageInterpreter
log = get_logger(__name__)
class PDFInterpreter:
class AccessError(Exception): pass
def __init__(self,
input_file_path: str,
maxpages: int = 0,
caching: bool = True,
check_extractable: bool = True,
password: str = None,
pagenos: List[int] = None,
preload: bool = False,
debug_level: int = logging.WARNING,
):
self.input_file_path = input_file_path
self.maxpages = maxpages
self.caching = caching
self.check_extractable = check_extractable
self.password = password
self.pagenos = pagenos
self.debug_level = debug_level
self.__pages = []
if preload:
for _ in self.__load_pages():
pass
def __load_pages(self):
set_log_level(self.debug_level)
log.info("Opening file...")
with open(self.input_file_path, 'rb') as input_file:
font_cache = {}
log.info("Parsing file...")
pages = PDFPage.get_pages(
input_file,
pagenos=self.pagenos,
maxpages=self.maxpages,
password=self.password,
caching=self.caching,
check_extractable=self.check_extractable
)
for page_num, page in tqdm(enumerate(pages), desc="Reading pages...", disable=self.debug_level > logging.INFO):
self.__pages.append(PageInterpreter(page, page_num, font_cache))
yield self.__pages[-1]
log.info(f"Done Reading {len(self.__pages)} pages.")
set_log_level(logging.WARNING)
if len(self.__pages) == 0:
raise self.AccessError("No pages found in pdf-file")
def __iter__(self):
if len(self.__pages) > 0:
set_log_level(self.debug_level)
for page in tqdm(self.__pages, desc="Reading pages from store...", disable=self.debug_level > logging.INFO):
yield page
set_log_level(logging.WARNING)
else:
for page in self.__load_pages():
yield page
-432
View File
@@ -1,432 +0,0 @@
""" Adobe character mapping (CMap) support.
CMaps provide the mapping between character codes and Unicode
code-points to character ids (CIDs).
More information is available on the Adobe website:
http://opensource.adobe.com/wiki/display/cmap/CMap+Resources
"""
import sys
import os
import os.path
import gzip
import pickle as pickle
import struct
import logging
from ...utils import choplist, nunpack, name2unicode
from ..PSStackParser import PSStackParser
from ..PSStackParser import PSSyntaxError
from ..PSStackParser import PSEOF
from ..PSStackParser import PSLiteral
from ..PSStackParser import KWD
from ..PSStackParser import literal_name
log = logging.getLogger(__name__)
class CMapError(Exception):
pass
## CMapBase
##
class CMapBase(object):
debug = 0
def __init__(self, **kwargs):
self.attrs = kwargs.copy()
return
def is_vertical(self):
return self.attrs.get('WMode', 0) != 0
def set_attr(self, k, v):
self.attrs[k] = v
return
def add_code2cid(self, code, cid):
return
def add_cid2unichr(self, cid, code):
return
def use_cmap(self, cmap):
return
## CMap
##
class CMap(CMapBase):
def __init__(self, **kwargs):
CMapBase.__init__(self, **kwargs)
self.code2cid = {}
return
def __repr__(self):
return '<CMap: %s>' % self.attrs.get('CMapName')
def use_cmap(self, cmap):
assert isinstance(cmap, CMap), str(type(cmap))
def copy(dst, src):
for (k, v) in src.iteritems():
if isinstance(v, dict):
d = {}
dst[k] = d
copy(d, v)
else:
dst[k] = v
copy(self.code2cid, cmap.code2cid)
return
def decode(self, code):
log.debug('decode: %r, %r', self, code)
d = self.code2cid
for i in iter(code):
if i in d:
d = d[i]
if isinstance(d, int):
yield d
d = self.code2cid
else:
d = self.code2cid
return
def dump(self, out=sys.stdout, code2cid=None, code=None):
if code2cid is None:
code2cid = self.code2cid
code = ()
for (k, v) in sorted(code2cid.iteritems()):
c = code+(k,)
if isinstance(v, int):
out.write('code %r = cid %d\n' % (c, v))
else:
self.dump(out=out, code2cid=v, code=c)
return
## IdentityCMap
##
class IdentityCMap(CMapBase):
def decode(self, code):
n = len(code)//2
if n:
return struct.unpack('>%dH' % n, code)
else:
return ()
## UnicodeMap
##
class UnicodeMap(CMapBase):
def __init__(self, **kwargs):
CMapBase.__init__(self, **kwargs)
self.cid2unichr = {}
return
def __repr__(self):
return '<UnicodeMap: %s>' % self.attrs.get('CMapName')
def get_unichr(self, cid):
log.debug('get_unichr: %r, %r', self, cid)
return self.cid2unichr[cid]
def dump(self, out=sys.stdout):
for (k, v) in sorted(self.cid2unichr.items()):
out.write('cid %d = unicode %r\n' % (k, v))
return
## FileCMap
##
class FileCMap(CMap):
def add_code2cid(self, code, cid):
assert isinstance(code, str) and isinstance(cid, int), str((type(code), type(cid)))
d = self.code2cid
for c in code[:-1]:
c = ord(c)
if c in d:
d = d[c]
else:
t = {}
d[c] = t
d = t
c = ord(code[-1])
d[c] = cid
return
## FileUnicodeMap
##
class FileUnicodeMap(UnicodeMap):
def add_cid2unichr(self, cid, code):
assert isinstance(cid, int), str(type(cid))
if isinstance(code, PSLiteral):
# Interpret as an Adobe glyph name.
self.cid2unichr[cid] = name2unicode(code.name)
elif isinstance(code, bytes):
# Interpret as UTF-16BE.
self.cid2unichr[cid] = code.decode('UTF-16BE', 'ignore')
elif isinstance(code, int):
self.cid2unichr[cid] = chr(code)
else:
raise TypeError(code)
return
## PyCMap
##
class PyCMap(CMap):
def __init__(self, name, module):
CMap.__init__(self, CMapName=name)
self.code2cid = module.CODE2CID
if module.IS_VERTICAL:
self.attrs['WMode'] = 1
return
## PyUnicodeMap
##
class PyUnicodeMap(UnicodeMap):
def __init__(self, name, module, vertical):
UnicodeMap.__init__(self, CMapName=name)
if vertical:
self.cid2unichr = module.CID2UNICHR_V
self.attrs['WMode'] = 1
else:
self.cid2unichr = module.CID2UNICHR_H
return
## CMapDB
##
class CMapDB(object):
_cmap_cache = {}
_umap_cache = {}
class CMapNotFound(CMapError):
pass
@classmethod
def _load_data(klass, name):
name = name.replace("\0", "")
filename = '%s.pickle.gz' % name
log.info('loading: %r', name)
cmap_paths = (os.environ.get('CMAP_PATH', '/usr/share/pdfminer/'),
os.path.join(os.path.dirname(__file__), 'cmap'),)
for directory in cmap_paths:
path = os.path.join(directory, filename)
if os.path.exists(path):
gzfile = gzip.open(path)
try:
return type(str(name), (), pickle.loads(gzfile.read()))
finally:
gzfile.close()
else:
raise CMapDB.CMapNotFound(name)
@classmethod
def get_cmap(klass, name):
if name == 'Identity-H':
return IdentityCMap(WMode=0)
elif name == 'Identity-V':
return IdentityCMap(WMode=1)
try:
return klass._cmap_cache[name]
except KeyError:
pass
data = klass._load_data(name)
klass._cmap_cache[name] = cmap = PyCMap(name, data)
return cmap
@classmethod
def get_unicode_map(klass, name, vertical=False):
try:
return klass._umap_cache[name][vertical]
except KeyError:
pass
data = klass._load_data('to-unicode-%s' % name)
klass._umap_cache[name] = umaps = [PyUnicodeMap(name, data, v) for v in (False, True)]
return umaps[vertical]
## CMapParser
##
class CMapParser(PSStackParser):
def __init__(self, cmap, fp):
PSStackParser.__init__(self, fp)
self.cmap = cmap
# some ToUnicode maps don't have "begincmap" keyword.
self._in_cmap = True
return
def run(self):
try:
self.nextobject()
except PSEOF:
pass
return
KEYWORD_BEGINCMAP = KWD(b'begincmap')
KEYWORD_ENDCMAP = KWD(b'endcmap')
KEYWORD_USECMAP = KWD(b'usecmap')
KEYWORD_DEF = KWD(b'def')
KEYWORD_BEGINCODESPACERANGE = KWD(b'begincodespacerange')
KEYWORD_ENDCODESPACERANGE = KWD(b'endcodespacerange')
KEYWORD_BEGINCIDRANGE = KWD(b'begincidrange')
KEYWORD_ENDCIDRANGE = KWD(b'endcidrange')
KEYWORD_BEGINCIDCHAR = KWD(b'begincidchar')
KEYWORD_ENDCIDCHAR = KWD(b'endcidchar')
KEYWORD_BEGINBFRANGE = KWD(b'beginbfrange')
KEYWORD_ENDBFRANGE = KWD(b'endbfrange')
KEYWORD_BEGINBFCHAR = KWD(b'beginbfchar')
KEYWORD_ENDBFCHAR = KWD(b'endbfchar')
KEYWORD_BEGINNOTDEFRANGE = KWD(b'beginnotdefrange')
KEYWORD_ENDNOTDEFRANGE = KWD(b'endnotdefrange')
def do_keyword(self, pos, token):
if token is self.KEYWORD_BEGINCMAP:
self._in_cmap = True
self.popall()
return
elif token is self.KEYWORD_ENDCMAP:
self._in_cmap = False
return
if not self._in_cmap:
return
#
if token is self.KEYWORD_DEF:
try:
((_, k), (_, v)) = self.pop(2)
self.cmap.set_attr(literal_name(k), v)
except PSSyntaxError:
pass
return
if token is self.KEYWORD_USECMAP:
try:
((_, cmapname),) = self.pop(1)
self.cmap.use_cmap(CMapDB.get_cmap(literal_name(cmapname)))
except PSSyntaxError:
pass
except CMapDB.CMapNotFound:
pass
return
if token is self.KEYWORD_BEGINCODESPACERANGE:
self.popall()
return
if token is self.KEYWORD_ENDCODESPACERANGE:
self.popall()
return
if token is self.KEYWORD_BEGINCIDRANGE:
self.popall()
return
if token is self.KEYWORD_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, str((s1, e1))
for i in range(e1-s1+1):
x = sprefix+struct.pack('>L', s1+i)[-vlen:]
self.cmap.add_code2cid(x, cid+i)
return
if token is self.KEYWORD_BEGINCIDCHAR:
self.popall()
return
if token is self.KEYWORD_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.add_code2cid(code, nunpack(cid))
return
if token is self.KEYWORD_BEGINBFRANGE:
self.popall()
return
if token is self.KEYWORD_ENDBFRANGE:
objs = [obj for (__, obj) in self.popall()]
for (s, e, code) in choplist(3, objs):
if (not isinstance(s, bytes) or not isinstance(e, bytes) or
len(s) != len(e)):
continue
s1 = nunpack(s)
e1 = nunpack(e)
#assert s1 <= e1, str((s1, e1))
if isinstance(code, list):
for i in range(e1-s1+1):
self.cmap.add_cid2unichr(s1+i, code[i])
else:
var = code[-4:]
base = nunpack(var)
prefix = code[:-4]
vlen = len(var)
for i in range(e1-s1+1):
x = prefix+struct.pack('>L', base+i)[-vlen:]
self.cmap.add_cid2unichr(s1+i, x)
return
if token is self.KEYWORD_BEGINBFCHAR:
self.popall()
return
if token is self.KEYWORD_ENDBFCHAR:
objs = [obj for (__, obj) in self.popall()]
for (cid, code) in choplist(2, objs):
if isinstance(cid, bytes) and isinstance(code, bytes):
self.cmap.add_cid2unichr(nunpack(cid), code)
return
if token is self.KEYWORD_BEGINNOTDEFRANGE:
self.popall()
return
if token is self.KEYWORD_ENDNOTDEFRANGE:
self.popall()
return
self.push((pos, token))
return
# test
def main(argv):
args = argv[1:]
for fname in args:
fp = open(fname, 'rb')
cmap = FileUnicodeMap()
#cmap = FileCMap()
CMapParser(cmap, fp).run()
fp.close()
cmap.dump()
return
if __name__ == '__main__':
sys.exit(main(sys.argv))
@@ -0,0 +1,30 @@
from typing import List
from pdfmajor.utils import MATRIX_IDENTITY, mult_matrix
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFStream import PDFStream, list_value, dict_value
from pdfmajor.parser.constants import LITERAL_FORM, LITERAL_IMAGE
from .state import PDFStateStack, PDFColorSpace
from .state.Curves import CurveMethod, CurvePath, CurvePoint
class PDFCommands:
commands = {}
class InvalidOperation(Exception): pass
class RepeatedCommand(Exception): pass
@classmethod
def __add(cls, cmd_name: str, cmd: callable):
if not (cmd_name in cls.commands.keys()):
cls.commands[cmd_name] = cmd
return cmd
raise cls.RepeatedCommand(cmd_name)
@classmethod
def add(cls, *cmd_names: List[str]):
def wrap(cmd: callable):
for cmd_name in cmd_names:
cls.__add(cmd_name, cmd)
return cmd
return wrap
@@ -0,0 +1,9 @@
from .commands import PDFCommands
from .state import PDFStateStack, PDFGraphicState, PDFTextState
from .state.PDFGraphicState import PDFColor, PDFColorSpace, PREDEFINED_COLORSPACE
from .state import PDFFont, get_font
from .state import LTCharBlock, LTChar
from .state import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
from .state import LTImage
from .state import LTXObject
from .state import LTItem, LTComponent, LTContainer
+485
View File
@@ -0,0 +1,485 @@
from pdfmajor.utils import mult_matrix, MATRIX_IDENTITY, apply_matrix_pt
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFStream import PDFStream, list_value, dict_value
from pdfmajor.parser.constants import LITERAL_FORM, LITERAL_IMAGE
from .PDFCommands import PDFCommands
from .state import PDFStateStack, PDFGraphicState
from .state.Curves import CurveMethod, CurvePath, CurvePoint
# from .state.PDFItem import PDFImage, PDFShape, PDFText, PDFXObject
from .state import make_char_block, make_curve, make_image, make_xobject
# No Support
@PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs')
def do_no_support(stack: PDFStateStack) -> PDFStateStack:
return stack
# marked content operators
@PDFCommands.add('MP', 'DP')
def do_MP(stack: PDFStateStack) -> PDFStateStack:
# self.device.do_tag(tag)
return stack
@PDFCommands.add('BMC', 'BDC')
def do_BMC(stack: PDFStateStack) -> PDFStateStack:
# self.device.begin_tag(tag)
return stack
@PDFCommands.add('EMC')
def do_EMC(stack: PDFStateStack) -> PDFStateStack:
# self.device.end_tag(tag)
return stack
@PDFCommands.add('q')
def do_q(stack: PDFStateStack) -> PDFStateStack:
stack.graphicstack.append(stack.graphics)
stack.graphics = PDFGraphicState()
return stack
@PDFCommands.add('Q')
def do_Q(stack: PDFStateStack) -> PDFStateStack:
if len(stack.graphicstack) > 0:
stack.graphicstate = stack.graphicstack.pop()
return stack
# concat-matrix
@PDFCommands.add('cm')
def do_cm(stack: PDFStateStack, a1, b1, c1, d1, e1, f1) -> PDFStateStack:
stack.t_matrix = mult_matrix((a1, b1, c1, d1, e1, f1), stack.t_matrix)
return stack
# setlinewidth
@PDFCommands.add('w')
def do_w(stack: PDFStateStack, linewidth) -> PDFStateStack:
stack.graphics.linewidth = linewidth
return stack
# setlinecap
@PDFCommands.add('J')
def do_J(stack: PDFStateStack, linecap) -> PDFStateStack:
stack.graphics.linecap = linecap
return stack
# setlinejoin
@PDFCommands.add('j')
def do_j(stack: PDFStateStack, linejoin) -> PDFStateStack:
stack.graphics.linejoin = linejoin
return stack
# setmiterlimit
@PDFCommands.add('M')
def do_M(stack: PDFStateStack, miterlimit) -> PDFStateStack:
stack.graphics.miterlimit = miterlimit
return stack
# setdash
@PDFCommands.add('d')
def do_d(stack: PDFStateStack, dash, phase) -> PDFStateStack:
stack.graphics.dash = (dash, phase)
return stack
# setintent
@PDFCommands.add('ri')
def do_ri(stack: PDFStateStack, intent) -> PDFStateStack:
stack.graphics.intent = intent
return stack
# setflatness
@PDFCommands.add('i')
def do_i(stack: PDFStateStack, flatness) -> PDFStateStack:
stack.graphics.flatness = flatness
return stack
# moveto
@PDFCommands.add('m')
def do_m(stack: PDFStateStack, x: float, y: float) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.MOVE_TO,
CurvePoint(x, y)
))
return stack
# lineto
@PDFCommands.add('l')
def do_l(stack: PDFStateStack, x: float, y: float) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.LINE_TO,
CurvePoint(x, y)
))
return stack
# curveto
@PDFCommands.add('c')
def do_c(stack: PDFStateStack,
x1: float, y1: float,
x2: float, y2: float,
x3: float, y3: float ) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.CURVE_BOTH_TO,
CurvePoint(x1, y1),
CurvePoint(x2, y2),
CurvePoint(x3, y3),
))
return stack
# urveto
@PDFCommands.add('v')
def do_v(stack: PDFStateStack,
x2: float, y2: float,
x3: float, y3: float) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.CURVE_NEXT_TO,
CurvePoint(x2, y2),
CurvePoint(x3, y3),
))
return stack
# rveto
@PDFCommands.add('y')
def do_y(stack: PDFStateStack,
x1: float, y1: float,
x3: float, y3: float) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.CURVE_FIRST_TO,
CurvePoint(x1, y1),
CurvePoint(x3, y3)
))
return stack
# closepath
@PDFCommands.add('h')
def do_h(stack: PDFStateStack) -> PDFStateStack:
stack.curvestacks.append(CurvePath(CurvePath.METHOD.CLOSE_PATH))
return stack
# rectangle
@PDFCommands.add('re')
def do_re(stack: PDFStateStack, x, y, w, h) -> PDFStateStack:
stack.curvestacks.extend([
CurvePath(CurvePath.METHOD.MOVE_TO, CurvePoint(x, y) ),
CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x+w, y) ),
CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x+w, y+h) ),
CurvePath(CurvePath.METHOD.LINE_TO, CurvePoint(x, y+h) ),
CurvePath(CurvePath.METHOD.CLOSE_PATH)
])
return stack
# stroke, fill
@PDFCommands.add('S', 'f', 'F', 'B')
def do_complete_path(stack: PDFStateStack) -> PDFStateStack:
stack.complete_layout_items.append(make_curve(
stack.t_matrix,
stack.graphics.copy(),
False,
stack.curvestacks
))
stack.curvestacks = []
return stack
# sroke, fill-even-odd
@PDFCommands.add('f_a', 'B_a')
def do_complete_path_evenodd(stack: PDFStateStack) -> PDFStateStack:
stack.complete_layout_items.append(make_curve(
stack.t_matrix,
stack.graphics.copy(),
True,
stack.curvestacks
))
stack.curvestacks = []
return stack
# close-and-stroke (do_h, do_S)
@PDFCommands.add('s', 'b')
def do_close_complete(stack: PDFStateStack) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.CLOSE_PATH
))
stack.complete_layout_items.append(make_curve(
stack.t_matrix,
stack.graphics.copy(),
False,
stack.curvestacks
))
stack.curvestacks = []
return stack
# close-fill-and-stroke-even-odd
@PDFCommands.add('b_a')
def do_close_complete_evenodd(stack: PDFStateStack) -> PDFStateStack:
stack.curvestacks.append(CurvePath(
CurvePath.METHOD.CLOSE_PATH
))
stack.complete_layout_items.append(make_curve(
stack.t_matrix,
stack.graphics.copy(),
True,
stack.curvestacks
))
stack.curvestacks = []
return stack
# close-only (empty path)
@PDFCommands.add('n')
def do_n(stack: PDFStateStack) -> PDFStateStack:
stack.curvestacks = []
return stack
# setcolorspace-stroking
@PDFCommands.add('CS')
def do_CS(stack: PDFStateStack, name: str) -> PDFStateStack:
try:
stack.graphics.set_stroke_color(stack.colorspace_map[literal_name(name)])
return stack
# stack.set_scolor()
except KeyError:
raise PDFCommands.InvalidOperation('Undefined ColorSpace: %r' % name)
# setcolorspace-non-strokine
@PDFCommands.add('cs')
def do_cs(stack: PDFStateStack, name: str) -> PDFStateStack:
try:
stack.graphics.set_nostroke_color(stack.colorspace_map[literal_name(name)])
return stack
# stack.set_scolor()
except KeyError:
raise PDFCommands.InvalidOperation('Undefined ColorSpace: %r' % name)
# setgray-stroking
@PDFCommands.add('G')
def do_G(stack: PDFStateStack, gray: float) -> PDFStateStack:
stack.graphics.set_stroke_color(
stack.colorspace_map['DeviceGray'],
gray
)
return stack
# setgray-non-stroking
@PDFCommands.add('g')
def do_g(stack: PDFStateStack, gray: float) -> PDFStateStack:
stack.graphics.set_nostroke_color(
stack.colorspace_map['DeviceGray'],
gray
)
return stack
# setrgb-stroking
@PDFCommands.add('RG')
def do_RG(stack: PDFStateStack, r, g, b) -> PDFStateStack:
stack.graphics.set_stroke_color(
stack.colorspace_map['DeviceRGB'],
round(r*255),
round(g*255),
round(b*255)
)
return stack
# setrgb-non-stroking
@PDFCommands.add('rg')
def do_rg(stack: PDFStateStack, r, g, b) -> PDFStateStack:
stack.graphics.set_nostroke_color(
stack.colorspace_map['DeviceRGB'],
round(r*255),
round(g*255),
round(b*255)
)
return stack
# setcmyk-stroking
@PDFCommands.add('K')
def do_K(stack: PDFStateStack, c, m, y, k) -> PDFStateStack:
stack.graphics.set_stroke_color(
stack.colorspace_map['DeviceCMYK'],
c, m, y, k
)
return stack
# setcmyk-non-stroking
@PDFCommands.add('k')
def do_k(stack: PDFStateStack, c, m, y, k) -> PDFStateStack:
stack.graphics.set_stroke_color(
stack.colorspace_map['DeviceCMYK'],
c, m, y, k
)
return stack
@PDFCommands.add('SCN','SC')
def do_SCN(stack: PDFStateStack) -> PDFStateStack:
if stack.graphics.scolor.color_space is None:
raise PDFCommands.InvalidOperation('No colorspace specified!')
components = stack.pop(stack.graphics.scolor.color_space.ncomponents)
stack.graphics.set_stroke_color(
stack.graphics.scolor.color_space,
*components
)
return stack
@PDFCommands.add('scn','sc')
def do_scn(stack: PDFStateStack) -> PDFStateStack:
if stack.graphics.ncolor.color_space is None:
raise PDFCommands.InvalidOperation('No colorspace specified!')
components = stack.pop(stack.graphics.ncolor.color_space.ncomponents)
stack.graphics.set_nostroke_color(
stack.graphics.ncolor.color_space,
*components
)
return stack
# begin-text
@PDFCommands.add('BT')
def do_BT(stack: PDFStateStack) -> PDFStateStack:
stack.text.matrix = MATRIX_IDENTITY
stack.text.linematrix = [0, 0]
return stack
# setcharspace
@PDFCommands.add('Tc')
def do_Tc(stack: PDFStateStack, charspace) -> PDFStateStack:
stack.text.charspace = charspace
return stack
# setwordspace
@PDFCommands.add('Tw')
def do_Tw(stack: PDFStateStack, wordspace) -> PDFStateStack:
stack.text.wordspace = wordspace
return stack
# textscale
@PDFCommands.add('Tz')
def do_Tz(stack: PDFStateStack, scaling) -> PDFStateStack:
stack.text.scaling = scaling
return stack
# setleading
@PDFCommands.add('TL')
def do_TL(stack: PDFStateStack, leading) -> PDFStateStack:
stack.text.leading = -leading
return stack
# selectfont
@PDFCommands.add('Tf')
def do_Tf(stack: PDFStateStack, fontid, fontsize) -> PDFStateStack:
stack.text.fontsize = fontsize
try:
stack.text.font = stack.fontmap[literal_name(fontid)]
return stack
except KeyError:
raise PDFCommands.InvalidOperation('Undefined Font id: %r' % fontid)
# setrendering
@PDFCommands.add('Tr')
def do_Tr(stack: PDFStateStack, render) -> PDFStateStack:
stack.text.render = render
return stack
# settextrise
@PDFCommands.add('Ts')
def do_Ts(stack: PDFStateStack, rise) -> PDFStateStack:
stack.text.rise = rise
return stack
# text-move
@PDFCommands.add('Td')
def do_Td(stack: PDFStateStack, tx, ty) -> PDFStateStack:
(a, b, c, d, e, f) = stack.text.matrix
stack.text.matrix = (a, b, c, d, tx*a+ty*c+e, tx*b+ty*d+f)
stack.text.linematrix = [0, 0]
return stack
# text-move
@PDFCommands.add('TD')
def do_TD(stack: PDFStateStack, tx, ty) -> PDFStateStack:
(a, b, c, d, e, f) = stack.text.matrix
stack.text.matrix = (a, b, c, d, tx*a+ty*c+e, tx*b+ty*d+f)
stack.text.leading = ty
stack.text.linematrix = [0, 0]
return stack
# textmatrix
@PDFCommands.add('Tm')
def do_Tm(stack: PDFStateStack, a, b, c, d, e, f) -> PDFStateStack:
stack.text.matrix = (a, b, c, d, e, f)
stack.text.linematrix = [0, 0]
return stack
# nextline
@PDFCommands.add('T_a')
def do_T_a(stack: PDFStateStack) -> PDFStateStack:
(a, b, c, d, e, f) = stack.text.matrix
stack.text.matrix = (a, b, c, d, stack.text.leading*c+e, stack.text.leading*d+f)
stack.text.linematrix = [0, 0]
return stack
# show-pos
@PDFCommands.add('TJ')
def do_TJ(stack: PDFStateStack, seq: bytearray) -> PDFStateStack:
if stack.text.font is None:
raise PDFCommands.InvalidOperation("No Font Specified")
stack.complete_layout_items.append(make_char_block(
seq,
stack.t_matrix,
stack.text.copy(),
stack.graphics.ncolor.copy()
))
return stack
# show
@PDFCommands.add('Tj')
def do_Tj(stack: PDFStateStack, b: bytes) -> PDFStateStack:
return do_TJ(stack, [b])
# quote
@PDFCommands.add('_q')
def do__q(stack: PDFStateStack, b: bytes) -> PDFStateStack:
return do_TJ(do_T_a(stack), [b])
# doublequote
@PDFCommands.add('_w')
def do__w(stack: PDFStateStack, wordspace, charspace, b) -> PDFStateStack:
stack.text.wordspace = wordspace
stack.text.charspace = charspace
return do_TJ(stack, [b])
@PDFCommands.add('EI')
def do_EI(stack: PDFStateStack, obj) -> PDFStateStack:
if 'W' in obj and 'H' in obj:
stack.complete_layout_items.append(make_image(
obj, stack.ctm
))
return stack
# invoke an XObject
@PDFCommands.add('Do')
def do_Do(stack: PDFStateStack, xobjid) -> PDFStateStack:
xobjid = literal_name(xobjid)
try:
xobj = PDFStream.validated_stream(stack.xobjmap[xobjid])
except KeyError:
raise PDFCommands.InvalidOperation('Undefined xobject id: %r' % xobjid)
# log.info('Processing xobj: %r', xobj)
subtype = xobj.get('Subtype')
if subtype is LITERAL_FORM and 'BBox' in xobj:
bbox = list_value(xobj['BBox'])
matrix = list_value(xobj.get('Matrix', MATRIX_IDENTITY))
# According to PDF reference 1.7 section 4.9.1, XObjects in
# earlier PDFs (prior to v1.2) use the page's Resources entry
# instead of having their own Resources entry.
xobjres = xobj.get('Resources')
resources = dict_value(xobjres) if xobjres else stack.resources.copy()
stack.complete_layout_items.append(make_xobject(
obj=xobj,
bbox=bbox,
ctm=stack.t_matrix,
matrix=matrix,
resources=resources
))
elif subtype is LITERAL_IMAGE and 'Width' in xobj and 'Height' in xobj:
stack.complete_layout_items.append(make_image(
obj, stack.ctm
))
else:
# unsupported xobject type.
pass
return stack
File renamed without changes.
@@ -0,0 +1,16 @@
from typing import List
from .PDFColorSpace import PDFColorSpace
class PDFColor:
def __init__(self, color_space: PDFColorSpace, *values: List[float]):
self.color_space: PDFColorSpace = color_space
self.values: list = values
def copy(self):
return self.__class__(
self.color_space,
*self.values
)
def __repr__(self):
return f"<PDFColor space={self.color_space} values={self.values}/>"
@@ -1,12 +1,4 @@
import collections
from .PSStackParser import LIT
## PDFColorSpace
##
LITERAL_DEVICE_GRAY = LIT('DeviceGray')
LITERAL_DEVICE_RGB = LIT('DeviceRGB')
LITERAL_DEVICE_CMYK = LIT('DeviceCMYK')
from collections import OrderedDict
class PDFColorSpace(object):
@@ -19,7 +11,7 @@ class PDFColorSpace(object):
return '<PDFColorSpace: %s, ncomponents=%d>' % (self.name, self.ncomponents)
PREDEFINED_COLORSPACE = collections.OrderedDict()
PREDEFINED_COLORSPACE = OrderedDict()
for (name, n) in [
('DeviceGray', 1), # default value first
@@ -0,0 +1,49 @@
from .PDFColor import PDFColor, PDFColorSpace
from .PDFColorSpace import PREDEFINED_COLORSPACE
class PDFGraphicState(object):
class InvalidOperation(Exception): pass
def __init__(self):
self.linewidth = 0
self.linecap = None
self.linejoin = None
self.miterlimit = None
self.dash = None
self.intent = None
self.flatness = None
self.scolor:PDFColor = PDFColor(None)
self.ncolor:PDFColor = PDFColor(None)
def set_stroke_color(self, colspace: PDFColorSpace, *values):
if self.scolor is None:
self.scolor = PDFColor(colspace, *values)
def set_nostroke_color(self, colspace: PDFColorSpace, *values):
if self.ncolor is None:
self.ncolor = PDFColor(colspace, *values)
def copy(self):
obj = PDFGraphicState()
obj.linewidth = self.linewidth
obj.linecap = self.linecap
obj.linejoin = self.linejoin
obj.miterlimit = self.miterlimit
obj.dash = self.dash
obj.intent = self.intent
obj.flatness = self.flatness
obj.scolor = self.scolor.copy() if self.scolor is not None else None
obj.ncolor = self.ncolor.copy() if self.ncolor is not None else None
return obj
def __repr__(self):
return ('<PDFGraphicState linewidth=%r, linecap=%r, linejoin=%r, '
' miterlimit=%r, dash=%r, intent=%r, flatness=%r, '
' stroking color=%r, non stroking color=%r'
'>' %
(
self.linewidth, self.linecap, self.linejoin,
self.miterlimit, self.dash, self.intent, self.flatness,
str(self.scolor), str(self.ncolor)
)
)
@@ -0,0 +1,6 @@
from pdfmajor.parser.PDFStream import PDFStream
from ._base import PDFItem
class PDFImage(PDFItem):
def __init__(self, stream: PDFStream):
self.stream = stream
@@ -0,0 +1,12 @@
from typing import List
from ..PDFGraphicState import PDFGraphicState, PDFColorSpace
from ..Curves import CurvePath
from ._base import PDFItem
class PDFShape(PDFItem):
def __init__(self, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]):
self.gstate = gstate
self.evenodd = evenodd
self.paths = paths
@@ -1,96 +1,51 @@
# -*- coding: utf-8 -*-
from typing import List
from .PDFFont import PDFUnicodeNotDefined
from pdfmajor.utils import mult_matrix, isnumber, apply_matrix_pt, translate_matrix
from pdfmajor.utils import Bbox
from ..utils import isnumber, apply_matrix_pt
from ..utils import Bbox, mult_matrix, translate_matrix
from ..PDFTextState import PDFTextState
from ..PDFGraphicState import PDFGraphicState, PDFColorSpace
from ._base import PDFItem
from .PDFResourceManager import PDFResourceManager
from .PDFGraphicState import PDFGraphicState
from .PDFTextState import PDFTextState
from .PDFColorSpace import PDFColorSpace
from .PDFFont import PDFFont
## PDFDevice
##
class PDFDevice(object):
def __init__(self, rsrcmgr: PDFResourceManager):
self.rsrcmgr = rsrcmgr
self.ctm = None
return
class Char:
def __init__(self, char: str, bbox: Bbox):
self.char = char
self.bbox = bbox
def __str__(self):
return '"%s" %s' % (self.char, self.bbox)
def __repr__(self):
return '<PDFDevice>'
return f"<Char:{str(self)}/>"
def __enter__(self):
return self
class PDFText(PDFItem):
def __init__(self, chars: List[Char], textstate: PDFTextState, graphicstate: PDFGraphicState):
self.chars = chars
self.tstate = textstate
self.gstate = graphicstate
def __exit__(self, exc_type, exc_val, exc_tb):
self.close()
def close(self):
return
def set_ctm(self, ctm):
self.ctm = ctm
return
def begin_tag(self, tag, props=None):
return
def end_tag(self):
return
def do_tag(self, tag, props=None):
return
def begin_page(self, page, ctm):
return
def end_page(self, page):
return
def begin_figure(self, name, bbox: Bbox, matrix):
return
def end_figure(self, name):
return
def paint_path(self, graphicstate: PDFGraphicState, stroke, fill, evenodd, path):
return
def render_image(self, name, stream):
return
def render_string(self, textstate, seq, ncs, graphicstate: PDFGraphicState):
return
## PDFTextDevice
##
class PDFTextDevice(PDFDevice):
def render_string(self, textstate: PDFTextState, seq: bytearray, ncs: PDFColorSpace, graphicstate: PDFGraphicState):
textstate.matrix = mult_matrix(textstate.matrix, self.ctm)
@classmethod
def make(cls, seq: bytearray, ctm, textstate: PDFTextState, graphicstate: PDFGraphicState):
textstate.matrix = mult_matrix(textstate.matrix, ctm)
textstate.scaling = textstate.scaling * .01
textstate.charspace = textstate.charspace * textstate.scaling
textstate.wordspace = textstate.wordspace * textstate.scaling
ncs = graphicstate.ncolor.color_space
if textstate.font.is_multibyte():
textstate.wordspace = 0
dxscale = .001 * textstate.fontsize * textstate.scaling
if textstate.font.is_vertical():
self.__render_string_along(1,
return cls.__render_string_along(1,
seq, textstate, dxscale, ncs, graphicstate
)
else:
self.__render_string_along(0,
return cls.__render_string_along(0,
seq, textstate, dxscale, ncs, graphicstate
)
def __render_string_along(self, idx: int, seq: bytearray,
@classmethod
def __render_string_along(cls, idx: int, seq: bytearray,
textstate: PDFTextState, dxscale: float, ncs: PDFColorSpace, graphicstate: PDFGraphicState):
needcharspace = False
char_meta_datas = []
@@ -107,19 +62,20 @@ class PDFTextDevice(PDFDevice):
assert isinstance(text, str), str(type(text))
matrix = translate_matrix(textstate.matrix, textstate.linematrix)
adv, bbox = self.__compute_char_bbox(matrix, cid, textstate)
adv, bbox = cls.__compute_char_bbox(matrix, cid, textstate)
textstate.linematrix[idx] += adv
char_meta_datas.append(
( text, bbox, )
Char( text, Bbox.from_points(bbox), )
)
if cid == 32 and textstate.wordspace:
textstate.linematrix[idx] += textstate.wordspace
needcharspace = True
self.render_char_box(char_meta_datas, textstate, graphicstate, ncs)
return PDFText(char_meta_datas, textstate, graphicstate)
def __compute_char_bbox(self, matrix, char_id: int, textstate: PDFTextState):
@classmethod
def __compute_char_bbox(cls, matrix, char_id: int, textstate: PDFTextState):
font = textstate.font
fontsize = textstate.fontsize
rise = textstate.rise
@@ -157,7 +113,4 @@ class PDFTextDevice(PDFDevice):
if y1 < y0:
(y0, y1) = (y1, y0)
return (adv, ((x0, y0), (x1, y1)))
def render_char_box(self, chars: list, textstate: PDFTextState, graphicstate: PDFGraphicState, ncs: PDFColorSpace):
pass
return (adv, ((x0, y0), (x1, y1)))
@@ -0,0 +1,9 @@
from ._base import PDFItem
class PDFXObject(PDFItem):
def __init__(self, xobj_strem, bbox, resources, ctm):
self.stream = xobj_strem
self.bbox = bbox
self.resources = resources
self.ctm = ctm
self.children = []
@@ -0,0 +1,100 @@
from pdfmajor.utils import mult_matrix, isnumber, apply_matrix_pt, translate_matrix
from ..PDFTextState import PDFTextState
from ..PDFGraphicState import PDFGraphicState, PDFColorSpace
from .PDFItem import PDFText
from .Char import Char, Bbox
class TextDecoder:
@classmethod
def make_pdftext(cls, seq: bytearray, ctm, textstate: PDFTextState, graphicstate: PDFGraphicState) -> PDFText:
textstate.matrix = mult_matrix(textstate.matrix, ctm)
textstate.scaling = textstate.scaling * .01
textstate.charspace = textstate.charspace * textstate.scaling
textstate.wordspace = textstate.wordspace * textstate.scaling
ncs = graphicstate.ncolor.color_space
if textstate.font.is_multibyte():
textstate.wordspace = 0
dxscale = .001 * textstate.fontsize * textstate.scaling
if textstate.font.is_vertical():
return cls.__render_string_along(1,
seq, textstate, dxscale, ncs, graphicstate
)
else:
return cls.__render_string_along(0,
seq, textstate, dxscale, ncs, graphicstate
)
@classmethod
def __render_string_along(cls, idx: int, seq: bytearray,
textstate: PDFTextState, dxscale: float, ncs: PDFColorSpace, graphicstate: PDFGraphicState) -> PDFText:
needcharspace = False
char_meta_datas = []
for obj in seq:
if isnumber(obj):
textstate.linematrix[idx] -= obj*dxscale
needcharspace = True
else:
for cid in textstate.font.decode(obj):
if needcharspace:
textstate.linematrix[idx] += textstate.charspace
text = textstate.font.to_unichr(cid)
assert isinstance(text, str), str(type(text))
matrix = translate_matrix(textstate.matrix, textstate.linematrix)
adv, bbox = cls.__compute_char_bbox(matrix, cid, textstate)
textstate.linematrix[idx] += adv
char_meta_datas.append(
Char( text, Bbox.from_points(bbox), )
)
if cid == 32 and textstate.wordspace:
textstate.linematrix[idx] += textstate.wordspace
needcharspace = True
return PDFText(char_meta_datas, textstate, graphicstate)
@classmethod
def __compute_char_bbox(cls, matrix, char_id: int, textstate: PDFTextState):
font = textstate.font
fontsize = textstate.fontsize
rise = textstate.rise
scaling = textstate.scaling
textwidth = font.char_width(char_id)
textdisp = font.char_disp(char_id)
adv = textwidth * fontsize * scaling
# compute the boundary rectangle.
if font.is_vertical():
# vertical
width = font.get_width() * fontsize
(vx, vy) = textdisp
if vx is None:
vx = width * 0.5
else:
vx = vx * fontsize * .001
vy = (1000 - vy) * fontsize * .001
tx = -vx
ty = vy + rise
bll = (tx, ty+adv)
bur = (tx+width, ty)
else:
# horizontal
height = font.get_height() * fontsize
descent = font.get_descent() * fontsize
ty = descent + rise
bll = (0, ty)
bur = (adv, ty+height)
(x0, y0) = apply_matrix_pt(matrix, bll)
(x1, y1) = apply_matrix_pt(matrix, bur)
if x1 < x0:
(x0, x1) = (x1, x0)
if y1 < y0:
(y0, y1) = (y1, y0)
return (adv, ((x0, y0), (x1, y1)))
@@ -0,0 +1,5 @@
from ._base import PDFItem
from .PDFShape import PDFShape
from .PDFImage import PDFImage
from .PDFText import PDFText
from .PDFXObject import PDFXObject
@@ -0,0 +1,6 @@
class PDFItem:
def __repr__(self) -> str:
return f"""<{self.__class__.__name__} {' '.join([
f'{name}="{val}"'
for name, val in vars(self).items()
])}/>"""
@@ -0,0 +1,246 @@
## CFFFont
## (Format specified in Adobe Technical Note: #5176
## "The Compact Font Format Specification")
##
import struct
from io import BytesIO
from pdfmajor.utils import nunpack
NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-')
def getdict(data):
d = {}
fp = BytesIO(data)
stack = []
while 1:
c = fp.read(1)
if not c:
break
b0 = ord(c)
if b0 <= 21:
d[b0] = stack
stack = []
continue
if b0 == 30:
s = ''
loop = True
while loop:
b = ord(fp.read(1))
for n in (b >> 4, b & 15):
if n == 15:
loop = False
else:
s += NIBBLES[n]
value = float(s)
elif 32 <= b0 and b0 <= 246:
value = b0-139
else:
b1 = ord(fp.read(1))
if 247 <= b0 and b0 <= 250:
value = ((b0-247) << 8)+b1+108
elif 251 <= b0 and b0 <= 254:
value = -((b0-251) << 8)-b1-108
else:
b2 = ord(fp.read(1))
if 128 <= b1:
b1 -= 256
if b0 == 28:
value = b1 << 8 | b2
else:
value = b1 << 24 | b2 << 16 | struct.unpack('>H', fp.read(2))[0]
stack.append(value)
return d
class CFFFont(object):
STANDARD_STRINGS = (
'.notdef', 'space', 'exclam', 'quotedbl', 'numbersign',
'dollar', 'percent', 'ampersand', 'quoteright', 'parenleft',
'parenright', 'asterisk', 'plus', 'comma', 'hyphen', 'period',
'slash', 'zero', 'one', 'two', 'three', 'four', 'five', 'six',
'seven', 'eight', 'nine', 'colon', 'semicolon', 'less', 'equal',
'greater', 'question', 'at', 'A', 'B', 'C', 'D', 'E', 'F', 'G',
'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T',
'U', 'V', 'W', 'X', 'Y', 'Z', 'bracketleft', 'backslash',
'bracketright', 'asciicircum', 'underscore', 'quoteleft', 'a',
'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n',
'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
'braceleft', 'bar', 'braceright', 'asciitilde', 'exclamdown',
'cent', 'sterling', 'fraction', 'yen', 'florin', 'section',
'currency', 'quotesingle', 'quotedblleft', 'guillemotleft',
'guilsinglleft', 'guilsinglright', 'fi', 'fl', 'endash',
'dagger', 'daggerdbl', 'periodcentered', 'paragraph', 'bullet',
'quotesinglbase', 'quotedblbase', 'quotedblright',
'guillemotright', 'ellipsis', 'perthousand', 'questiondown',
'grave', 'acute', 'circumflex', 'tilde', 'macron', 'breve',
'dotaccent', 'dieresis', 'ring', 'cedilla', 'hungarumlaut',
'ogonek', 'caron', 'emdash', 'AE', 'ordfeminine', 'Lslash',
'Oslash', 'OE', 'ordmasculine', 'ae', 'dotlessi', 'lslash',
'oslash', 'oe', 'germandbls', 'onesuperior', 'logicalnot', 'mu',
'trademark', 'Eth', 'onehalf', 'plusminus', 'Thorn',
'onequarter', 'divide', 'brokenbar', 'degree', 'thorn',
'threequarters', 'twosuperior', 'registered', 'minus', 'eth',
'multiply', 'threesuperior', 'copyright', 'Aacute',
'Acircumflex', 'Adieresis', 'Agrave', 'Aring', 'Atilde',
'Ccedilla', 'Eacute', 'Ecircumflex', 'Edieresis', 'Egrave',
'Iacute', 'Icircumflex', 'Idieresis', 'Igrave', 'Ntilde',
'Oacute', 'Ocircumflex', 'Odieresis', 'Ograve', 'Otilde',
'Scaron', 'Uacute', 'Ucircumflex', 'Udieresis', 'Ugrave',
'Yacute', 'Ydieresis', 'Zcaron', 'aacute', 'acircumflex',
'adieresis', 'agrave', 'aring', 'atilde', 'ccedilla', 'eacute',
'ecircumflex', 'edieresis', 'egrave', 'iacute', 'icircumflex',
'idieresis', 'igrave', 'ntilde', 'oacute', 'ocircumflex',
'odieresis', 'ograve', 'otilde', 'scaron', 'uacute',
'ucircumflex', 'udieresis', 'ugrave', 'yacute', 'ydieresis',
'zcaron', 'exclamsmall', 'Hungarumlautsmall', 'dollaroldstyle',
'dollarsuperior', 'ampersandsmall', 'Acutesmall',
'parenleftsuperior', 'parenrightsuperior', 'twodotenleader',
'onedotenleader', 'zerooldstyle', 'oneoldstyle', 'twooldstyle',
'threeoldstyle', 'fouroldstyle', 'fiveoldstyle', 'sixoldstyle',
'sevenoldstyle', 'eightoldstyle', 'nineoldstyle',
'commasuperior', 'threequartersemdash', 'periodsuperior',
'questionsmall', 'asuperior', 'bsuperior', 'centsuperior',
'dsuperior', 'esuperior', 'isuperior', 'lsuperior', 'msuperior',
'nsuperior', 'osuperior', 'rsuperior', 'ssuperior', 'tsuperior',
'ff', 'ffi', 'ffl', 'parenleftinferior', 'parenrightinferior',
'Circumflexsmall', 'hyphensuperior', 'Gravesmall', 'Asmall',
'Bsmall', 'Csmall', 'Dsmall', 'Esmall', 'Fsmall', 'Gsmall',
'Hsmall', 'Ismall', 'Jsmall', 'Ksmall', 'Lsmall', 'Msmall',
'Nsmall', 'Osmall', 'Psmall', 'Qsmall', 'Rsmall', 'Ssmall',
'Tsmall', 'Usmall', 'Vsmall', 'Wsmall', 'Xsmall', 'Ysmall',
'Zsmall', 'colonmonetary', 'onefitted', 'rupiah', 'Tildesmall',
'exclamdownsmall', 'centoldstyle', 'Lslashsmall', 'Scaronsmall',
'Zcaronsmall', 'Dieresissmall', 'Brevesmall', 'Caronsmall',
'Dotaccentsmall', 'Macronsmall', 'figuredash', 'hypheninferior',
'Ogoneksmall', 'Ringsmall', 'Cedillasmall', 'questiondownsmall',
'oneeighth', 'threeeighths', 'fiveeighths', 'seveneighths',
'onethird', 'twothirds', 'zerosuperior', 'foursuperior',
'fivesuperior', 'sixsuperior', 'sevensuperior', 'eightsuperior',
'ninesuperior', 'zeroinferior', 'oneinferior', 'twoinferior',
'threeinferior', 'fourinferior', 'fiveinferior', 'sixinferior',
'seveninferior', 'eightinferior', 'nineinferior',
'centinferior', 'dollarinferior', 'periodinferior',
'commainferior', 'Agravesmall', 'Aacutesmall',
'Acircumflexsmall', 'Atildesmall', 'Adieresissmall',
'Aringsmall', 'AEsmall', 'Ccedillasmall', 'Egravesmall',
'Eacutesmall', 'Ecircumflexsmall', 'Edieresissmall',
'Igravesmall', 'Iacutesmall', 'Icircumflexsmall',
'Idieresissmall', 'Ethsmall', 'Ntildesmall', 'Ogravesmall',
'Oacutesmall', 'Ocircumflexsmall', 'Otildesmall',
'Odieresissmall', 'OEsmall', 'Oslashsmall', 'Ugravesmall',
'Uacutesmall', 'Ucircumflexsmall', 'Udieresissmall',
'Yacutesmall', 'Thornsmall', 'Ydieresissmall', '001.000',
'001.001', '001.002', '001.003', 'Black', 'Bold', 'Book',
'Light', 'Medium', 'Regular', 'Roman', 'Semibold',
)
class INDEX(object):
def __init__(self, fp):
self.fp = fp
self.offsets = []
(count, offsize) = struct.unpack('>HB', self.fp.read(3))
for _ in range(count+1):
self.offsets.append(nunpack(self.fp.read(offsize)))
self.base = self.fp.tell()-1
self.fp.seek(self.base+self.offsets[-1])
return
def __repr__(self):
return '<INDEX: size=%d>' % len(self)
def __len__(self): # pylint: disable=E0303
return len(self.offsets)-1
def __getitem__(self, i):
self.fp.seek(self.base+self.offsets[i])
return self.fp.read(self.offsets[i+1]-self.offsets[i])
def __iter__(self):
return iter(self[i] for i in range(len(self)))
def __init__(self, name, fp):
self.name = name
self.fp = fp
# Header
(_major, _minor, hdrsize, _) = struct.unpack('BBBB', self.fp.read(4))
self.fp.read(hdrsize-4)
# Name INDEX
self.name_index = self.INDEX(self.fp)
# Top DICT INDEX
self.dict_index = self.INDEX(self.fp)
# String INDEX
self.string_index = self.INDEX(self.fp)
# Global Subr INDEX
self.subr_index = self.INDEX(self.fp)
# Top DICT DATA
self.top_dict = getdict(self.dict_index[0])
(charset_pos,) = self.top_dict.get(15, [0])
(encoding_pos,) = self.top_dict.get(16, [0])
(charstring_pos,) = self.top_dict.get(17, [0])
# CharStrings
self.fp.seek(charstring_pos)
self.charstring = self.INDEX(self.fp)
self.nglyphs = len(self.charstring)
# Encodings
self.code2gid = {}
self.gid2code = {}
self.fp.seek(encoding_pos)
format = self.fp.read(1)
if format == b'\x00':
# Format 0
(n,) = struct.unpack('B', self.fp.read(1))
for (code, gid) in enumerate(struct.unpack('B'*n, self.fp.read(n))):
self.code2gid[code] = gid
self.gid2code[gid] = code
elif format == b'\x01':
# Format 1
(n,) = struct.unpack('B', self.fp.read(1))
code = 0
for _ in range(n):
(first, nleft) = struct.unpack('BB', self.fp.read(2))
for gid in range(first, first+nleft+1):
self.code2gid[code] = gid
self.gid2code[gid] = code
code += 1
else:
raise ValueError('unsupported encoding format: %r' % format)
# Charsets
self.name2gid = {}
self.gid2name = {}
self.fp.seek(charset_pos)
format = self.fp.read(1)
if format == b'\x00':
# Format 0
n = self.nglyphs-1
for (gid, sid) in enumerate(struct.unpack('>'+'H'*n, self.fp.read(2*n))):
gid += 1
name = self.getstr(sid)
self.name2gid[name] = gid
self.gid2name[gid] = name
elif format == b'\x01':
# Format 1
(n,) = struct.unpack('B', self.fp.read(1))
sid = 0
for _ in range(n):
(first, nleft) = struct.unpack('BB', self.fp.read(2))
for gid in range(first, first+nleft+1):
name = self.getstr(sid)
self.name2gid[name] = gid
self.gid2name[gid] = name
sid += 1
elif format == b'\x02':
# Format 2
assert False, str(('Unhandled', format))
else:
raise ValueError('unsupported charset format: %r' % format)
#print self.code2gid
#print self.name2gid
#assert 0
return
def getstr(self, sid):
if sid < len(self.STANDARD_STRINGS):
return self.STANDARD_STRINGS[sid]
return self.string_index[sid-len(self.STANDARD_STRINGS)]
@@ -0,0 +1,140 @@
# PDFFont
from abc import abstractmethod
from io import BytesIO
from pdfmajor.parser.EncodingDB import EncodingDB
from pdfmajor.parser.PSStackParser import LIT
from pdfmajor.parser.PSStackParser import PSLiteral
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFStream import resolve1
from pdfmajor.parser.PDFStream import int_value
from pdfmajor.parser.PDFStream import num_value
from pdfmajor.parser.PDFStream import list_value
from pdfmajor.parser.PDFStream import PDFException
from pdfmajor.parser.PDFStream import PDFStream
from pdfmajor.parser.cmapdb import FileUnicodeMap
from pdfmajor.parser.cmapdb import CMapParser
## Literals
##
LITERAL_STANDARD_ENCODING = LIT('StandardEncoding')
LITERAL_TYPE1C = LIT('Type1C')
class PDFFont(object):
class PDFFontError(PDFException): pass
class PDFUnicodeNotDefined(PDFFontError): pass
def __init__(self, descriptor, widths, default_width=None):
self.descriptor = descriptor
self.widths = widths
self.fontname = resolve1(descriptor.get('FontName', 'unknown'))
if isinstance(self.fontname, PSLiteral):
self.fontname = literal_name(self.fontname)
self.flags = int_value(descriptor.get('Flags', 0))
self.ascent = num_value(descriptor.get('Ascent', 0))
self.descent = num_value(descriptor.get('Descent', 0))
self.italic_angle = num_value(descriptor.get('ItalicAngle', 0))
self.default_width = default_width or num_value(descriptor.get('MissingWidth', 0))
self.leading = num_value(descriptor.get('Leading', 0))
self.bbox = list_value(descriptor.get('FontBBox', (0, 0, 0, 0)))
self.hscale = self.vscale = .001
return
def __repr__(self):
return '<PDFFont>'
def is_vertical(self):
return False
def is_multibyte(self):
return False
def decode(self, bytes):
return bytearray(bytes) # map(ord, bytes)
def get_ascent(self):
return self.ascent * self.vscale
def get_descent(self):
return self.descent * self.vscale
def get_width(self):
w = self.bbox[2]-self.bbox[0]
if w == 0:
w = -self.default_width
return w * self.hscale
def get_height(self):
h = self.bbox[3]-self.bbox[1]
if h == 0:
h = self.ascent - self.descent
return h * self.vscale
def char_width(self, cid):
try:
return self.widths[cid] * self.hscale
except KeyError:
try:
return self.widths[self.to_unichr(cid)] * self.hscale
except (KeyError, self.PDFUnicodeNotDefined):
return self.default_width * self.hscale
def char_disp(self, cid):
return 0
def string_width(self, s):
return sum(self.char_width(cid) for cid in self.decode(s))
@abstractmethod
def to_unichr(self, cid):
raise NotImplementedError
@property
def font_weight(self):
if 'FontWeight' in self.descriptor.keys():
return self.descriptor['FontWeight']
elif 'bold' in self.fontname.lower():
return 'bold'
else:
return None
@property
def is_bold(self):
return self.font_weight == 'bold' or self.font_weight > 549
# PDFSimpleFont
class PDFSimpleFont(PDFFont):
def __init__(self, descriptor, widths, spec):
# Font encoding is specified either by a name of
# built-in encoding or a dictionary that describes
# the differences.
if 'Encoding' in spec:
encoding = resolve1(spec['Encoding'])
else:
encoding = LITERAL_STANDARD_ENCODING
if isinstance(encoding, dict):
name = literal_name(encoding.get('BaseEncoding', LITERAL_STANDARD_ENCODING))
diff = list_value(encoding.get('Differences', []))
self.cid2unicode = EncodingDB.get_encoding(name, diff)
else:
self.cid2unicode = EncodingDB.get_encoding(literal_name(encoding))
self.unicode_map = None
if 'ToUnicode' in spec:
strm = PDFStream.validated_stream(spec['ToUnicode'])
self.unicode_map = FileUnicodeMap()
CMapParser(self.unicode_map, BytesIO(strm.get_data())).run()
PDFFont.__init__(self, descriptor, widths)
return
def to_unichr(self, cid):
if self.unicode_map:
try:
return self.unicode_map.get_unichr(cid)
except KeyError:
pass
try:
return self.cid2unicode[cid]
except KeyError:
raise self.PDFUnicodeNotDefined(None, cid)
@@ -0,0 +1,89 @@
## TrueTypeFont
##
import struct
from pdfmajor.parser.cmapdb import FileUnicodeMap
class TrueTypeFont(object):
class CMapNotFound(Exception):
pass
def __init__(self, name, fp):
self.name = name
self.fp = fp
self.tables = {}
self.fonttype = fp.read(4)
try:
(ntables, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8))
for _ in range(ntables):
(name, _, offset, length) = struct.unpack('>4sLLL', fp.read(16))
self.tables[name] = (offset, length)
except struct.error:
# Do not fail if there are not enough bytes to read. Even for
# corrupted PDFs we would like to get as much information as
# possible, so continue.
pass
return
def create_unicode_map(self):
if 'cmap' not in self.tables:
raise TrueTypeFont.CMapNotFound
(base_offset, _) = self.tables['cmap']
fp = self.fp
fp.seek(base_offset)
(_, nsubtables) = struct.unpack('>HH', fp.read(4))
subtables = []
for i in range(nsubtables):
subtables.append(struct.unpack('>HHL', fp.read(8)))
char2gid = {}
# Only supports subtable type 0, 2 and 4.
for (_1, _2, st_offset) in subtables:
fp.seek(base_offset+st_offset)
(fmttype, _, _) = struct.unpack('>HHH', fp.read(6))
if fmttype == 0:
char2gid.update(enumerate(struct.unpack('>256B', fp.read(256))))
elif fmttype == 2:
subheaderkeys = struct.unpack('>256H', fp.read(512))
firstbytes = [0]*8192
for (i, k) in enumerate(subheaderkeys):
firstbytes[k//8] = i
nhdrs = max(subheaderkeys)//8 + 1
hdrs = []
for i in range(nhdrs):
(firstcode, entcount, delta, offset) = struct.unpack('>HHhH', fp.read(8))
hdrs.append((i, firstcode, entcount, delta, fp.tell()-2+offset))
for (i, firstcode, entcount, delta, pos) in hdrs:
if not entcount:
continue
first = firstcode + (firstbytes[i] << 8)
fp.seek(pos)
for c in range(entcount):
gid = struct.unpack('>H', fp.read(2))
if gid:
gid += delta
char2gid[first+c] = gid
elif fmttype == 4:
(segcount, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8))
segcount //= 2
ecs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
fp.read(2)
scs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
idds = struct.unpack('>%dh' % segcount, fp.read(2*segcount))
pos = fp.tell()
idrs = struct.unpack('>%dH' % segcount, fp.read(2*segcount))
for (ec, sc, idd, idr) in zip(ecs, scs, idds, idrs):
if idr:
fp.seek(pos+idr)
for c in range(sc, ec+1):
char2gid[c] = (struct.unpack('>H', fp.read(2))[0] + idd) & 0xffff
else:
for c in range(sc, ec+1):
char2gid[c] = (c + idd) & 0xffff
else:
assert False, str(('Unhandled', fmttype))
# create unicode map
unicode_map = FileUnicodeMap()
for (char, gid) in char2gid.items():
unicode_map.add_cid2unichr(gid, char)
return unicode_map
@@ -0,0 +1,45 @@
from pdfmajor.parser.PSStackParser import PSStackParser
from pdfmajor.parser.PSStackParser import PSEOF, KWD
from pdfmajor.parser.PSStackParser import PSLiteral
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.EncodingDB import name2unicode
## Type1FontHeaderParser
##
class Type1FontHeaderParser(PSStackParser):
KEYWORD_BEGIN = KWD(b'begin')
KEYWORD_END = KWD(b'end')
KEYWORD_DEF = KWD(b'def')
KEYWORD_PUT = KWD(b'put')
KEYWORD_DICT = KWD(b'dict')
KEYWORD_ARRAY = KWD(b'array')
KEYWORD_READONLY = KWD(b'readonly')
KEYWORD_FOR = KWD(b'for')
KEYWORD_FOR = KWD(b'for')
def __init__(self, data):
PSStackParser.__init__(self, data)
self._cid2unicode = {}
return
def get_encoding(self):
while 1:
try:
(cid, name) = self.nextobject()
except PSEOF:
break
try:
self._cid2unicode[cid] = name2unicode(name)
except KeyError:
pass
return self._cid2unicode
def do_keyword(self, pos, token):
if token is self.KEYWORD_PUT:
((_, key), (_, value)) = self.pop(2) # pylint: disable=E0632
if (isinstance(key, int) and
isinstance(value, PSLiteral)):
self.add_results((key, literal_name(value)))
return
@@ -0,0 +1,57 @@
from pdfmajor.utils import get_logger, settings
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFStream import list_value
from pdfmajor.parser.PDFStream import dict_value
from pdfmajor.parser.PDFStream import resolve1
from pdfmajor.parser.constants import LITERAL_FONT
from .fonts import PDFType1Font, PDFTrueTypeFont, PDFType3Font, PDFCIDFont, PDFFont
log = get_logger(__file__)
def get_font(objid: int, spec: dict, cached_fonts: dict = {}):
if objid and objid in cached_fonts:
font = cached_fonts[objid]
else:
log.debug('get_font: create: objid=%r, spec=%r', objid, spec)
if settings.STRICT:
if spec['Type'] is not LITERAL_FONT:
raise PDFFont.PDFFontError('Type is not /Font')
# Create a Font object.
if 'Subtype' in spec:
subtype = literal_name(spec['Subtype'])
else:
if settings.STRICT:
raise PDFFont.PDFFontError('Font Subtype is not specified.')
subtype = 'Type1'
if subtype in ('Type1', 'MMType1'):
# Type1 Font
font = PDFType1Font(spec)
elif subtype == 'TrueType':
# TrueType Font
font = PDFTrueTypeFont(spec)
elif subtype == 'Type3':
# Type3 Font
font = PDFType3Font(spec)
elif subtype in ('CIDFontType0', 'CIDFontType2'):
# CID Font
font = PDFCIDFont(spec)
elif subtype == 'Type0':
# Type0 Font
dfonts = list_value(spec['DescendantFonts'])
assert dfonts
subspec = dict_value(dfonts[0]).copy()
for k in ('Encoding', 'ToUnicode'):
if k in spec:
subspec[k] = resolve1(spec[k])
font = get_font(None, subspec, cached_fonts)
else:
if settings.STRICT:
raise PDFFont.PDFFontError('Invalid Font spec: %r' % spec)
font = PDFType1Font(spec) # this is so wrong!
if objid:
cached_fonts[objid] = font
return font
@@ -0,0 +1,174 @@
from io import BytesIO
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.PDFStream import int_value
from pdfmajor.parser.PDFStream import num_value
from pdfmajor.parser.PDFStream import list_value
from pdfmajor.parser.PDFStream import dict_value
from pdfmajor.parser.PDFStream import PDFStream
from pdfmajor.parser.PDFStream import resolve1
from pdfmajor.parser.cmapdb import CMap, CMapDB, CMapParser
from pdfmajor.parser.cmapdb import FileUnicodeMap
from pdfmajor.utils import settings, apply_matrix_norm
from .PDFFont import PDFFont, PDFSimpleFont
from .util import FontMetricsDB, get_widths, get_widths2
from .Type1FontHeaderParser import Type1FontHeaderParser
from .TrueTypeFont import TrueTypeFont
# PDFType1Font
class PDFType1Font(PDFSimpleFont):
def __init__(self, spec):
try:
self.basefont = literal_name(spec['BaseFont'])
except KeyError:
if settings.STRICT:
raise self.PDFFontError('BaseFont is missing')
self.basefont = 'unknown'
try:
(descriptor, widths) = FontMetricsDB.get_metrics(self.basefont)
except KeyError:
descriptor = dict_value(spec.get('FontDescriptor', {}))
firstchar = int_value(spec.get('FirstChar', 0))
#lastchar = int_value(spec.get('LastChar', 255))
widths = list_value(spec.get('Widths', [0]*256))
widths = dict((i+firstchar, w) for (i, w) in enumerate(widths))
PDFSimpleFont.__init__(self, descriptor, widths, spec)
if 'Encoding' not in spec and 'FontFile' in descriptor:
# try to recover the missing encoding info from the font file.
self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile'))
length1 = int_value(self.fontfile['Length1'])
data = self.fontfile.get_data()[:length1]
parser = Type1FontHeaderParser(BytesIO(data))
self.cid2unicode = parser.get_encoding()
return
def __repr__(self):
return '<PDFType1Font: basefont=%r>' % self.basefont
# PDFTrueTypeFont
class PDFTrueTypeFont(PDFType1Font):
def __repr__(self):
return '<PDFTrueTypeFont: basefont=%r>' % self.basefont
# PDFType3Font
class PDFType3Font(PDFSimpleFont):
def __init__(self, spec):
firstchar = int_value(spec.get('FirstChar', 0))
#lastchar = int_value(spec.get('LastChar', 0))
widths = list_value(spec.get('Widths', [0]*256))
widths = dict((i+firstchar, w) for (i, w) in enumerate(widths))
if 'FontDescriptor' in spec:
descriptor = dict_value(spec['FontDescriptor'])
else:
descriptor = {'Ascent': 0, 'Descent': 0,
'FontBBox': spec['FontBBox']}
PDFSimpleFont.__init__(self, descriptor, widths, spec)
self.matrix = tuple(list_value(spec.get('FontMatrix')))
(_, self.descent, _, self.ascent) = self.bbox
(self.hscale, self.vscale) = apply_matrix_norm(self.matrix, (1, 1))
return
def __repr__(self):
return '<PDFType3Font>'
# PDFCIDFont
class PDFCIDFont(PDFFont):
def __init__(self, spec, strict=settings.STRICT):
try:
self.basefont = literal_name(spec['BaseFont'])
except KeyError:
if strict:
raise self.PDFFontError('BaseFont is missing')
self.basefont = 'unknown'
self.cidsysteminfo = dict_value(spec.get('CIDSystemInfo', {}))
self.cidcoding = '%s-%s' % (resolve1(self.cidsysteminfo.get('Registry', b'unknown')).decode("latin1"),
resolve1(self.cidsysteminfo.get('Ordering', b'unknown')).decode("latin1"))
try:
name = literal_name(spec['Encoding'])
except KeyError:
if strict:
raise self.PDFFontError('Encoding is unspecified')
name = 'unknown'
try:
self.cmap = CMapDB.get_cmap(name)
except CMapDB.CMapNotFound as e:
if strict:
raise self.PDFFontError(e)
self.cmap = CMap()
try:
descriptor = dict_value(spec['FontDescriptor'])
except KeyError:
if strict:
raise self.PDFFontError('FontDescriptor is missing')
descriptor = {}
ttf = None
if 'FontFile2' in descriptor:
self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile2'))
ttf = TrueTypeFont(self.basefont,
BytesIO(self.fontfile.get_data()))
self.unicode_map = None
if 'ToUnicode' in spec:
strm = PDFStream.validated_stream(spec['ToUnicode'])
self.unicode_map = FileUnicodeMap()
CMapParser(self.unicode_map, BytesIO(strm.get_data())).run()
elif self.cidcoding in ('Adobe-Identity', 'Adobe-UCS'):
if ttf:
try:
self.unicode_map = ttf.create_unicode_map()
except TrueTypeFont.CMapNotFound:
pass
else:
try:
self.unicode_map = CMapDB.get_unicode_map(self.cidcoding, self.cmap.is_vertical())
except CMapDB.CMapNotFound as e:
pass
self.vertical = self.cmap.is_vertical()
if self.vertical:
# writing mode: vertical
widths = get_widths2(list_value(spec.get('W2', [])))
self.disps = dict((cid, (vx, vy)) for (cid, (_, (vx, vy))) in iter(widths.items()))
(vy, w) = spec.get('DW2', [880, -1000])
self.default_disp = (None, vy)
widths = dict((cid, w) for (cid, (w, _)) in iter(widths.items()))
default_width = w
else:
# writing mode: horizontal
self.disps = {}
self.default_disp = 0
widths = get_widths(list_value(spec.get('W', [])))
default_width = spec.get('DW', 1000)
PDFFont.__init__(self, descriptor, widths, default_width=default_width)
return
def __repr__(self):
return '<PDFCIDFont: basefont=%r, cidcoding=%r>' % (self.basefont, self.cidcoding)
def is_vertical(self):
return self.vertical
def is_multibyte(self):
return True
def decode(self, bytes):
return self.cmap.decode(bytes)
def char_disp(self, cid):
"Returns an integer for horizontal fonts, a tuple for vertical fonts."
return self.disps.get(cid, self.default_disp)
def to_unichr(self, cid):
try:
if not self.unicode_map:
raise KeyError(cid)
return self.unicode_map.get_unichr(cid)
except KeyError:
raise self.PDFUnicodeNotDefined(self.cidcoding, cid)
@@ -0,0 +1,57 @@
from pdfmajor.utils import FONT_METRICS, choplist, isnumber
def get_widths(seq):
widths = {}
r = []
for v in seq:
if isinstance(v, list):
if r:
char1 = r[-1]
for (i, w) in enumerate(v):
widths[char1+i] = w
r = []
elif isnumber(v):
r.append(v)
if len(r) == 3:
(char1, char2, w) = r
for i in range(char1, char2+1):
widths[i] = w
r = []
return widths
assert get_widths([1]) == {}
assert get_widths([1,2,3]) == {1:3, 2:3}
assert get_widths([1,[2,3],6,[7,8]]) == {1:2,2:3, 6:7,7:8}
def get_widths2(seq):
widths = {}
r = []
for v in seq:
if isinstance(v, list):
if r:
char1 = r[-1]
for (i, (w, vx, vy)) in enumerate(choplist(3, v)):
widths[char1+i] = (w, (vx, vy))
r = []
elif isnumber(v):
r.append(v)
if len(r) == 5:
(char1, char2, w, vx, vy) = r # pylint: disable=E0632
for i in range(char1, char2+1):
widths[i] = (w, (vx, vy))
r = []
return widths
assert get_widths2([1]) == {}
assert get_widths2([1,2,3,4,5]) == {1:(3, (4,5)), 2:(3, (4,5))}
assert get_widths2([1,[2,3,4,5],6,[7,8,9]]) == {1:(2, (3,4)), 6:(7, (8,9))}
## FontMetricsDB
##
class FontMetricsDB(object):
@classmethod
def get_metrics(klass, fontname):
return FONT_METRICS[fontname]
@@ -1,11 +1,10 @@
from .PDFFont import PDFFont
from ..utils import MATRIX_IDENTITY
from .PDFFont import PDFFont, get_font
from pdfmajor.utils import MATRIX_IDENTITY
## PDFTextState
##
class PDFTextState(object):
def __init__(self):
self.font: PDFFont = None
self.fontsize: int = 0
@@ -15,10 +14,8 @@ class PDFTextState(object):
self.leading: int = 0
self.render: int = 0
self.rise: int = 0
self.reset()
# self.matrix is set
# self.linematrix is set
return
self.matrix = MATRIX_IDENTITY
self.linematrix = [0, 0]
def __repr__(self):
return ('<PDFTextState: font=%r, fontsize=%r, charspace=%r, wordspace=%r, '
@@ -29,7 +26,7 @@ class PDFTextState(object):
self.matrix, self.linematrix))
def copy(self):
obj = PDFTextState()
obj = self.__class__()
obj.font = self.font
obj.fontsize = self.fontsize
obj.charspace = self.charspace
@@ -42,12 +39,10 @@ class PDFTextState(object):
obj.linematrix = self.linematrix
return obj
def reset(self):
def reset_matrices(self):
self.matrix = MATRIX_IDENTITY
self.linematrix = [0, 0]
return
@property
def is_upright(self):
(a, b, c, d, _, _) = self.matrix
return (0 < a*d*self.scaling and b*c <= 0)
return (0 < self.matrix[:,0].prod()*self.scaling and self.matrix[0,1:].prod() <= 0)
@@ -0,0 +1,39 @@
from typing import List
from pdfmajor.utils import MATRIX_IDENTITY
from .PDFGraphicState import PDFGraphicState
from .PDFGraphicState.PDFColor import PDFColor
from .PDFGraphicState.PDFColorSpace import PREDEFINED_COLORSPACE, OrderedDict, PDFColorSpace
from .PDFTextState import PDFTextState, PDFFont, get_font
from .Curves import CurvePath
from .layout import LTItem, LTComponent, LTContainer
from .layout import LTCharBlock, LTChar
from .layout import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
from .layout import LTXObject
from .layout import LTImage
from .layout import make_char_block, make_curve, make_image, make_xobject
class PDFStateStack:
def __init__(self):
self.t_matrix = MATRIX_IDENTITY
self.text = PDFTextState()
self.graphics = PDFGraphicState()
self.graphicstack : List[PDFGraphicState] = []
self.curvestacks: List[CurvePath] = []
self.argstack: List[bytes] = []
self.complete_layout_items : List[LTItem] = []
self.colorspace_map: OrderedDict = PREDEFINED_COLORSPACE.copy()
self.fontmap = {}
self.xobjmap = {}
self.resources = {}
def pop(self, n: int) -> bytearray:
if n == 0:
return []
x = self.argstack[-n:]
self.argstack = self.argstack[:-n]
return x
@@ -0,0 +1,46 @@
from ._base import LTComponent
from pdfmajor.utils import apply_matrix_norm, apply_matrix_pt, matrix2str, Bbox
from ..PDFGraphicState import PDFColor
from ..PDFTextState import PDFTextState, PDFFont
## LTChar
##
class LTChar(LTComponent):
def __init__(self,
bbox: Bbox,
char: str,
textstate: PDFTextState,
color: PDFColor
):
self._text = char
self.textstate = textstate
self.color = color
LTComponent.__init__(self, bbox)
if self.textstate.font.is_vertical():
self.size = self.width
else:
self.size = self.height
return
def __repr__(self):
return (
'<%s %s font=%r text=%r>' %
(
self.__class__.__name__,
str(self.bbox),
self.fontname,
self.get_text()
)
)
def get_text(self):
return self._text
@property
def font(self):
return self.textstate.font
@property
def fontname(self) -> str:
return self.font.fontname
@@ -0,0 +1,66 @@
from typing import List
from ._base import LTContainer
from .LTChar import LTChar
from pdfmajor.utils import apply_matrix_norm, apply_matrix_pt, matrix2str, Bbox, Point
from ..PDFGraphicState import PDFColor
from ..PDFTextState import PDFFont
from ..PDFTextState import PDFTextState
## LTCharBlock
##
class LTCharBlock(LTContainer):
def __init__(self,
chars: list, # [str, List[Point]]
textstate: PDFTextState,
color: PDFColor
):
c, ((x0, y0), (x1, y1)) = chars[0]
ltchars = [
LTChar(
bbox=Bbox(x0, y0, x1, y1),
char= c,
textstate=textstate,
color=color
)
]
for c, pts in chars[1:]:
((nx0, ny0), (nx1, ny1)) = pts
x0 = min([ x0, x1, nx0, nx1 ])
x1 = max([ x0, x1, nx0, nx1 ])
y1 = max([ y0, y1, ny0, ny1 ])
y0 = min([ y0, y1, ny0, ny1 ])
ltchars.append(LTChar(
bbox=Bbox(nx0, ny0, nx1, ny1),
char= c,
textstate=textstate,
color=color
))
LTContainer.__init__(self, Bbox(x0, y0, x1, y1), ltchars)
self.textstate = textstate
self.color = color
if self.textstate.font.is_vertical():
self.size = self.width
else:
self.size = self.height
return
def __repr__(self):
return f"""<{self.__class__.__name__} font="{self.fontname}" text="{self.get_text()}"/>"""
def get_text(self):
return "".join([ c.get_text() for c in self ])
@property
def font(self):
return self.textstate.font
@property
def fontname(self) -> str:
return self.textstate.font.fontname
@@ -0,0 +1,45 @@
from typing import List, Tuple
from pdfmajor.utils import Bbox
from ..PDFGraphicState import PDFColor
from ..Curves import CurvePath
from ._base import LTComponent, Bbox
class LTCurve(LTComponent):
def __init__(self,
linewidth: float,
paths: List[CurvePath],
bbox: Bbox,
stroke: PDFColor,
fill: PDFColor,
evenodd: bool,
):
LTComponent.__init__(self, bbox)
self.paths = paths
self.linewidth = linewidth
self.evenodd = evenodd
self.stroke = stroke
self.fill = fill
## LTLine
##
class LTLine(LTCurve):
pass
## LTHorizontalLine
##
class LTHorizontalLine(LTLine):
pass
## LTHorizontalLine
##
class LTVerticalLine(LTLine):
pass
## LTRect
##
class LTRect(LTCurve):
pass
@@ -0,0 +1,25 @@
from pdfmajor.utils import Bbox
from pdfmajor.parser.PDFStream import PDFStream
from ._base import LTComponent
## LTImage
##
class LTImage(LTComponent):
def __init__(self, name: str, stream: PDFStream, bbox: Bbox):
LTComponent.__init__(self, bbox)
self.name = name
self.stream = stream
self.srcsize = (stream.get_any(('W', 'Width')),
stream.get_any(('H', 'Height')))
self.imagemask = stream.get_any(('IM', 'ImageMask'))
self.bits = stream.get_any(('BPC', 'BitsPerComponent'), 1)
self.colorspace = stream.get_any(('CS', 'ColorSpace'))
if not isinstance(self.colorspace, list):
self.colorspace = [self.colorspace]
return
def __repr__(self):
return ('<%s(%s) %s %r>' %
(self.__class__.__name__, self.name,
str(self.bbox), self.srcsize))
@@ -0,0 +1,22 @@
from typing import List
from pdfmajor.utils import Bbox, Point, matrix2str
from pdfmajor.parser.PDFStream import PDFStream
from ._base import LTContainer
## LTXObject
##
class LTXObject(LTContainer):
def __init__(self, name: str, bbox: Bbox, xobj_stream: PDFStream, resources: dict, t_matrix: List[Point] ):
LTContainer.__init__(self, bbox)
self.name = name
self.stream = xobj_stream
self.t_matrix = t_matrix
self.resources = resources
def __repr__(self):
return ('<%s(%s) %s matrix=%s>' %
(self.__class__.__name__, self.name,
str(self.bbox), matrix2str(self.t_matrix)))
@@ -0,0 +1,133 @@
from typing import List
from pdfmajor.utils import apply_matrix_pt, INF, Bbox, Point, MATRIX_IDENTITY, mult_matrix
from ..Curves import CurvePath
from ..PDFGraphicState import PDFGraphicState, PDFColor
from ..PDFTextState import PDFTextState
from ._base import LTItem, LTComponent, LTContainer
from .LTCharBlock import LTCharBlock
from .LTCurves import LTCurve, LTLine, LTHorizontalLine, LTRect, LTVerticalLine
from .LTImage import LTImage
from .LTXObject import LTXObject
from .utils.textdecoder import decode_text_seq
def make_char_block(seq: bytearray, ctm: tuple, textstate: PDFTextState, color: PDFColor):
(char_meta_datas, textstate, color) = decode_text_seq(seq, ctm, textstate, color)
return LTCharBlock(
chars=char_meta_datas,
color=color,
textstate=textstate
)
def make_curve(ctm: tuple, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]):
shape = ''.join(x.method.value for x in paths)
if shape == 'ml':
# horizontal/vertical line
pt0 = paths[0].points[0]
pt1 = paths[1].points[0]
(x0, y0) = apply_matrix_pt(ctm, (pt0.x, pt0.y))
(x1, y1) = apply_matrix_pt(ctm, (pt1.x, pt1.y))
if x0 == x1:
return LTVerticalLine(
linewidth=gstate.linewidth,
paths=paths,
bbox=Bbox( x0, y0, x1, y1 ),
stroke=gstate.scolor,
fill=gstate.ncolor,
evenodd=evenodd,
)
elif y0 == y1:
return LTHorizontalLine(
linewidth=gstate.linewidth,
paths=paths,
bbox=Bbox( x0, y0, x1, y1 ),
stroke=gstate.scolor,
fill=gstate.ncolor,
evenodd=evenodd,
)
else:
return LTLine(
linewidth=gstate.linewidth,
paths=paths,
bbox=Bbox( x0, y0, x1, y1 ),
stroke=gstate.scolor,
fill=gstate.ncolor,
evenodd=evenodd,
)
elif shape == 'mlllh':
# rectangle
pt0 = paths[0].points[0]
pt1 = paths[1].points[0]
pt2 = paths[2].points[0]
pt3 = paths[3].points[0]
(x0, y0) = apply_matrix_pt(ctm, (pt0.x, pt0.y))
(x1, y1) = apply_matrix_pt(ctm, (pt1.x, pt1.y))
(x2, y2) = apply_matrix_pt(ctm, (pt2.x, pt2.y))
(x3, y3) = apply_matrix_pt(ctm, (pt3.x, pt3.y))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or
(y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
return LTRect(
linewidth=gstate.linewidth,
paths=paths,
bbox=Bbox(
x0=min(x0,x1,x2,x3),
x1=max(x0,x1,x2,x3),
y0=min(y0,y1,y2,y3),
y1=max(y0,y1,y2,y3),
),
evenodd=evenodd,
stroke=gstate.scolor,
fill=gstate.ncolor,
)
else:
(x0, y0, x1, y1) = (INF, INF, -INF, -INF)
for path in paths:
if path.method in [CurvePath.METHOD.MOVE_TO, CurvePath.METHOD.LINE_TO]:
x0 = min(x0, path.points[0].x)
y0 = min(y0, path.points[0].y)
x1 = max(x1, path.points[0].x)
y1 = max(y1, path.points[0].y)
elif path.method in [CurvePath.METHOD.CURVE_BOTH_TO]:
x0 = min(x0, path.points[2].x)
y0 = min(y0, path.points[2].y)
x1 = max(x1, path.points[2].x)
y1 = max(y1, path.points[2].y)
elif path.method in [CurvePath.METHOD.CURVE_NEXT_TO, CurvePath.METHOD.CURVE_FIRST_TO]:
x0 = min(x0, path.points[1].x)
y0 = min(y0, path.points[1].y)
x1 = max(x1, path.points[1].x)
y1 = max(y1, path.points[1].y)
return LTCurve(
linewidth=gstate.linewidth,
paths=paths,
bbox=Bbox(x0,y0,x1,y1),
evenodd=evenodd,
stroke=gstate.scolor,
fill=gstate.ncolor
)
def make_image(obj, ctm):
matrix = mult_matrix(MATRIX_IDENTITY, ctm)
return LTImage(
name=str(id(obj)),
stream=obj,
bbox=Bbox.from_points([
apply_matrix_pt(matrix, (p, q))
for (p, q) in ((0, 0), (1, 0), (0, 1), (1, 1))
])
)
def make_xobject(obj, bbox, ctm, matrix, resources):
(x, y, w, h) = bbox
return LTXObject(
name=str(id(obj)),
bbox=Bbox.from_points([
apply_matrix_pt(matrix, (p, q))
for (p, q) in ((x, y), (x+w, y), (x, y+h), (x+w, y+h))
]),
xobj_stream=obj,
resources=resources,
t_matrix = ctm
)
@@ -0,0 +1,115 @@
from typing import List
from pdfmajor.utils import Bbox
## LTItem
##
class LTItem(object):
pass
## LTComponent
##
class LTComponent(LTItem):
def __init__(self, bbox: Bbox):
if isinstance(bbox, Bbox):
self.bbox = bbox
else:
raise Exception("Invalid BBox")
def __repr__(self):
return ('<%s %s>' %
(self.__class__.__name__, str(self.bbox)))
# Disable comparison.
def __lt__(self, _):
raise ValueError
def __le__(self, _):
raise ValueError
def __gt__(self, _):
raise ValueError
def __ge__(self, _):
raise ValueError
@property
def width(self): return self.bbox.width
@property
def height(self): return self.bbox.height
@property
def x0(self): return self.bbox.x0
@property
def x1(self): return self.bbox.x1
@property
def y0(self): return self.bbox.y0
@property
def y1(self): return self.bbox.y1
def is_empty(self):
return self.width <= 0 or self.height <= 0
def is_hoverlap(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
return obj.x0 <= self.x1 and self.x0 <= obj.x1
def hdistance(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
if self.is_hoverlap(obj):
return 0
else:
return min(abs(self.x0-obj.x1), abs(self.x1-obj.x0))
def hoverlap(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
if self.is_hoverlap(obj):
return min(abs(self.x0-obj.x1), abs(self.x1-obj.x0))
else:
return 0
def is_voverlap(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
return obj.y0 <= self.y1 and self.y0 <= obj.y1
def vdistance(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
if self.is_voverlap(obj):
return 0
else:
return min(abs(self.y0-obj.y1), abs(self.y1-obj.y0))
def voverlap(self, obj):
assert isinstance(obj, LTComponent), str(type(obj))
if self.is_voverlap(obj):
return min(abs(self.y0-obj.y1), abs(self.y1-obj.y0))
else:
return 0
## LTContainer
##
class LTContainer(LTComponent):
def __init__(self, bbox: Bbox, objs: List[LTItem] = []):
LTComponent.__init__(self, bbox)
self._objs: List[LTItem] = objs
return
def __iter__(self):
return iter(self._objs)
def __len__(self) -> int:
return len(self._objs)
def add(self, obj: LTItem):
self._objs.append(obj)
return
def extend(self, objs: List[LTItem]):
for obj in objs:
self.add(obj)
return
@@ -0,0 +1,90 @@
from pdfmajor.utils import mult_matrix, translate_matrix, Bbox, apply_matrix_pt, isnumber
from ....state import PDFTextState, PDFColor
def decode_text_seq(seq: bytearray, ctm: tuple, textstate: PDFTextState, color: PDFColor):
textstate.matrix = mult_matrix(textstate.matrix, ctm)
textstate.scaling = textstate.scaling * .01
textstate.charspace = textstate.charspace * textstate.scaling
textstate.wordspace = textstate.wordspace * textstate.scaling
if textstate.font.is_multibyte():
textstate.wordspace = 0
dxscale = .001 * textstate.fontsize * textstate.scaling
if textstate.font.is_vertical():
return __render_string_along(1,
seq, textstate, dxscale, color
)
else:
return __render_string_along(0,
seq, textstate, dxscale, color
)
def __render_string_along(idx: int, seq: bytearray,
textstate: PDFTextState, dxscale: float, color: PDFColor):
needcharspace = False
char_meta_datas = []
for obj in seq:
if isnumber(obj):
textstate.linematrix[idx] -= obj*dxscale
needcharspace = True
else:
for cid in textstate.font.decode(obj):
if needcharspace:
textstate.linematrix[idx] += textstate.charspace
text = textstate.font.to_unichr(cid)
assert isinstance(text, str), str(type(text))
matrix = translate_matrix(textstate.matrix, textstate.linematrix)
adv, bbox = __compute_char_bbox(matrix, cid, textstate)
textstate.linematrix[idx] += adv
char_meta_datas.append(
( text, bbox )
)
if cid == 32 and textstate.wordspace:
textstate.linematrix[idx] += textstate.wordspace
needcharspace = True
return (char_meta_datas, textstate, color)
def __compute_char_bbox(matrix, char_id: int, textstate: PDFTextState):
font = textstate.font
fontsize = textstate.fontsize
rise = textstate.rise
scaling = textstate.scaling
textwidth = font.char_width(char_id)
textdisp = font.char_disp(char_id)
adv = textwidth * fontsize * scaling
# compute the boundary rectangle.
if font.is_vertical():
# vertical
width = font.get_width() * fontsize
(vx, vy) = textdisp
if vx is None:
vx = width * 0.5
else:
vx = vx * fontsize * .001
vy = (1000 - vy) * fontsize * .001
tx = -vx
ty = vy + rise
bll = (tx, ty+adv)
bur = (tx+width, ty)
else:
# horizontal
height = font.get_height() * fontsize
descent = font.get_descent() * fontsize
ty = descent + rise
bll = (0, ty)
bur = (adv, ty+height)
(x0, y0) = apply_matrix_pt(matrix, bll)
(x1, y1) = apply_matrix_pt(matrix, bur)
if x1 < x0:
(x0, x1) = (x1, x0)
if y1 < y0:
(y0, y1) = (y1, y0)
return (adv, ((x0, y0), (x1, y1)))
-35
View File
@@ -1,35 +0,0 @@
from .PSStackParser import LIT
## Constants
##
LITERAL_PDF = LIT("PDF")
LITERAL_TEXT = LIT("Text")
LITERAL_FONT = LIT("Font")
LITERAL_FORM = LIT("Form")
LITERAL_IMAGE = LIT("Image")
COMMON_FONT_ATTRIBUTES = [
"Type",
"FontName",
"FontFamily",
"FontStretch",
"FontWeight",
"Flags",
"FontBBox",
"ItalicAngle",
"Ascent",
"Descent",
"Leading",
"CapHeight",
"XHeight",
"StemV",
"StemH",
"AvgWidth",
"MaxWidth",
"MissingWidth",
# "FontFile",
# "FontFile2",
# "FontFile3",
"Charset",
]
+50
View File
@@ -0,0 +1,50 @@
from pdfmajor.parser.PDFParser import PDFStream
from pdfmajor.parser.PDFStream import list_value, dict_value, resolve1
from pdfmajor.parser.PDFStream.PDFObjRef import PDFObjRef
from pdfmajor.parser.PSStackParser import literal_name
from pdfmajor.parser.constants import LITERAL_PDF, LITERAL_TEXT
from .commands import PDFColorSpace, PREDEFINED_COLORSPACE
from .commands import PDFStateStack, get_font
def get_colorspace(spec):
if isinstance(spec, list):
name = literal_name(spec[0])
else:
name = literal_name(spec)
if name == 'ICCBased' and isinstance(spec, list) and 2 <= len(spec):
return PDFColorSpace(name, PDFStream.validated_stream(spec[1])['N'])
elif name == 'DeviceN' and isinstance(spec, list) and 2 <= len(spec):
return PDFColorSpace(name, len(list_value(spec[1])))
else:
return PREDEFINED_COLORSPACE.get(name)
def get_procset(procs):
for proc in procs:
if proc is LITERAL_PDF:
pass
elif proc is LITERAL_TEXT:
pass
else:
#raise PDFResourceError('ProcSet %r is not supported.' % proc)
pass
return
def init_resources(state: PDFStateStack, font_cache: dict = {}):
for (k, v) in iter(dict_value(state.resources).items()):
# log.debug('Resource: %r: %r', k, v)
if k == 'Font':
for (fontid, spec) in iter(dict_value(v).items()):
objid = None
if isinstance(spec, PDFObjRef):
objid = spec.objid
spec = dict_value(spec)
state.fontmap[fontid] = get_font(objid, spec, font_cache)
elif k == 'ColorSpace':
for (csid, spec) in iter(dict_value(v).items()):
state.colorspace_map[csid] = get_colorspace(resolve1(spec))
elif k == 'ProcSet':
get_procset(list_value(v))
elif k == 'XObject':
for (xobjid, xobjstrm) in iter(dict_value(v).items()):
state.xobjmap[xobjid] = xobjstrm