split out xref from parser
This commit is contained in:
1 parent
5af06a3f1e
commit
78175d3a64
14 files changed
+290
-154
No files matched your search
@@ -1,2 +1,6 @@
|
|||||||
class PDFMajorException(Exception):
|
class PDFMajorException(Exception):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class BrokenFile(PDFMajorException):
|
||||||
|
pass
|
||||||
+69
-58
@@ -1,6 +1,4 @@
|
|||||||
from pdfmajor.parser.stream.PDFStream import PDFStream
|
from pdfmajor.parser.stream.PDFStream import PDFStream
|
||||||
from pdfmajor.parser.xref import XRefRow
|
|
||||||
from pdfmajor.parser.xrefdb import XRefDB
|
|
||||||
from typing import Iterator, Optional
|
from typing import Iterator, Optional
|
||||||
from pdfmajor.streambuffer import BufferStream
|
from pdfmajor.streambuffer import BufferStream
|
||||||
from pdfmajor.lexer.token import (
|
from pdfmajor.lexer.token import (
|
||||||
@@ -12,7 +10,7 @@ from pdfmajor.lexer.token import (
|
|||||||
TokenKeyword,
|
TokenKeyword,
|
||||||
)
|
)
|
||||||
from pdfmajor.lexer import iter_tokens
|
from pdfmajor.lexer import iter_tokens
|
||||||
from .objects.indirect import IndirectObject
|
from .objects.indirect import IndirectObject, ObjectRef
|
||||||
from .objects.base import PDFObject
|
from .objects.base import PDFObject
|
||||||
from .objects.collections import PDFArray, PDFDictionary
|
from .objects.collections import PDFArray, PDFDictionary
|
||||||
from .objects.comment import PDFComment
|
from .objects.comment import PDFComment
|
||||||
@@ -22,14 +20,12 @@ from .exceptions import (
|
|||||||
InvalidKeywordPos,
|
InvalidKeywordPos,
|
||||||
BrokenFile,
|
BrokenFile,
|
||||||
InvalidKeywordPos,
|
InvalidKeywordPos,
|
||||||
InvalidXref,
|
|
||||||
ParserError,
|
ParserError,
|
||||||
)
|
)
|
||||||
|
|
||||||
|
|
||||||
def iter_objects(
|
def iter_objects(
|
||||||
buffer: BufferStream,
|
buffer: BufferStream,
|
||||||
xrefdb: XRefDB,
|
|
||||||
state: Optional[ParsingState] = None,
|
state: Optional[ParsingState] = None,
|
||||||
) -> Iterator[PDFObject]:
|
) -> Iterator[PDFObject]:
|
||||||
"""Iterates over the objects in the current byte stream
|
"""Iterates over the objects in the current byte stream
|
||||||
@@ -51,18 +47,16 @@ def iter_objects(
|
|||||||
)
|
)
|
||||||
elif isinstance(token, TokenKeyword):
|
elif isinstance(token, TokenKeyword):
|
||||||
if token.value == b"R":
|
if token.value == b"R":
|
||||||
yield from _on_indirect_ref_close(buffer, state, xrefdb)
|
yield from _on_indirect_ref_close(state)
|
||||||
elif token.value == b"obj":
|
elif token.value == b"obj":
|
||||||
state.initialize_indirect_obj(token.start_loc)
|
state.initialize_indirect_obj(token.start_loc)
|
||||||
elif token.value == b"endobj":
|
elif token.value == b"endobj":
|
||||||
obj = _on_endobj(state, token)
|
obj = _on_endobj(state, token)
|
||||||
xrefdb.save_indobject(obj)
|
|
||||||
yield obj
|
yield obj
|
||||||
elif token.value == b"stream":
|
elif token.value == b"stream":
|
||||||
_on_stream(
|
_on_stream(
|
||||||
buffer,
|
buffer,
|
||||||
state,
|
state,
|
||||||
xrefdb,
|
|
||||||
token,
|
token,
|
||||||
)
|
)
|
||||||
else:
|
else:
|
||||||
@@ -106,9 +100,7 @@ def _on_endobj(state: ParsingState, token: TokenKeyword) -> IndirectObject:
|
|||||||
raise ParserError(f"Recived an endobj while not procesing an indirect object")
|
raise ParserError(f"Recived an endobj while not procesing an indirect object")
|
||||||
|
|
||||||
|
|
||||||
def _on_stream(
|
def _on_stream(buffer: BufferStream, state: ParsingState, token: TokenKeyword):
|
||||||
buffer: BufferStream, state: ParsingState, xrefdb: XRefDB, token: TokenKeyword
|
|
||||||
):
|
|
||||||
"""Initilize a PDFStream
|
"""Initilize a PDFStream
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
@@ -119,10 +111,10 @@ def _on_stream(
|
|||||||
"""
|
"""
|
||||||
last_ctx = state.current_context
|
last_ctx = state.current_context
|
||||||
if last_ctx is not None and isinstance(last_ctx, IndirectObject):
|
if last_ctx is not None and isinstance(last_ctx, IndirectObject):
|
||||||
xrefdb.save_indobject(last_ctx)
|
obj = last_ctx.get_object()
|
||||||
# stream = PDFStream(token.end_loc + 1, 0)
|
if not isinstance(obj, PDFDictionary):
|
||||||
obj = _save_or_get_object(buffer, xrefdb, last_ctx.obj_num, last_ctx.gen_num)
|
raise ParserError(f"Cannot initilize a pdfstream with {type(obj)}")
|
||||||
last_ctx.stream = stream = PDFStream.from_pdfdict(token.end_loc, obj)
|
last_ctx.stream = stream = PDFStream.from_pdfdict(token.end_loc + 1, obj)
|
||||||
buffer.seek(stream.offset + stream.length)
|
buffer.seek(stream.offset + stream.length)
|
||||||
next_token = next(iter_tokens(buffer))
|
next_token = next(iter_tokens(buffer))
|
||||||
if not (
|
if not (
|
||||||
@@ -135,9 +127,9 @@ def _on_stream(
|
|||||||
raise ParserError(f"PDFStream is missing initilization dictionary")
|
raise ParserError(f"PDFStream is missing initilization dictionary")
|
||||||
|
|
||||||
|
|
||||||
def _on_indirect_ref_close(buffer: BufferStream, state: ParsingState, xrefdb: XRefDB):
|
def _on_indirect_ref_close(state: ParsingState):
|
||||||
obj_num, gen_num = state.get_indobj_values()
|
obj_num, gen_num = state.get_indobj_values()
|
||||||
obj = _save_or_get_object(buffer, xrefdb, obj_num, gen_num)
|
obj = ObjectRef(obj_num, gen_num)
|
||||||
cur_ctx = state.current_context
|
cur_ctx = state.current_context
|
||||||
if cur_ctx is not None:
|
if cur_ctx is not None:
|
||||||
cur_ctx.pass_item(obj)
|
cur_ctx.pass_item(obj)
|
||||||
@@ -145,50 +137,69 @@ def _on_indirect_ref_close(buffer: BufferStream, state: ParsingState, xrefdb: XR
|
|||||||
yield obj
|
yield obj
|
||||||
|
|
||||||
|
|
||||||
def _save_or_get_object(
|
# def _save_or_get_object(
|
||||||
buffer: BufferStream, xrefdb: XRefDB, obj_num: int, gen_num: int
|
# buffer: BufferStream, xrefdb: XRefDB, obj_num: int, gen_num: int
|
||||||
) -> PDFObject:
|
# ) -> PDFObject:
|
||||||
"""get the actual value of the reference, if the value does not exists return the Null object as specified in PDF spec 1.7 section 7.3.10
|
# """get the actual value of the reference, if the value does not exists return the Null object as specified in PDF spec 1.7 section 7.3.10
|
||||||
|
|
||||||
Args:
|
# Args:
|
||||||
obj_num (int): object number
|
# obj_num (int): object number
|
||||||
gen_num (int): generation number
|
# gen_num (int): generation number
|
||||||
|
|
||||||
Returns:
|
# Returns:
|
||||||
PDFObject
|
# PDFObject
|
||||||
"""
|
# """
|
||||||
indobj = xrefdb.objs.get((obj_num, gen_num), None)
|
# indobj = xrefdb.objs.get((obj_num, gen_num), None)
|
||||||
if indobj is None:
|
# print("trying to read", obj_num, gen_num)
|
||||||
xref = xrefdb.xrefs.get((obj_num, gen_num), None)
|
# if indobj is None:
|
||||||
if xref is None:
|
# xref = xrefdb.xrefs.get((obj_num, gen_num), None)
|
||||||
raise InvalidXref(
|
# if xref is None:
|
||||||
f"xref not found for object retrieval {obj_num}, {gen_num}"
|
# raise InvalidXref(
|
||||||
)
|
# f"xref not found for object retrieval {obj_num}, {gen_num}"
|
||||||
return _read_object_from_xref(xref, xrefdb, buffer)
|
# )
|
||||||
else:
|
# return _read_object_from_xref(xref, xrefdb, buffer)
|
||||||
return _read_object_from_db_or_buffer(indobj, xrefdb, buffer)
|
# else:
|
||||||
|
# return _read_object_from_db_or_buffer(indobj, xrefdb, buffer)
|
||||||
|
|
||||||
|
|
||||||
def _read_object_from_xref(
|
# def _read_object_from_xref(
|
||||||
xref: XRefRow, xrefdb: XRefDB, buffer: BufferStream
|
# xref: XRefRow, xrefdb: XRefDB, buffer: BufferStream
|
||||||
) -> PDFObject:
|
# ) -> PDFObject:
|
||||||
indobj = IndirectObject(xref.obj_num, xref.gen_num, xref.offset)
|
# with buffer.get_window():
|
||||||
with buffer.get_window():
|
# buffer.seek(xref.offset)
|
||||||
buffer.seek(indobj.offset)
|
# obj = next(iter_objects(buffer, xrefdb))
|
||||||
obj = next(iter_objects(buffer, xrefdb))
|
# if (
|
||||||
indobj.save_object(obj)
|
# not isinstance(obj, IndirectObject)
|
||||||
xrefdb.save_indobject(indobj)
|
# or obj.gen_num != xref.gen_num
|
||||||
return obj
|
# or obj.obj_num != xref.obj_num
|
||||||
|
# ):
|
||||||
|
# raise ParserError(
|
||||||
|
# f"Invalid offset for indirect object, expected {xref} but got {obj}"
|
||||||
|
# )
|
||||||
|
# return obj
|
||||||
|
|
||||||
|
|
||||||
def _read_object_from_db_or_buffer(
|
# def _read_object_from_db_or_buffer(
|
||||||
indobj: IndirectObject, xrefdb: XRefDB, buffer: BufferStream
|
# indobj: IndirectObject, xrefdb: XRefDB, buffer: BufferStream
|
||||||
) -> PDFObject:
|
# ) -> PDFObject:
|
||||||
obj = indobj.get_object()
|
# if indobj.read:
|
||||||
if obj is not None:
|
# return indobj
|
||||||
return obj
|
# with buffer.get_window():
|
||||||
with buffer.get_window():
|
# buffer.seek(indobj.offset)
|
||||||
buffer.seek(indobj.offset)
|
# next_token = next(iter_tokens(buffer))
|
||||||
obj = next(iter_objects(buffer, xrefdb))
|
# if next_token.value != b"obj":
|
||||||
indobj.save_object(obj)
|
# if isinstance(next_token, TokenKeyword):
|
||||||
return obj
|
# raise InvalidKeywordPos(next_token, [b"obj"])
|
||||||
|
# else:
|
||||||
|
# raise ParserError(f"expected a keyword instead got {next_token}")
|
||||||
|
# obj = next(
|
||||||
|
# iter_objects(
|
||||||
|
# buffer,
|
||||||
|
# xrefdb,
|
||||||
|
# ParsingState([indobj], []),
|
||||||
|
# )
|
||||||
|
# )
|
||||||
|
# if obj != indobj:
|
||||||
|
# raise ParserError(f"got a different indobj {obj} != {indobj}")
|
||||||
|
# indobj.read = True
|
||||||
|
# return obj
|
||||||
@@ -1,21 +1,21 @@
|
|||||||
from typing import List
|
from typing import List
|
||||||
from pdfmajor.lexer.token import TokenKeyword
|
from pdfmajor.lexer.token import TokenKeyword
|
||||||
from pdfmajor.exceptions import PDFMajorException
|
from pdfmajor.exceptions import PDFMajorException, BrokenFile
|
||||||
|
|
||||||
|
|
||||||
class ParserError(PDFMajorException):
|
class ParserError(PDFMajorException):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
class BrokenFile(ParserError):
|
class BrokenFileParserError(BrokenFile):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
class UnexpectedEOF(BrokenFile, EOFError):
|
class UnexpectedEOF(BrokenFileParserError, EOFError):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
class InvalidKeywordPos(BrokenFile):
|
class InvalidKeywordPos(BrokenFileParserError):
|
||||||
def __init__(self, token: TokenKeyword, expected_token: List[bytes]) -> None:
|
def __init__(self, token: TokenKeyword, expected_token: List[bytes]) -> None:
|
||||||
self.token = token
|
self.token = token
|
||||||
super().__init__(
|
super().__init__(
|
||||||
@@ -23,21 +23,9 @@ class InvalidKeywordPos(BrokenFile):
|
|||||||
)
|
)
|
||||||
|
|
||||||
|
|
||||||
class InvalidIndirectObjAccess(ParserError):
|
class DecodeFailed(BrokenFileParserError):
|
||||||
pass
|
|
||||||
|
|
||||||
|
|
||||||
class DecodeFailed(BrokenFile):
|
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
class InvalidDecoderOrNotImplemented(DecodeFailed, NotImplementedError):
|
class InvalidDecoderOrNotImplemented(DecodeFailed, NotImplementedError):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
|
||||||
class PDFNoValidXRef(BrokenFile, EOFError):
|
|
||||||
pass
|
|
||||||
|
|
||||||
|
|
||||||
class InvalidXref(ParserError, KeyError):
|
|
||||||
pass
|
|
||||||
@@ -4,9 +4,20 @@ from pdfmajor.parser.stream.PDFStream import PDFStream
|
|||||||
from .base import PDFContextualObject, PDFObject
|
from .base import PDFContextualObject, PDFObject
|
||||||
|
|
||||||
|
|
||||||
class IndirectObject(PDFContextualObject):
|
class ObjectRef(PDFObject):
|
||||||
"""A class representing a reference for other objects"""
|
"""A class representing a reference for other objects"""
|
||||||
|
|
||||||
|
def __init__(self, obj_num: int, gen_num: int) -> None:
|
||||||
|
self.obj_num = obj_num
|
||||||
|
self.gen_num = gen_num
|
||||||
|
|
||||||
|
def to_python(self):
|
||||||
|
return (self.obj_num, self.gen_num)
|
||||||
|
|
||||||
|
|
||||||
|
class IndirectObject(PDFContextualObject):
|
||||||
|
"""A class representing a referencable object"""
|
||||||
|
|
||||||
def __init__(
|
def __init__(
|
||||||
self,
|
self,
|
||||||
obj_num: int,
|
obj_num: int,
|
||||||
@@ -20,6 +31,7 @@ class IndirectObject(PDFContextualObject):
|
|||||||
self.gen_num = gen_num
|
self.gen_num = gen_num
|
||||||
self.__data = data
|
self.__data = data
|
||||||
self.stream = stream
|
self.stream = stream
|
||||||
|
self.read: bool = False
|
||||||
|
|
||||||
def __repr__(self) -> str:
|
def __repr__(self) -> str:
|
||||||
return f"IndirectObject(obj_num={self.obj_num}, gen_num={self.gen_num}, offset={self.offset}, data={self.get_object()})"
|
return f"IndirectObject(obj_num={self.obj_num}, gen_num={self.gen_num}, offset={self.offset}, data={self.get_object()})"
|
||||||
@@ -37,10 +49,10 @@ class IndirectObject(PDFContextualObject):
|
|||||||
self.__data = obj
|
self.__data = obj
|
||||||
|
|
||||||
def to_python(self):
|
def to_python(self):
|
||||||
obj = self.get_object()
|
data = self.get_object()
|
||||||
if obj is None:
|
if data is not None:
|
||||||
return None
|
data = data.to_python()
|
||||||
return obj.to_python()
|
return (self.obj_num, self.gen_num, self.offset, data)
|
||||||
|
|
||||||
def pass_item(self, item: PDFObject):
|
def pass_item(self, item: PDFObject):
|
||||||
"""This method will either propogate the item to it's child (in case the child is a contexual object) or override it's own data with the incoming
|
"""This method will either propogate the item to it's child (in case the child is a contexual object) or override it's own data with the incoming
|
||||||
|
|||||||
@@ -32,10 +32,8 @@ def attempt_parse_prim(token: Token, state: ParsingState) -> ParsingAttempt:
|
|||||||
if len(state.int_collection) <= 1:
|
if len(state.int_collection) <= 1:
|
||||||
state.int_collection.append(prim_obj)
|
state.int_collection.append(prim_obj)
|
||||||
else:
|
else:
|
||||||
for last_num in state.int_collection:
|
for obj in state.flush_int_collections():
|
||||||
out.append(last_num)
|
out.append(obj)
|
||||||
state.int_collection = []
|
|
||||||
out.append(prim_obj)
|
|
||||||
else:
|
else:
|
||||||
for obj in state.flush_int_collections():
|
for obj in state.flush_int_collections():
|
||||||
out.append(obj)
|
out.append(obj)
|
||||||
|
|||||||
@@ -1,4 +1,4 @@
|
|||||||
from typing import Iterator, List, Optional, Tuple
|
from typing import Iterator, List, Optional, Tuple, Union
|
||||||
|
|
||||||
from .exceptions import ParserError
|
from .exceptions import ParserError
|
||||||
from .objects.base import PDFContextualObject, PDFObject
|
from .objects.base import PDFContextualObject, PDFObject
|
||||||
@@ -14,7 +14,7 @@ class ParsingState:
|
|||||||
def __init__(
|
def __init__(
|
||||||
self,
|
self,
|
||||||
context_stack: List[PDFContextualObject],
|
context_stack: List[PDFContextualObject],
|
||||||
int_collection: List[PDFInteger],
|
int_collection: List[Union[PDFInteger, int]],
|
||||||
last_obj: Optional[PDFObject] = None,
|
last_obj: Optional[PDFObject] = None,
|
||||||
) -> None:
|
) -> None:
|
||||||
self.context_stack = context_stack
|
self.context_stack = context_stack
|
||||||
@@ -68,8 +68,12 @@ class ParsingState:
|
|||||||
"""
|
"""
|
||||||
if len(self.int_collection) == 2:
|
if len(self.int_collection) == 2:
|
||||||
gen_num = self.int_collection.pop()
|
gen_num = self.int_collection.pop()
|
||||||
|
if isinstance(gen_num, PDFInteger):
|
||||||
|
gen_num = gen_num.to_python()
|
||||||
obj_num = self.int_collection.pop()
|
obj_num = self.int_collection.pop()
|
||||||
return (obj_num.to_python(), gen_num.to_python())
|
if isinstance(obj_num, PDFInteger):
|
||||||
|
obj_num = obj_num.to_python()
|
||||||
|
return (obj_num, gen_num)
|
||||||
raise ParserError("missing leading tokens for object reference")
|
raise ParserError("missing leading tokens for object reference")
|
||||||
|
|
||||||
def initialize_indirect_obj(self, offset: int):
|
def initialize_indirect_obj(self, offset: int):
|
||||||
@@ -92,9 +96,17 @@ class ParsingState:
|
|||||||
cur_ctx = self.current_context
|
cur_ctx = self.current_context
|
||||||
if cur_ctx is not None:
|
if cur_ctx is not None:
|
||||||
for obj in self.int_collection:
|
for obj in self.int_collection:
|
||||||
|
if not isinstance(obj, PDFInteger):
|
||||||
|
raise ParserError(
|
||||||
|
f"Invalid int-collection contains none PDFInteger types {obj}"
|
||||||
|
)
|
||||||
cur_ctx.pass_item(obj)
|
cur_ctx.pass_item(obj)
|
||||||
else:
|
else:
|
||||||
for obj in self.int_collection:
|
for obj in self.int_collection:
|
||||||
|
if not isinstance(obj, PDFInteger):
|
||||||
|
raise ParserError(
|
||||||
|
f"Invalid int-collection contains none PDFInteger types {obj}"
|
||||||
|
)
|
||||||
yield obj
|
yield obj
|
||||||
self.int_collection = []
|
self.int_collection = []
|
||||||
|
|
||||||
|
|||||||
@@ -40,9 +40,7 @@ class PDFStream:
|
|||||||
self.dl: Optional[int] = dl
|
self.dl: Optional[int] = dl
|
||||||
|
|
||||||
@classmethod
|
@classmethod
|
||||||
def from_pdfdict(cls, offset: int, item: PDFObject) -> "PDFStream":
|
def from_pdfdict(cls, offset: int, item: PDFDictionary) -> "PDFStream":
|
||||||
if not isinstance(item, PDFDictionary):
|
|
||||||
raise ParserError(f"Cannot initilize a pdfstream with {type(item)}")
|
|
||||||
data = item.to_python()
|
data = item.to_python()
|
||||||
stream = cls(offset, data["Length"])
|
stream = cls(offset, data["Length"])
|
||||||
stream.filter = get_single_or_list(data.get("Filter", stream.filter))
|
stream.filter = get_single_or_list(data.get("Filter", stream.filter))
|
||||||
|
|||||||
@@ -1,27 +0,0 @@
|
|||||||
from pdfmajor.parser.xref import XRefRow
|
|
||||||
|
|
||||||
from typing import Dict, Iterator, Tuple
|
|
||||||
|
|
||||||
from pdfmajor.parser.exceptions import InvalidXref
|
|
||||||
from pdfmajor.parser.objects.base import PDFObject
|
|
||||||
from .objects.indirect import IndirectObject
|
|
||||||
from .objects.base import PDFObject
|
|
||||||
|
|
||||||
|
|
||||||
class XRefDB:
|
|
||||||
"""A class for collection references to objects"""
|
|
||||||
|
|
||||||
def __init__(self) -> None:
|
|
||||||
self.objs: Dict[Tuple[int, int], IndirectObject] = {}
|
|
||||||
self.xrefs: Dict[Tuple[int, int], XRefRow] = {}
|
|
||||||
|
|
||||||
def update_from_xrefiter(self, it_xref: Iterator[XRefRow]):
|
|
||||||
for xref in it_xref:
|
|
||||||
if xref.use == b"n":
|
|
||||||
self.xrefs[(xref.obj_num, xref.gen_num)] = xref
|
|
||||||
|
|
||||||
def save_indobject(self, indobj: IndirectObject):
|
|
||||||
if self.objs.get((indobj.obj_num, indobj.gen_num), None) is None:
|
|
||||||
self.objs[(indobj.obj_num, indobj.gen_num)] = indobj
|
|
||||||
else:
|
|
||||||
print("WARN: saving twice", indobj)
|
|
||||||
@@ -0,0 +1,21 @@
|
|||||||
|
from pdfmajor.exceptions import PDFMajorException, BrokenFile
|
||||||
|
|
||||||
|
|
||||||
|
class XRefError(PDFMajorException):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class PDFNoValidXRef(XRefError, BrokenFile, EOFError):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class InvalidXref(XRefError, KeyError):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class InvalidIndirectObjAccess(XRefError):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class UnexpectedEOF(BrokenFile, EOFError):
|
||||||
|
pass
|
||||||
@@ -1,8 +1,8 @@
|
|||||||
# from pdfmajor.parser.objects.indirect import IndirectObjectCollection
|
from typing import Iterator, Literal, NamedTuple, Union
|
||||||
from pdfmajor.lexer.token import TokenInteger, TokenKeyword
|
from pdfmajor.lexer.token import TokenInteger, TokenKeyword
|
||||||
from pdfmajor.lexer import iter_tokens
|
from pdfmajor.lexer import iter_tokens
|
||||||
from typing import Iterator, Literal, NamedTuple, Union
|
|
||||||
from pdfmajor.streambuffer import BufferStream
|
from pdfmajor.streambuffer import BufferStream
|
||||||
|
|
||||||
from .exceptions import PDFNoValidXRef, BrokenFile, UnexpectedEOF
|
from .exceptions import PDFNoValidXRef, BrokenFile, UnexpectedEOF
|
||||||
|
|
||||||
|
|
||||||
@@ -0,0 +1,51 @@
|
|||||||
|
from typing import Dict, Iterator, Tuple
|
||||||
|
|
||||||
|
from pdfmajor.parser.objects.indirect import IndirectObject
|
||||||
|
from .xref import XRefRow
|
||||||
|
from .exceptions import InvalidIndirectObjAccess
|
||||||
|
|
||||||
|
|
||||||
|
class XRefDB:
|
||||||
|
"""A class for collection references to objects"""
|
||||||
|
|
||||||
|
def __init__(self) -> None:
|
||||||
|
self.objs: Dict[Tuple[int, int], IndirectObject] = {}
|
||||||
|
self.xrefs: Dict[Tuple[int, int], XRefRow] = {}
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def from_xrefiter(cls, it_xref: Iterator[XRefRow]) -> "XRefDB":
|
||||||
|
db = cls()
|
||||||
|
db.update_from_xrefiter(it_xref)
|
||||||
|
return db
|
||||||
|
|
||||||
|
def has_xref(self, obj_num: int, gen_num: int) -> bool:
|
||||||
|
return (obj_num, gen_num) in self.xrefs.keys()
|
||||||
|
|
||||||
|
def has_obj(self, obj_num: int, gen_num: int) -> bool:
|
||||||
|
return (obj_num, gen_num) in self.objs.keys()
|
||||||
|
|
||||||
|
def update_from_xrefiter(self, it_xref: Iterator[XRefRow]):
|
||||||
|
for xref in it_xref:
|
||||||
|
if xref.use == b"n":
|
||||||
|
self.xrefs[(xref.obj_num, xref.gen_num)] = xref
|
||||||
|
|
||||||
|
def create_indobject(
|
||||||
|
self, obj_num: int, gen_num: int, offset: int
|
||||||
|
) -> IndirectObject:
|
||||||
|
if self.has_obj(obj_num, gen_num):
|
||||||
|
raise InvalidIndirectObjAccess(
|
||||||
|
f"Attempted to recreate {(obj_num, gen_num)}"
|
||||||
|
)
|
||||||
|
cur_obj = self.objs[(obj_num, gen_num)] = IndirectObject(
|
||||||
|
obj_num, gen_num, offset
|
||||||
|
)
|
||||||
|
return cur_obj
|
||||||
|
|
||||||
|
def get_or_create_indobject(self, obj_num: int, gen_num: int) -> IndirectObject:
|
||||||
|
cur_obj = self.objs.get((obj_num, gen_num), None)
|
||||||
|
if cur_obj is None:
|
||||||
|
print("WARN: got a non-existent reference")
|
||||||
|
cur_obj = self.objs[(obj_num, gen_num)] = IndirectObject(
|
||||||
|
obj_num, gen_num, -1
|
||||||
|
)
|
||||||
|
return cur_obj
|
||||||
@@ -0,0 +1,44 @@
|
|||||||
|
from pdfmajor.parser.xref import iter_over_xref
|
||||||
|
from pdfmajor.streambuffer import BufferStream
|
||||||
|
from pdfmajor.parser.objects.indirect import IndirectObject
|
||||||
|
from pdfmajor.parser.xrefdb import XRefDB
|
||||||
|
from datetime import datetime
|
||||||
|
from typing import List, NamedTuple, Tuple
|
||||||
|
|
||||||
|
|
||||||
|
class PDFDocumentMetadata(NamedTuple):
|
||||||
|
version: str
|
||||||
|
creator: str
|
||||||
|
producer: str
|
||||||
|
creation_date: datetime
|
||||||
|
mediabox: Tuple[int, int, int, int]
|
||||||
|
|
||||||
|
|
||||||
|
class PDFPage(IndirectObject):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class PDFResoruce(IndirectObject):
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
|
class PDFDocument:
|
||||||
|
metadata: PDFDocumentMetadata
|
||||||
|
_xrefdb: XRefDB
|
||||||
|
_resources: List[PDFResoruce]
|
||||||
|
_pages: List[PDFPage]
|
||||||
|
|
||||||
|
def __init__(self, buffer: BufferStream, strict: bool = False) -> None:
|
||||||
|
self._xrefdb = XRefDB()
|
||||||
|
self._xrefdb.update_from_xrefiter(iter_over_xref(buffer, strict))
|
||||||
|
self.metadata = get_metadata_from_buffer(buffer, self._xrefdb)
|
||||||
|
self._pages = get_pages_from_buffer(buffer, self._xrefdb)
|
||||||
|
self._resources = get_resources_from_buffer(buffer, self._xrefdb)
|
||||||
|
|
||||||
|
def read_page(self, page_num: int) -> PDFPage:
|
||||||
|
page_obj = self._pages[page_num]
|
||||||
|
# for
|
||||||
|
|
||||||
|
|
||||||
|
# pdf_doc = PDFDocument.from_path('path/to/file.pdf')
|
||||||
|
# pdf_doc.metadata
|
||||||
+21
-30
@@ -1,15 +1,14 @@
|
|||||||
from pathlib import Path
|
from pathlib import Path
|
||||||
from pdfmajor.parser.xrefdb import XRefDB
|
|
||||||
from pdfmajor.parser import iter_objects
|
from pdfmajor.parser import iter_objects
|
||||||
from pdfmajor.parser.objects.collections import PDFDictionary
|
from pdfmajor.parser.objects.collections import PDFDictionary
|
||||||
from pdfmajor.parser.objects.primitives import PDFHexString, PDFInteger, PDFString
|
from pdfmajor.parser.objects.primitives import PDFHexString, PDFInteger, PDFString
|
||||||
|
|
||||||
from typing import Any, List
|
from typing import Any, Callable, List, Optional
|
||||||
from decimal import Decimal
|
from decimal import Decimal
|
||||||
from unittest import TestCase
|
from unittest import TestCase
|
||||||
|
|
||||||
from pdfmajor.streambuffer import BufferStream
|
from pdfmajor.streambuffer import BufferStream
|
||||||
from pdfmajor.parser.objects.indirect import IndirectObject
|
from pdfmajor.parser.objects.indirect import IndirectObject, ObjectRef
|
||||||
from pdfmajor.parser.objects.base import PDFObject
|
from pdfmajor.parser.objects.base import PDFObject
|
||||||
from pdfmajor.parser.objects.comment import PDFComment
|
from pdfmajor.parser.objects.comment import PDFComment
|
||||||
from pdfmajor.parser.stream.PDFStream import PDFStream
|
from pdfmajor.parser.stream.PDFStream import PDFStream
|
||||||
@@ -32,7 +31,7 @@ class Collections(TestCase):
|
|||||||
4 * size,
|
4 * size,
|
||||||
]:
|
]:
|
||||||
buffer = BufferStream.from_bytes(raw, buffer_size=buf_size)
|
buffer = BufferStream.from_bytes(raw, buffer_size=buf_size)
|
||||||
it = iter_objects(buffer, XRefDB())
|
it = iter_objects(buffer)
|
||||||
obj = next(it)
|
obj = next(it)
|
||||||
self.assertEqual(obj.to_python(), expected)
|
self.assertEqual(obj.to_python(), expected)
|
||||||
with self.assertRaises(StopIteration):
|
with self.assertRaises(StopIteration):
|
||||||
@@ -90,8 +89,11 @@ class Collections(TestCase):
|
|||||||
|
|
||||||
|
|
||||||
class IndirectObjects(TestCase):
|
class IndirectObjects(TestCase):
|
||||||
def run_test(self, raw: bytes, expected: List[PDFObject]) -> XRefDB:
|
def run_test(
|
||||||
db: XRefDB = XRefDB()
|
self,
|
||||||
|
raw: bytes,
|
||||||
|
expected: List[PDFObject],
|
||||||
|
):
|
||||||
size = len(raw)
|
size = len(raw)
|
||||||
for buf_size in [
|
for buf_size in [
|
||||||
2,
|
2,
|
||||||
@@ -104,14 +106,16 @@ class IndirectObjects(TestCase):
|
|||||||
2 * size,
|
2 * size,
|
||||||
4 * size,
|
4 * size,
|
||||||
]:
|
]:
|
||||||
db = XRefDB()
|
|
||||||
buffer = BufferStream.from_bytes(raw, buffer_size=buf_size)
|
buffer = BufferStream.from_bytes(raw, buffer_size=buf_size)
|
||||||
it = iter_objects(buffer, db)
|
it = iter_objects(buffer)
|
||||||
for obj, eobj in zip(it, expected):
|
for obj, eobj in zip(it, expected):
|
||||||
self.assertEqual(obj, eobj)
|
self.assertEqual(
|
||||||
|
obj,
|
||||||
|
eobj,
|
||||||
|
f"objects do not match\n incoming={obj}\n expected={eobj}",
|
||||||
|
)
|
||||||
with self.assertRaises((StopIteration, EOFError)):
|
with self.assertRaises((StopIteration, EOFError)):
|
||||||
next(it)
|
next(it)
|
||||||
return db
|
|
||||||
|
|
||||||
def test_parse_simple_indobj(self):
|
def test_parse_simple_indobj(self):
|
||||||
self.run_test(
|
self.run_test(
|
||||||
@@ -126,11 +130,11 @@ class IndirectObjects(TestCase):
|
|||||||
""",
|
""",
|
||||||
expected=[
|
expected=[
|
||||||
IndirectObject(12, 1, 18, PDFString("Bring", 0, 0)),
|
IndirectObject(12, 1, 18, PDFString("Bring", 0, 0)),
|
||||||
IndirectObject(100, 0, 25, PDFHexString(b"3", 0, 0)),
|
IndirectObject(100, 0, 55, PDFHexString(b"3", 0, 0)),
|
||||||
IndirectObject(
|
IndirectObject(
|
||||||
13,
|
13,
|
||||||
8,
|
8,
|
||||||
144,
|
124,
|
||||||
PDFDictionary.from_dict(
|
PDFDictionary.from_dict(
|
||||||
{"x": PDFInteger(1, 0, 0), "y": PDFInteger(3, 0, 0)}
|
{"x": PDFInteger(1, 0, 0), "y": PDFInteger(3, 0, 0)}
|
||||||
),
|
),
|
||||||
@@ -139,7 +143,7 @@ class IndirectObjects(TestCase):
|
|||||||
)
|
)
|
||||||
|
|
||||||
def test_parse_obj_with_references(self):
|
def test_parse_obj_with_references(self):
|
||||||
db = self.run_test(
|
self.run_test(
|
||||||
raw=br"""
|
raw=br"""
|
||||||
12 1 obj (Bring) endobj
|
12 1 obj (Bring) endobj
|
||||||
10 2 obj << /lastone 12 1 R >> endobj
|
10 2 obj << /lastone 12 1 R >> endobj
|
||||||
@@ -148,18 +152,13 @@ class IndirectObjects(TestCase):
|
|||||||
expected=[
|
expected=[
|
||||||
IndirectObject(12, 1, 18, PDFString("Bring", 0, 0)),
|
IndirectObject(12, 1, 18, PDFString("Bring", 0, 0)),
|
||||||
IndirectObject(
|
IndirectObject(
|
||||||
0,
|
10,
|
||||||
2,
|
2,
|
||||||
54,
|
54,
|
||||||
data=PDFDictionary.from_dict({"lastone": PDFString("Bring", 0, 0)}),
|
data=PDFDictionary.from_dict({"lastone": ObjectRef(12, 1)}),
|
||||||
),
|
),
|
||||||
],
|
],
|
||||||
)
|
)
|
||||||
sec_obj = db.objs[(10, 2)]
|
|
||||||
self.assertEqual(
|
|
||||||
sec_obj.to_python(),
|
|
||||||
{"lastone": "Bring"},
|
|
||||||
)
|
|
||||||
|
|
||||||
def test_parse_comments(self):
|
def test_parse_comments(self):
|
||||||
self.run_test(
|
self.run_test(
|
||||||
@@ -191,14 +190,14 @@ stream
|
|||||||
endobj
|
endobj
|
||||||
"""
|
"""
|
||||||
)
|
)
|
||||||
for obj in iter_objects(buffer, XRefDB()):
|
for obj in iter_objects(buffer):
|
||||||
self.assertIsInstance(obj, IndirectObject)
|
self.assertIsInstance(obj, IndirectObject)
|
||||||
self.assertIsInstance(obj.get_object(), PDFDictionary)
|
self.assertIsInstance(obj.get_object(), PDFDictionary)
|
||||||
self.assertIsInstance(obj.stream, PDFStream) # type: ignore
|
self.assertIsInstance(obj.stream, PDFStream) # type: ignore
|
||||||
self.assertDictEqual(
|
self.assertDictEqual(
|
||||||
obj.stream.to_python(), # type: ignore
|
obj.stream.to_python(), # type: ignore
|
||||||
{
|
{
|
||||||
"offset": 99,
|
"offset": 100,
|
||||||
"length": 123,
|
"length": 123,
|
||||||
"filter": ["/FlateDecode"],
|
"filter": ["/FlateDecode"],
|
||||||
"decode_parms": [],
|
"decode_parms": [],
|
||||||
@@ -217,11 +216,3 @@ endobj
|
|||||||
"otherfeature": [10, 20],
|
"otherfeature": [10, 20],
|
||||||
},
|
},
|
||||||
)
|
)
|
||||||
|
|
||||||
|
|
||||||
# class PDFStreamParsing(TestCase):
|
|
||||||
# def test_simple(self):
|
|
||||||
# with (CURRENT_FOLDER / "samples" / "pdf" / "chars.pdf").open("br") as fp:
|
|
||||||
# parser = PDFParser(fp)
|
|
||||||
# for obj in parser.iter_objects():
|
|
||||||
# print(obj)
|
|
||||||
+38
-5
@@ -1,10 +1,9 @@
|
|||||||
from pathlib import Path
|
from pathlib import Path
|
||||||
from pdfmajor.parser.xrefdb import XRefDB
|
from pdfmajor.xref.xrefdb import XRefDB
|
||||||
from pdfmajor.parser import _save_or_get_object
|
|
||||||
from unittest import TestCase
|
from unittest import TestCase
|
||||||
|
|
||||||
from pdfmajor.streambuffer import BufferStream
|
from pdfmajor.streambuffer import BufferStream
|
||||||
from pdfmajor.parser.xref import iter_over_xref
|
from pdfmajor.xref.xref import XRefRow, iter_over_xref
|
||||||
|
|
||||||
CURRENT_FOLDER = Path(__file__).parent
|
CURRENT_FOLDER = Path(__file__).parent
|
||||||
|
|
||||||
@@ -15,5 +14,39 @@ class Standard(TestCase):
|
|||||||
buffer = BufferStream(fp)
|
buffer = BufferStream(fp)
|
||||||
db = XRefDB()
|
db = XRefDB()
|
||||||
db.update_from_xrefiter(iter_over_xref(buffer))
|
db.update_from_xrefiter(iter_over_xref(buffer))
|
||||||
for obj, gen in db.xrefs.keys():
|
self.assertDictEqual(
|
||||||
print(_save_or_get_object(buffer, db, obj, gen))
|
db.xrefs,
|
||||||
|
{
|
||||||
|
(1, 0): XRefRow(offset=27259, obj_num=1, gen_num=0, use=b"n"),
|
||||||
|
(2, 0): XRefRow(offset=19, obj_num=2, gen_num=0, use=b"n"),
|
||||||
|
(3, 0): XRefRow(offset=344, obj_num=3, gen_num=0, use=b"n"),
|
||||||
|
(4, 0): XRefRow(offset=27402, obj_num=4, gen_num=0, use=b"n"),
|
||||||
|
(5, 0): XRefRow(offset=364, obj_num=5, gen_num=0, use=b"n"),
|
||||||
|
(6, 0): XRefRow(offset=714, obj_num=6, gen_num=0, use=b"n"),
|
||||||
|
(7, 0): XRefRow(offset=27545, obj_num=7, gen_num=0, use=b"n"),
|
||||||
|
(8, 0): XRefRow(offset=734, obj_num=8, gen_num=0, use=b"n"),
|
||||||
|
(9, 0): XRefRow(offset=5949, obj_num=9, gen_num=0, use=b"n"),
|
||||||
|
(10, 0): XRefRow(offset=5970, obj_num=10, gen_num=0, use=b"n"),
|
||||||
|
(11, 0): XRefRow(offset=6176, obj_num=11, gen_num=0, use=b"n"),
|
||||||
|
(12, 0): XRefRow(offset=6503, obj_num=12, gen_num=0, use=b"n"),
|
||||||
|
(13, 0): XRefRow(offset=6700, obj_num=13, gen_num=0, use=b"n"),
|
||||||
|
(14, 0): XRefRow(offset=11679, obj_num=14, gen_num=0, use=b"n"),
|
||||||
|
(15, 0): XRefRow(offset=11701, obj_num=15, gen_num=0, use=b"n"),
|
||||||
|
(16, 0): XRefRow(offset=11905, obj_num=16, gen_num=0, use=b"n"),
|
||||||
|
(17, 0): XRefRow(offset=12232, obj_num=17, gen_num=0, use=b"n"),
|
||||||
|
(18, 0): XRefRow(offset=12427, obj_num=18, gen_num=0, use=b"n"),
|
||||||
|
(19, 0): XRefRow(offset=17804, obj_num=19, gen_num=0, use=b"n"),
|
||||||
|
(20, 0): XRefRow(offset=17826, obj_num=20, gen_num=0, use=b"n"),
|
||||||
|
(21, 0): XRefRow(offset=18037, obj_num=21, gen_num=0, use=b"n"),
|
||||||
|
(22, 0): XRefRow(offset=18364, obj_num=22, gen_num=0, use=b"n"),
|
||||||
|
(23, 0): XRefRow(offset=18565, obj_num=23, gen_num=0, use=b"n"),
|
||||||
|
(24, 0): XRefRow(offset=26258, obj_num=24, gen_num=0, use=b"n"),
|
||||||
|
(25, 0): XRefRow(offset=26280, obj_num=25, gen_num=0, use=b"n"),
|
||||||
|
(26, 0): XRefRow(offset=26477, obj_num=26, gen_num=0, use=b"n"),
|
||||||
|
(27, 0): XRefRow(offset=26882, obj_num=27, gen_num=0, use=b"n"),
|
||||||
|
(28, 0): XRefRow(offset=27141, obj_num=28, gen_num=0, use=b"n"),
|
||||||
|
(29, 0): XRefRow(offset=27204, obj_num=29, gen_num=0, use=b"n"),
|
||||||
|
(30, 0): XRefRow(offset=27650, obj_num=30, gen_num=0, use=b"n"),
|
||||||
|
(31, 0): XRefRow(offset=27747, obj_num=31, gen_num=0, use=b"n"),
|
||||||
|
},
|
||||||
|
)
|
||||||
Reference in new issue
Block a user