This commit is contained in:
ari committed 2020-10-30 19:52:57 -04:00
1 parent 3f5bac98b3
commit e33d13a3e1
5 files changed
+145 -12

No files matched your search

+9 -6
View File
@@ -1,9 +1,8 @@
from pathlib import Path
from pdfmajor.document.PDFParsingContext import PDFParsingContext
from typing import BinaryIO
from pdfmajor.healthlog import PDFHealthReport
from pdfmajor.streambuffer import BufferStream
from pdfmajor.xref.xrefdb import XRefDB
class PDFDocument:
@@ -22,7 +21,11 @@ class PDFDocument:
return cls(BufferStream(path.open("rb"), buffer_size))
def __init__(self, buffer: BufferStream) -> None:
self.buffer = buffer
self.health_report = PDFHealthReport()
with self.buffer.get_window():
self.xref_db = XRefDB(buffer)
self.__parser = PDFParsingContext(buffer)
self.catalog = self.__parser.get_catalog()
self.info = self.__parser.get_info()
@property
def health_report(self):
"""A log of errors that occured in the document during the processing of the file"""
return self.__parser.health_report
+16 -5
View File
@@ -1,4 +1,5 @@
from typing import Set
from decimal import Decimal
from typing import List, Set
from pdfmajor.exceptions import BrokenFile, PDFMajorException
from pdfmajor.parser.objects.indirect import ObjectRef
@@ -8,15 +9,25 @@ class PDFDocumentError(PDFMajorException):
pass
class MissingCatalogObj(PDFDocumentError, BrokenFile):
class BrokenFilePDF(BrokenFile):
pass
class InvalidCatalogObj(PDFDocumentError, BrokenFile):
class MissingCatalogObj(BrokenFilePDF):
pass
class TooManyInfoObj(PDFDocumentError, BrokenFile):
class InvalidCatalogObj(BrokenFilePDF):
pass
class TooManyInfoObj(BrokenFilePDF):
def __init__(self, elms: Set[ObjectRef]) -> None:
super().__init__(f"Too many info elements found | total={elms}")
super().__init__(f"Too many info elements found | elms={elms}")
self.elms = elms
class TooManyRectField(BrokenFilePDF):
def __init__(self, values: List[Decimal]) -> None:
super().__init__(f"Too many elements found for rectangle | values={values}")
self.values = values
+17
View File
@@ -0,0 +1,17 @@
from typing import List, NamedTuple, Optional
from pdfmajor.parser.objects.indirect import ObjectRef
from pdfmajor.parser.objects.collections import PDFDictionary
class PDFPageTreeNode(NamedTuple):
"""A PDF Pages Node representation as it confirms with PDF spec 1.7 section 7.7.3.2"""
kids: List[ObjectRef]
parent: Optional[ObjectRef] = None
class PDFPage(NamedTuple):
"""A PDF Page representation as it confirms with PDF Spec 1.7 section 7.7.3.3"""
parent: ObjectRef
resources: PDFDictionary
+32
View File
@@ -0,0 +1,32 @@
from decimal import Decimal
from pdfmajor.document.exceptions import BrokenFilePDF, TooManyRectField
from pdfmajor.parser.objects.primitives import PDFInteger, PDFReal
from typing import List, NamedTuple
from pdfmajor.parser.objects.collections import PDFArray
class PDFRectangle(NamedTuple):
"""A PDF Rectangle as specified by PDF spec 1.7 section 7.9.5"""
llx: Decimal
lly: Decimal
urx: Decimal
ury: Decimal
@classmethod
def from_pdfarray(cls, arr: PDFArray):
vals: List[Decimal] = []
for obj in arr:
if isinstance(obj, PDFInteger):
vals.append(Decimal(obj.to_python()))
elif isinstance(obj, PDFReal):
vals.append(obj.to_python())
else:
raise BrokenFilePDF(f"Invalid type for initilizing rectangle {obj}")
if len(vals) > 4:
raise TooManyRectField(vals)
elif len(vals) < 4:
raise BrokenFilePDF(
f"Insufficient number of fields for rectangle {arr} {vals}"
)
return cls(*vals)
+71 -1
View File
@@ -1,3 +1,9 @@
from decimal import Decimal
from pdfmajor.document.exceptions import BrokenFilePDF, TooManyRectField
from typing import List
from pdfmajor.parser.objects.base import PDFObject
from pdfmajor.parser.objects.collections import PDFArray
from pdfmajor.document.structures import PDFRectangle
from unittest import TestCase
from pdfmajor.document.PDFDocumentCatalog import PDFDocumentCatalog
@@ -7,12 +13,13 @@ from pdfmajor.parser.objects.primitives import (
PDFBoolean,
PDFInteger,
PDFName,
PDFReal,
PDFString,
)
from tests.util import all_pdf_files
class Catalog(TestCase):
class ParsingState(TestCase):
def run_test(self, file_name: str, expcat: PDFDocumentCatalog):
with all_pdf_files[file_name].get_window() as buffer:
pdf = PDFParsingContext(buffer)
@@ -77,3 +84,66 @@ class Catalog(TestCase):
parser.get_catalog()
self.assertEqual(len(parser.health_report), 0)
parser.get_info()
class Structures(TestCase):
def run_rect_test(self, arr_vals: List[PDFObject], exprect: PDFRectangle):
self.assertEqual(
PDFRectangle.from_pdfarray(PDFArray.from_list(arr_vals)),
exprect,
)
def test_rect(self):
self.run_rect_test(
[
PDFInteger(20, 0, 0),
PDFInteger(10, 0, 0),
PDFInteger(0, 0, 0),
PDFInteger(3, 0, 0),
],
PDFRectangle(Decimal(20), Decimal(10), Decimal(0), Decimal(3)),
)
self.run_rect_test(
[
PDFReal(Decimal("0.23"), 0, 0),
PDFInteger(10, 0, 0),
PDFInteger(Decimal("3.3"), 0, 0),
PDFInteger(3, 0, 0),
],
PDFRectangle(Decimal("0.23"), Decimal(10), Decimal("3.3"), Decimal(3)),
)
def test_rect_fail(self):
with self.assertRaises(BrokenFilePDF):
PDFRectangle.from_pdfarray(
PDFArray.from_list(
[
PDFString("h0", 0, 0),
PDFString("h0", 0, 0),
PDFString("h0", 0, 0),
PDFString("h0", 0, 0),
]
)
)
with self.assertRaises(TooManyRectField):
PDFRectangle.from_pdfarray(
PDFArray.from_list(
[
PDFInteger(3, 0, 0),
PDFInteger(3, 0, 0),
PDFInteger(3, 0, 0),
PDFInteger(3, 0, 0),
PDFInteger(3, 0, 0),
]
)
)
with self.assertRaises(BrokenFilePDF):
PDFRectangle.from_pdfarray(
PDFArray.from_list(
[
PDFInteger(3, 0, 0),
PDFInteger(3, 0, 0),
]
)
)