1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
|
#!/usr/bin/env python
# Copyright 2014 The PDFium Authors. All rights reserved.
# Use of this source code is governed by a BSD-style license that can be
# found in the LICENSE file.
"""Expands a hand-written PDF testcase (template) into a valid PDF file.
There are several places in a PDF file where byte-offsets are required. This
script replaces {{name}}-style variables in the input with calculated results
{{header}} - expands to the header comment required for PDF files.
{{xref}} - expands to a generated xref table, noting the offset.
{{trailer}} - expands to a standard trailer with "1 0 R" as the /Root.
{{startxref} - expands to a startxref directive followed by correct offset.
{{object x y}} - expands to |x y obj| declaration, noting the offset.
{{streamlen}} - expands to |/Length n|.
"""
import cStringIO
import optparse
import os
import re
import sys
class StreamLenState:
START = 1
FIND_STREAM = 2
FIND_ENDSTREAM = 3
class TemplateProcessor:
HEADER_TOKEN = '{{header}}'
HEADER_REPLACEMENT = '%PDF-1.7\n%\xa0\xf2\xa4\xf4'
XREF_TOKEN = '{{xref}}'
XREF_REPLACEMENT = 'xref\n%d %d\n'
XREF_REPLACEMENT_N = '%010d %05d n \n'
XREF_REPLACEMENT_F = '0000000000 65535 f \n'
# XREF rows must be exactly 20 bytes - space required.
assert(len(XREF_REPLACEMENT_F) == 20)
TRAILER_TOKEN = '{{trailer}}'
TRAILER_REPLACEMENT = 'trailer<< /Root 1 0 R /Size %d >>'
STARTXREF_TOKEN = '{{startxref}}'
STARTXREF_REPLACEMENT = 'startxref\n%d'
OBJECT_PATTERN = r'\{\{object\s+(\d+)\s+(\d+)\}\}'
OBJECT_REPLACEMENT = r'\1 \2 obj'
STREAMLEN_TOKEN = '{{streamlen}}'
STREAMLEN_REPLACEMENT = '/Length %d'
def __init__(self):
self.streamlen_state = StreamLenState.START
self.streamlens = []
self.offset = 0
self.xref_offset = 0
self.max_object_number = 0
self.objects = {}
def insert_xref_entry(self, object_number, generation_number):
self.objects[object_number] = (self.offset, generation_number)
self.max_object_number = max(self.max_object_number, object_number)
def generate_xref_table(self):
result = self.XREF_REPLACEMENT % (0, self.max_object_number + 1)
for i in range(0, self.max_object_number + 1):
if i in self.objects:
result += self.XREF_REPLACEMENT_N % self.objects[i]
else:
result += self.XREF_REPLACEMENT_F
return result
def preprocess_line(self, line):
if self.STREAMLEN_TOKEN in line:
assert(self.streamlen_state == StreamLenState.START)
self.streamlen_state = StreamLenState.FIND_STREAM
self.streamlens.append(0)
return
if (self.streamlen_state == StreamLenState.FIND_STREAM and
line.rstrip() == 'stream'):
self.streamlen_state = StreamLenState.FIND_ENDSTREAM
return
if self.streamlen_state == StreamLenState.FIND_ENDSTREAM:
if line.rstrip() == 'endstream':
self.streamlen_state = StreamLenState.START
else:
self.streamlens[-1] += len(line)
def process_line(self, line):
if self.HEADER_TOKEN in line:
line = line.replace(self.HEADER_TOKEN, self.HEADER_REPLACEMENT)
if self.STREAMLEN_TOKEN in line:
sub = self.STREAMLEN_REPLACEMENT % self.streamlens.pop(0)
line = re.sub(self.STREAMLEN_TOKEN, sub, line)
if self.XREF_TOKEN in line:
self.xref_offset = self.offset
line = self.generate_xref_table()
if self.TRAILER_TOKEN in line:
replacement = self.TRAILER_REPLACEMENT % (self.max_object_number + 1)
line = line.replace(self.TRAILER_TOKEN, replacement)
if self.STARTXREF_TOKEN in line:
replacement = self.STARTXREF_REPLACEMENT % self.xref_offset
line = line.replace(self.STARTXREF_TOKEN, replacement)
match = re.match(self.OBJECT_PATTERN, line)
if match:
self.insert_xref_entry(int(match.group(1)), int(match.group(2)))
line = re.sub(self.OBJECT_PATTERN, self.OBJECT_REPLACEMENT, line)
self.offset += len(line)
return line
def expand_file(input_path, output_path):
processor = TemplateProcessor()
try:
with open(input_path, 'rb') as infile:
with open(output_path, 'wb') as outfile:
preprocessed = cStringIO.StringIO()
for line in infile:
preprocessed.write(line)
processor.preprocess_line(line)
preprocessed.seek(0)
for line in preprocessed:
outfile.write(processor.process_line(line))
except IOError:
print >> sys.stderr, 'failed to process %s' % input_path
def main():
parser = optparse.OptionParser()
parser.add_option('--output-dir', default='')
options, args = parser.parse_args()
for testcase_path in args:
testcase_filename = os.path.basename(testcase_path)
testcase_root, _ = os.path.splitext(testcase_filename)
output_dir = os.path.dirname(testcase_path)
if options.output_dir:
output_dir = options.output_dir
output_path = os.path.join(output_dir, testcase_root + '.pdf')
expand_file(testcase_path, output_path)
return 0
if __name__ == '__main__':
sys.exit(main())
|
