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