pub/boobies.py

   1 #!/usr/bin/python
   2 # Copyright (C) 2012  Roman Zimbelmann <hut@lavabit.com>
   3 # This software is distributed under the terms of the GNU GPL version 3.
   4 """
   5 boobies.py is a html preprocessor for pre-generated static websites.
   6
   7 It searches your html files for commands in the form of <!-- COMMAND [ARGS] -->
   8 and evaluates them IN-PLACE.  NESTING COMMANDS IS NOT SUPPORTED.
   9 The following commands exist:
  10
  11 run:
  12   <!-- run ls -l -->
  13   <!-- /run -->
  14
  15   This will execute "ls -l" in the shell and insert the output into the block
  16
  17 get:
  18   You can write this in index.html:
  19
  20     <!-- head -->
  21     <head><title>boobies</title></head>
  22     <!-- /head -->
  23
  24   and this in another file:
  25
  26     <!-- get head from index.html -->
  27     foo
  28     <!-- /get -->
  29
  30   and foo will be replaced by <head><title>boobies</title></head>.
  31
  32 relative:
  33   <!-- relative -->
  34   <a href="about.html">about</a>
  35   <!-- /relative -->
  36
  37   If you use this in a "get"-block, it will fix URLs with relative paths.
  38 """
  39
  40 import os.path
  41 import re
  42 import sys
  43 from subprocess import Popen, PIPE, call
  44
  45 PATTERN_GET = r'(<!--\s*get\s+(\w+)\s+from\s+([^\s]+)\s*-->\n?)(.*?)(<!--\s*/get\s*-->)'
  46 PATTERN_RUN = r'(<!--\s*run\s+(.*?)\s*-->(\n?))(.*?)(<!--\s*/run\s*-->)'
  47 PATTERN_RELATIVE = r'(<!--\s*relative\s*-->\n?)(.*?)(<!--\s*/relative\s*-->)'
  48 PATTERN_GETURL = r'href="([^"]+?\.[^"]+?)"'
  49 PATTERN_GETKEYWORD = r'<!--\s*{0}\s*-->\n?(.*?)<!--\s*/{0}\s*-->'
  50
  51 def main():
  52   if '-h' in sys.argv or '--help' in sys.argv or len(sys.argv) <= 1:
  53     return sys.stdout.write("usage: boobies.py [--dry] [-h|--help] files...\n")
  54
  55   for filename in [s for s in sys.argv[1:] if not s.startswith('-')]:
  56     tmpname = filename + ".__new"
  57
  58     f = open(filename, 'r')
  59     content = f.read()
  60     f.close()
  61     processed_content = process(content, filename)
  62     if processed_content == content:
  63       print("unchanged: %s" % filename)
  64       continue
  65
  66     tmp = open(tmpname, "w")
  67     tmp.write(processed_content)
  68     tmp.close()
  69     call(['diff', '-u', filename, tmpname])
  70
  71     if '--dry' in sys.argv:
  72       os.unlink(tmpname)
  73     else:
  74       print("writing: %s" % filename)
  75       os.rename(tmpname, filename)
  76
  77 def process(content, filename):
  78   global PATH
  79   PATH = os.path.dirname(filename)
  80   content = re.sub(PATTERN_GET, _command_get, content, 0, re.S)
  81   content = re.sub(PATTERN_RUN, _command_run, content, 0, re.S)
  82   return content
  83
  84
  85 def _command_get(m):
  86   filename = m.group(3)
  87   keyword = m.group(2)
  88   fallback = m.group(4)
  89
  90   f = open(filename, 'r')
  91   content = f.read()
  92   f.close()
  93   match = re.search(PATTERN_GETKEYWORD.format(keyword), content, re.DOTALL)
  94   if match is None:
  95     sys.stderr.write("Could not find keyword {0} in {1}\n".format(keyword, filename))
  96     result = fallback
  97   else:
  98     result = re.sub(PATTERN_RELATIVE, _command_relative, match.group(1), 0, re.S)
  99
 100   return m.group(1) + result + m.group(5)
 101
 102
 103 def _command_relative(m):
 104   return m.group(1) + re.sub(PATTERN_GETURL,
 105       lambda match: 'href="' + os.path.relpath(match.group(1), PATH) + '"',
 106       m.group(2)) + m.group(3)
 107
 108
 109 def _command_run(m):
 110   output = Popen(m.group(2), shell=True, stdout=PIPE).communicate()[0]
 111   if m.group(3) != "\n":
 112     output = output.strip()
 113   return m.group(1) + output.decode('utf-8') + m.group(5)
 114
 115
 116 if __name__ == '__main__':
 117   main()