twisted/python/htmlizer.py

   1 # -*- test-case-name: twisted.python.test.test_htmlizer -*-
   2 # Copyright (c) Twisted Matrix Laboratories.
   3 # See LICENSE for details.
   4
   5 """
   6 HTML rendering of Python source.
   7 """
   8
   9 import tokenize, cgi, keyword
  10 import reflect
  11
  12 class TokenPrinter:
  13
  14     currentCol, currentLine = 0, 1
  15     lastIdentifier = parameters = 0
  16
  17     def __init__(self, writer):
  18         self.writer = writer
  19
  20     def printtoken(self, type, token, (srow, scol), (erow, ecol), line):
  21         #print "printtoken(%r,%r,%r,(%r,%r),(%r,%r),%r), row=%r,col=%r" % (
  22         #    self, type, token, srow,scol, erow,ecol, line,
  23         #    self.currentLine, self.currentCol)
  24         if self.currentLine < srow:
  25             self.writer('\n'*(srow-self.currentLine))
  26             self.currentLine, self.currentCol = srow, 0
  27         self.writer(' '*(scol-self.currentCol))
  28         if self.lastIdentifier:
  29             type = "identifier"
  30             self.parameters = 1
  31         elif type == tokenize.NAME:
  32              if keyword.iskeyword(token):
  33                  type = 'keyword'
  34              else:
  35                  if self.parameters:
  36                      type = 'parameter'
  37                  else:
  38                      type = 'variable'
  39         else:
  40             type = tokenize.tok_name.get(type).lower()
  41         self.writer(token, type)
  42         self.currentCol = ecol
  43         self.currentLine += token.count('\n')
  44         if self.currentLine != erow:
  45             self.currentCol = 0
  46         self.lastIdentifier = token in ('def', 'class')
  47         if token == ':':
  48             self.parameters = 0
  49
  50
  51 class HTMLWriter:
  52
  53     noSpan = []
  54
  55     def __init__(self, writer):
  56         self.writer = writer
  57         noSpan = []
  58         reflect.accumulateClassList(self.__class__, "noSpan", noSpan)
  59         self.noSpan = noSpan
  60
  61     def write(self, token, type=None):
  62         token = cgi.escape(token)
  63         if (type is None) or (type in self.noSpan):
  64             self.writer(token)
  65         else:
  66             self.writer('<span class="py-src-%s">%s</span>' %
  67                         (type, token))
  68
  69
  70 class SmallerHTMLWriter(HTMLWriter):
  71     """HTMLWriter that doesn't generate spans for some junk.
  72
  73     Results in much smaller HTML output.
  74     """
  75     noSpan = ["endmarker", "indent", "dedent", "op", "newline", "nl"]
  76
  77 def filter(inp, out, writer=HTMLWriter):
  78     out.write('<pre>')
  79     printer = TokenPrinter(writer(out.write).write).printtoken
  80     try:
  81         tokenize.tokenize(inp.readline, printer)
  82     except tokenize.TokenError:
  83         pass
  84     out.write('</pre>\n')
  85
  86 def main():
  87     import sys
  88     filter(open(sys.argv[1]), sys.stdout)
  89
  90 if __name__ == '__main__':
  91    main()