Advanced Search

Robust pretty-printing for XML-like data

/ Published in: Python
Save to your folder(s)

Expand | Embed | Plain Text

Copy this code and paste it in your HTML 
  1. import re
  2. TAG = r"""
  3.         <\s*
  4.         (?!\s*[!"])
  5.         (?P<close>\s*\/)?
  6.         (?P<name>\w+)
  7.         (
  8.                 [^'"\t >]+
  9.             |
  10.                 "[^\"]*"['\"]*
  11.             |
  12.                 '[^']*'['\"]*
  13.             | 
  14.                 \s+
  15.         )*
  16.         (?P<selfcont>\s*\/\s*)?
  17.         \s*>
  18.       """
  19. UNI = set(["br", "hr", "img", "input", "area", "link"])
  20. INDENT = " "*4
  21. def pretty_xmlish(s):
  22.     """
  23.         This is a robust, general indenter for XML-ish data. 
  24.         Returns a list of lines.
  25.     """
  26.     data, offset, indent, prev = [], 0, 0, None
  27.     for i in re.finditer(TAG, s, re.VERBOSE|re.MULTILINE):
  28.         start, end = i.span()
  29.         name = i.group("name")
  30.         if start > offset:
  31.             txt = []
  32.             for x in textwrap.dedent(s[offset:start]).split("\n"):
  33.                 if x.strip():
  34.                     txt.append(indent*INDENT + x)
  35.             data.extend(txt)
  36.         if i.group("close") and not (name in UNI and name==prev):
  37.             indent = max(indent - 1, 0)
  38.         data.append(indent*INDENT + i.group().strip())
  39.         offset = end
  40.         if not any([i.group("close"), i.group("selfcont"), name in UNI]):
  41.             indent += 1
  42.         prev = name
  43.     return data

Report this snippet


Comments

RSS Icon Subscribe to comments

You need to login to post a comment.