1 #!/usr/bin/env python
   2 #coding: utf-8
   3 
   4 # Author: Firederich Lung, João S. O. Bueno
   5 
   6 # Copyright (c) 1999-2005 by Fredrik Lundh (geração de XML)
   7 #Copyrigjt (c) 2009 Fundação CPqD - Formatação do XML e partes auxiliares do script
   8 #
   9 # By obtaining, using, and/or copying this software and/or its
  10 # associated documentation, you agree that you have read, understood,
  11 # and will comply with the following terms and conditions:
  12 #
  13 # Permission to use, copy, modify, and distribute this software and
  14 # its associated documentation for any purpose and without fee is
  15 # hereby granted, provided that the above copyright notice appears in
  16 # all copies, and that both that copyright notice and this permission
  17 # notice appear in supporting documentation, and that the name of
  18 # Secret Labs AB or the author not be used in advertising or publicity
  19 # pertaining to distribution of the software without specific, written
  20 # prior permission.
  21 #
  22 # SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD
  23 # TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANT-
  24 # ABILITY AND FITNESS.  IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR
  25 # BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY
  26 # DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
  27 # WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
  28 # ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
  29 # OF THIS SOFTWARE.
  30 # --------------------------------------------------------------------
  31 
  32 # Licensed to PSF under a Contributor Agreement.
  33 # See http://www.python.org/2.4/license for licensing details.
  34 
  35 """
  36  Modifies the XML generator in ElementTree in order to create
  37  a "human-readable", indentend by level, XML output.
  38  Can be used as a stand alone script, or to change ElementTree XML generation
  39  on the fly  through a call to  "monkey_patch_etree" function.
  40 
  41 """
  42 import xml.etree.ElementTree
  43 from xml.etree.ElementTree import Comment, _encode, ProcessingInstruction,\
  44                            QName, fixtag, _escape_attrib, _escape_cdata
  45 
  46 def Etree_pretty__write(self, file, node, encoding, namespaces,
  47                         level=0, identator="    "):
  48     # write XML to file
  49     tag = node.tag
  50     if tag is Comment:
  51         file.write(level * identator + "<!-- %s -->" % _escape_cdata(node.text, encoding))
  52     elif tag is ProcessingInstruction:
  53         file.write("<?%s?>" % _escape_cdata(node.text, encoding))
  54     else:
  55         items = node.items()
  56         xmlns_items = [] # new namespaces in this scope
  57         try:
  58             if isinstance(tag, QName) or tag[:1] == "{":
  59                 tag, xmlns = fixtag(tag, namespaces)
  60                 if xmlns: xmlns_items.append(xmlns)
  61         except TypeError:
  62             _raise_serialization_error(tag)
  63         file.write("\n" + level * identator + "<" + _encode(tag, encoding))
  64         if items or xmlns_items:
  65             items.sort() # lexical order
  66             for k, v in items:
  67                 try:
  68                     if isinstance(k, QName) or k[:1] == "{":
  69                         k, xmlns = fixtag(k, namespaces)
  70                         if xmlns: xmlns_items.append(xmlns)
  71                 except TypeError:
  72                     _raise_serialization_error(k)
  73                 try:
  74                     if isinstance(v, QName):
  75                         v, xmlns = fixtag(v, namespaces)
  76                         if xmlns: xmlns_items.append(xmlns)
  77                 except TypeError:
  78                     _raise_serialization_error(v)
  79                 file.write(" %s=\"%s\"" % (_encode(k, encoding),
  80                                             _escape_attrib(v, encoding)))
  81             for k, v in xmlns_items:
  82                 file.write(" %s=\"%s\"" % (_encode(k, encoding),
  83                                             _escape_attrib(v, encoding)))
  84         if node.text or len(node):
  85             file.write(">")
  86             if node.text:
  87                 file.write(_escape_cdata(node.text.replace("\n", (level + 1) * identator + "\n"), encoding))
  88             for n in node:
  89                 self._write(file, n, encoding, namespaces, level + 1, identator)
  90             file.write("\n" + level * identator + "</" + _encode(tag, encoding) + ">")
  91         else:
  92             file.write(" />")
  93         for k, v in xmlns_items:
  94             del namespaces[v]
  95     if node.tail:
  96         file.write(_escape_cdata(node.tail.replace("\n", level * identator + "\n"), encoding))
  97 
  98 
  99 
 100 original__write = xml.etree.ElementTree.ElementTree._write
 101 def monkey_patch_etree():
 102     """
 103        Call this method to overwrite python's native library
 104        xml.etree.ElementTree.ElementTree _write method in order to produce
 105        humam-pleasant formated output
 106     """
 107     xml.etree.ElementTree.ElementTree._write = Etree_pretty__write
 108 
 109 def un_monkey_patch_etree():
 110     """Use to restore default behavior to ElementTree's XML generation"""
 111     xml.etree.ElementTree.ElementTree._write = original__write
 112 
 113 def de_tokenize(text):
 114     """removes html tokens from the "&#195;" type """
 115     r = re.compile(r"\&\#([0-9]{1,3})\;")
 116     caracteres = set(r.findall(text))
 117     textn = text
 118     for caracter in caracteres:
 119         caracter = int(caracter)
 120         r = re.compile(r"\&\#%d\;" % caracter)
 121         textn = r.subn(chr(caracter), textn)[0]
 122     return textn
 123 
 124 
 125 
 126 if __name__ == "__main__":
 127     import sys, re
 128     if len(sys.argv) != 3:
 129         sys.stderr.write("Usage: prettyxml.py <inputfile> <outputfile>\n")
 130         sys.exit(1)
 131 
 132     infile = open(sys.argv[1], "rt")
 133     outfile = open(sys.argv[2], "wt")
 134     monkey_patch_etree()
 135     xmldata = xml.etree.ElementTree.parse(infile)
 136     xmlstring = de_tokenize(xml.etree.ElementTree.tostring(xmldata.getroot()))
 137     outfile.write(xmlstring)
 138     outfile.close()