libs/elementtree/SgmlopXMLTreeBuilder.py

   1 #
   2 # ElementTree
   3 # $Id$
   4 #
   5 # A simple XML tree builder, based on the sgmlop library.
   6 #
   7 # Note that this version does not support namespaces.  This may be
   8 # changed in future versions.
   9 #
  10 # history:
  11 # 2004-03-28 fl   created
  12 #
  13 # Copyright (c) 1999-2004 by Fredrik Lundh.  All rights reserved.
  14 #
  15 # fredrik@pythonware.com
  16 # http://www.pythonware.com
  17 #
  18 # --------------------------------------------------------------------
  19 # The ElementTree toolkit is
  20 #
  21 # Copyright (c) 1999-2004 by Fredrik Lundh
  22 #
  23 # By obtaining, using, and/or copying this software and/or its
  24 # associated documentation, you agree that you have read, understood,
  25 # and will comply with the following terms and conditions:
  26 #
  27 # Permission to use, copy, modify, and distribute this software and
  28 # its associated documentation for any purpose and without fee is
  29 # hereby granted, provided that the above copyright notice appears in
  30 # all copies, and that both that copyright notice and this permission
  31 # notice appear in supporting documentation, and that the name of
  32 # Secret Labs AB or the author not be used in advertising or publicity
  33 # pertaining to distribution of the software without specific, written
  34 # prior permission.
  35 #
  36 # SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD
  37 # TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANT-
  38 # ABILITY AND FITNESS.  IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR
  39 # BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY
  40 # DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
  41 # WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
  42 # ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
  43 # OF THIS SOFTWARE.
  44 # --------------------------------------------------------------------
  45
  46 ##
  47 # Tools to build element trees from XML, based on the SGMLOP parser.
  48 # <p>
  49 # The current version does not support XML namespaces.
  50 # <p>
  51 # This tree builder requires the <b>sgmlop</b> extension module
  52 # (available from
  53 # <a href='http://effbot.org/downloads'>http://effbot.org/downloads</a>).
  54 ##
  55
  56 import ElementTree
  57
  58 ##
  59 # ElementTree builder for XML source data, based on the SGMLOP parser.
  60 #
  61 # @see elementtree.ElementTree
  62
  63 class TreeBuilder:
  64
  65     def __init__(self, html=0):
  66         try:
  67             import sgmlop
  68         except ImportError:
  69             raise RuntimeError("sgmlop parser not available")
  70         self.__builder = ElementTree.TreeBuilder()
  71         if html:
  72             import htmlentitydefs
  73             self.entitydefs.update(htmlentitydefs.entitydefs)
  74         self.__parser = sgmlop.XMLParser()
  75         self.__parser.register(self)
  76
  77     ##
  78     # Feeds data to the parser.
  79     #
  80     # @param data Encoded data.
  81
  82     def feed(self, data):
  83         self.__parser.feed(data)
  84
  85     ##
  86     # Finishes feeding data to the parser.
  87     #
  88     # @return An element structure.
  89     # @defreturn Element
  90
  91     def close(self):
  92         self.__parser.close()
  93         self.__parser = None
  94         return self.__builder.close()
  95
  96     def finish_starttag(self, tag, attrib):
  97         self.__builder.start(tag, attrib)
  98
  99     def finish_endtag(self, tag):
 100         self.__builder.end(tag)
 101
 102     def handle_data(self, data):
 103         self.__builder.data(data)