Source Code for Module nltk.etree.ElementInclude

  1  # 
  2  # ElementTree 
  3  # $Id: ElementInclude.py 1862 2004-06-18 07:31:02Z Fredrik $ 
  4  # 
  5  # limited xinclude support for element trees 
  6  # 
  7  # history: 
  8  # 2003-08-15 fl   created 
  9  # 2003-11-14 fl   fixed default loader 
 10  # 
 11  # Copyright (c) 2003-2004 by Fredrik Lundh.  All rights reserved. 
 12  # 
 13  # [email protected] 
 14  # http://www.pythonware.com 
 15  # 
 16  # -------------------------------------------------------------------- 
 17  # The ElementTree toolkit is 
 18  # 
 19  # Copyright (c) 1999-2004 by Fredrik Lundh 
 20  # 
 21  # By obtaining, using, and/or copying this software and/or its 
 22  # associated documentation, you agree that you have read, understood, 
 23  # and will comply with the following terms and conditions: 
 24  # 
 25  # Permission to use, copy, modify, and distribute this software and 
 26  # its associated documentation for any purpose and without fee is 
 27  # hereby granted, provided that the above copyright notice appears in 
 28  # all copies, and that both that copyright notice and this permission 
 29  # notice appear in supporting documentation, and that the name of 
 30  # Secret Labs AB or the author not be used in advertising or publicity 
 31  # pertaining to distribution of the software without specific, written 
 32  # prior permission. 
 33  # 
 34  # SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD 
 35  # TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANT- 
 36  # ABILITY AND FITNESS.  IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR 
 37  # BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY 
 38  # DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, 
 39  # WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS 
 40  # ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE 
 41  # OF THIS SOFTWARE. 
 42  # -------------------------------------------------------------------- 
 43   
 44  # Licensed to PSF under a Contributor Agreement. 
 45  # See http://www.python.org/2.4/license for licensing details. 
 46   
 47  ## 
 48  # Limited XInclude support for the ElementTree package. 
 49  ## 
 50   
 51  import copy 
 52  import ElementTree 
 53   
 54  XINCLUDE = "{http://www.w3.org/2001/XInclude}" 
 55   
 56  XINCLUDE_INCLUDE = XINCLUDE + "include" 
 57  XINCLUDE_FALLBACK = XINCLUDE + "fallback" 
 58   
 59  ## 
 60  # Fatal include error. 
 61   
 62 -class FatalIncludeError(SyntaxError): 
 63      pass 
 64   
 65  ## 
 66  # Default loader.  This loader reads an included resource from disk. 
 67  # 
 68  # @param href Resource reference. 
 69  # @param parse Parse mode.  Either "xml" or "text". 
 70  # @param encoding Optional text encoding. 
 71  # @return The expanded resource.  If the parse mode is "xml", this 
 72  #    is an ElementTree instance.  If the parse mode is "text", this 
 73  #    is a Unicode string.  If the loader fails, it can return None 
 74  #    or raise an IOError exception. 
 75  # @throws IOError If the loader fails to load the resource. 
 76   
 77 -def default_loader(href, parse, encoding=None): 
 78      file = open(href) 
 79      if parse == "xml": 
 80          data = ElementTree.parse(file).getroot() 
 81      else: 
 82          data = file.read() 
 83          if encoding: 
 84              data = data.decode(encoding) 
 85      file.close() 
 86      return data 
 87   
 88  ## 
 89  # Expand XInclude directives. 
 90  # 
 91  # @param elem Root element. 
 92  # @param loader Optional resource loader.  If omitted, it defaults 
 93  #     to {@link default_loader}.  If given, it should be a callable 
 94  #     that implements the same interface as <b>default_loader</b>. 
 95  # @throws FatalIncludeError If the function fails to include a given 
 96  #     resource, or if the tree contains malformed XInclude elements. 
 97  # @throws IOError If the function fails to load a given resource. 
 98   
 99 -def include(elem, loader=None): 
100      if loader is None: 
101          loader = default_loader 
102      # look for xinclude elements 
103      i = 0 
104      while i < len(elem): 
105          e = elem[i] 
106          if e.tag == XINCLUDE_INCLUDE: 
107              # process xinclude directive 
108              href = e.get("href") 
109              parse = e.get("parse", "xml") 
110              if parse == "xml": 
111                  node = loader(href, parse) 
112                  if node is None: 
113                      raise FatalIncludeError( 
114                          "cannot load %r as %r" % (href, parse) 
115                          ) 
116                  node = copy.copy(node) 
117                  if e.tail: 
118                      node.tail = (node.tail or "") + e.tail 
119                  elem[i] = node 
120              elif parse == "text": 
121                  text = loader(href, parse, e.get("encoding")) 
122                  if text is None: 
123                      raise FatalIncludeError( 
124                          "cannot load %r as %r" % (href, parse) 
125                          ) 
126                  if i: 
127                      node = elem[i-1] 
128                      node.tail = (node.tail or "") + text 
129                  else: 
130                      elem.text = (elem.text or "") + text + (e.tail or "") 
131                  del elem[i] 
132                  continue 
133              else: 
134                  raise FatalIncludeError( 
135                      "unknown parse type in xi:include tag (%r)" % parse 
136                  ) 
137          elif e.tag == XINCLUDE_FALLBACK: 
138              raise FatalIncludeError( 
139                  "xi:fallback tag must be child of xi:include (%r)" % e.tag 
140                  ) 
141          else: 
142              include(e, loader) 
143          i = i + 1 
144