Package lxml :: Module ElementInclude
[hide private]
[frames] | no frames]

Source Code for Module lxml.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  # fredrik@pythonware.com 
 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  """ 
 45  Limited XInclude support for the ElementTree package. 
 46   
 47  While lxml.etree has full support for XInclude (see 
 48  `etree.ElementTree.xinclude()`), this module provides a simpler, pure 
 49  Python, ElementTree compatible implementation that supports a simple 
 50  form of custom URL resolvers. 
 51  """ 
 52   
 53  from lxml import etree 
 54  import copy 
 55  try: 
 56      from urlparse import urljoin 
 57      from urllib2 import urlopen 
 58  except ImportError: 
 59      # Python 3 
 60      from urllib.parse import urljoin 
 61      from urllib.request import urlopen 
 62   
 63  try: 
 64      set 
 65  except NameError: 
 66      # Python 2.3 
 67      from sets import Set as set 
 68   
 69  XINCLUDE = "{http://www.w3.org/2001/XInclude}" 
 70   
 71  XINCLUDE_INCLUDE = XINCLUDE + "include" 
 72  XINCLUDE_FALLBACK = XINCLUDE + "fallback" 
 73   
 74  ## 
 75  # Fatal include error. 
 76   
77 -class FatalIncludeError(etree.LxmlSyntaxError):
78 pass
79 80 ## 81 # ET compatible default loader. 82 # This loader reads an included resource from disk. 83 # 84 # @param href Resource reference. 85 # @param parse Parse mode. Either "xml" or "text". 86 # @param encoding Optional text encoding. 87 # @return The expanded resource. If the parse mode is "xml", this 88 # is an ElementTree instance. If the parse mode is "text", this 89 # is a Unicode string. If the loader fails, it can return None 90 # or raise an IOError exception. 91 # @throws IOError If the loader fails to load the resource. 92
93 -def default_loader(href, parse, encoding=None):
94 file = open(href, 'rb') 95 if parse == "xml": 96 data = etree.parse(file).getroot() 97 else: 98 data = file.read() 99 if not encoding: 100 encoding = 'utf-8' 101 data = data.decode(encoding) 102 file.close() 103 return data
104 105 ## 106 # Default loader used by lxml.etree - handles custom resolvers properly 107 # 108
109 -def _lxml_default_loader(href, parse, encoding=None, parser=None):
110 if parse == "xml": 111 data = etree.parse(href, parser).getroot() 112 else: 113 if "://" in href: 114 f = urlopen(href) 115 else: 116 f = open(href, 'rb') 117 data = f.read() 118 f.close() 119 if not encoding: 120 encoding = 'utf-8' 121 data = data.decode(encoding) 122 return data
123 124 ## 125 # Wrapper for ET compatibility - drops the parser 126
127 -def _wrap_et_loader(loader):
128 def load(href, parse, encoding=None, parser=None): 129 return loader(href, parse, encoding)
130 return load 131 132 133 ## 134 # Expand XInclude directives. 135 # 136 # @param elem Root element. 137 # @param loader Optional resource loader. If omitted, it defaults 138 # to {@link default_loader}. If given, it should be a callable 139 # that implements the same interface as <b>default_loader</b>. 140 # @throws FatalIncludeError If the function fails to include a given 141 # resource, or if the tree contains malformed XInclude elements. 142 # @throws IOError If the function fails to load a given resource. 143 # @returns the node or its replacement if it was an XInclude node 144
145 -def include(elem, loader=None, base_url=None):
146 if base_url is None: 147 if hasattr(elem, 'getroot'): 148 tree = elem 149 elem = elem.getroot() 150 else: 151 tree = elem.getroottree() 152 if hasattr(tree, 'docinfo'): 153 base_url = tree.docinfo.URL 154 elif hasattr(elem, 'getroot'): 155 elem = elem.getroot() 156 _include(elem, loader, base_url=base_url)
157
158 -def _include(elem, loader=None, _parent_hrefs=None, base_url=None):
159 if loader is not None: 160 load_include = _wrap_et_loader(loader) 161 else: 162 load_include = _lxml_default_loader 163 164 if _parent_hrefs is None: 165 _parent_hrefs = set() 166 167 parser = elem.getroottree().parser 168 169 include_elements = list( 170 elem.iter('{http://www.w3.org/2001/XInclude}*')) 171 172 for e in include_elements: 173 if e.tag == XINCLUDE_INCLUDE: 174 # process xinclude directive 175 href = urljoin(base_url, e.get("href")) 176 parse = e.get("parse", "xml") 177 parent = e.getparent() 178 if parse == "xml": 179 if href in _parent_hrefs: 180 raise FatalIncludeError( 181 "recursive include of %r detected" % href 182 ) 183 _parent_hrefs.add(href) 184 node = load_include(href, parse, parser=parser) 185 if node is None: 186 raise FatalIncludeError( 187 "cannot load %r as %r" % (href, parse) 188 ) 189 node = _include(node, loader, _parent_hrefs) 190 if e.tail: 191 node.tail = (node.tail or "") + e.tail 192 if parent is None: 193 return node # replaced the root node! 194 parent.replace(e, node) 195 elif parse == "text": 196 text = load_include(href, parse, encoding=e.get("encoding")) 197 if text is None: 198 raise FatalIncludeError( 199 "cannot load %r as %r" % (href, parse) 200 ) 201 predecessor = e.getprevious() 202 if predecessor is not None: 203 predecessor.tail = (predecessor.tail or "") + text 204 elif parent is None: 205 return text # replaced the root node! 206 else: 207 parent.text = (parent.text or "") + text + (e.tail or "") 208 parent.remove(e) 209 else: 210 raise FatalIncludeError( 211 "unknown parse type in xi:include tag (%r)" % parse 212 ) 213 elif e.tag == XINCLUDE_FALLBACK: 214 parent = e.getparent() 215 if parent is not None and parent.tag != XINCLUDE_INCLUDE: 216 raise FatalIncludeError( 217 "xi:fallback tag must be child of xi:include (%r)" % e.tag 218 ) 219 else: 220 raise FatalIncludeError( 221 "Invalid element found in XInclude namespace (%r)" % e.tag 222 ) 223 return elem
224