Source Code for Module qm.xmlutil

  1  ######################################################################## 
  2  # 
  3  # File:   xmlutil.py 
  4  # Author: Alex Samuel 
  5  # Date:   2001-03-18 
  6  # 
  7  # Contents: 
  8  #   Miscellaneous XML-related functions. 
  9  # 
 10  # Copyright (c) 2001, 2002 by CodeSourcery, LLC.  All rights reserved.  
 11  # 
 12  # For license terms see the file COPYING. 
 13  # 
 14  ######################################################################## 
 15   
 16  ######################################################################## 
 17  # imports 
 18  ######################################################################## 
 19   
 20  import os 
 21  import qm 
 22  import re 
 23  import xml.dom 
 24  import xml.dom.minidom 
 25   
 26  ######################################################################## 
 27  # functions 
 28  ######################################################################## 
 29   
 30 -def make_public_id(name): 
 31      """Return a public ID for the DTD with the given 'name'. 
 32   
 33      'name' -- The name of the DTD. 
 34   
 35      returns -- A public ID for the DTD.""" 
 36   
 37      return "-//QM/%s/%s//EN" % (qm.version, name) 
 38   
 39       
 40 -def make_system_id(name): 
 41      """Return a system ID for the DTD with the given 'name'. 
 42   
 43      'name' -- The name of the DTD, as a relative UNIX path. 
 44   
 45      returns -- A URL for the DTD.""" 
 46   
 47      return "http://www.codesourcery.com/qm/dtds/%s/%s" % (qm.version, name) 
 48   
 49   
 50 -def load_xml_file(path): 
 51      """Return a DOM document loaded from the XML file 'path'.""" 
 52   
 53      # Open the file. 
 54      file = open(path, "r") 
 55      return load_xml(file) 
 56   
 57   
 58 -def load_xml(file): 
 59      """Return a DOM document loaded from the XML file object 'file'. 
 60   
 61      'file' -- A file object, opened for reading. 
 62       
 63      returns -- The DOM document contained in 'file'. 
 64       
 65      This function closes 'file', whether or not reading the document was 
 66      successful.""" 
 67   
 68      try: 
 69          document = xml.dom.minidom.parse(file) 
 70      finally: 
 71          file.close() 
 72      return document 
 73   
 74   
 75 -def get_dom_text(node): 
 76      """Return the text contained in DOM 'node'. 
 77   
 78      'node' -- A DOM element node. 
 79   
 80      prerequisites -- 'node' is an element node with exactly one child, 
 81      which is a text node.""" 
 82   
 83      assert node.nodeType == xml.dom.Node.ELEMENT_NODE 
 84      # Normalize the node so that multiple TEXT_NODEs are collapsed into 
 85      # a single node. 
 86      node.normalize() 
 87      # If there are no children, the string is empty. 
 88      if len(node.childNodes) == 0: 
 89          return "" 
 90      # If there is a child, there should be only one. 
 91      if len(node.childNodes) != 1: 
 92          raise QMException, "Invalid XML text node." 
 93      child = node.childNodes[0] 
 94      if child.nodeType != xml.dom.Node.TEXT_NODE: 
 95          raise QMException, "Invalid XML text node." 
 96      return child.data 
 97   
 98   
 99 -def child_tag_predicate(child_tag): 
100      """Return a predicate function for finding element nodes by tag. 
101   
102      returns -- A predicate function that takes a node as its argument 
103      and returns true if the node is an element node whose tag is 
104      'child_tag'.""" 
105   
106      return lambda node, tag=child_tag: \ 
107             node.nodeType == xml.dom.Node.ELEMENT_NODE \ 
108             and node.tagName == tag 
109   
110   
111 -def get_child(node, child_tag): 
112      """Return the child element node of 'node' whose tag is 'child_tag'. 
113   
114      'node' -- A DOM node.  It must have exactly one element child with 
115      the tag 'child_tag'. 
116   
117      'child_tag' -- The desired element tag. 
118   
119      returns -- A child DOM node of 'node'. 
120   
121      raises -- 'KeyError' if 'node' has no element child with tag 
122      'child_tag', or more than one.. """ 
123   
124      matching_children = \ 
125          filter(child_tag_predicate(child_tag), node.childNodes) 
126      if len(matching_children) != 1: 
127          raise KeyError, child_tag 
128      return matching_children[0] 
129   
130   
131 -def get_child_text(node, child_tag, default=None): 
132      """Return the text contained in a child of DOM 'node'. 
133   
134      'child_tag' -- The tag of the child node whose text is to be 
135      retrieved. 
136   
137      'default' -- If 'node' has no child element with tag 'child_tag', 
138      returns 'default', unless 'default' is 'None'. 
139   
140      raises -- 'KeyError' if 'default' is 'None' and 'node' has no child 
141      element with tag 'child_tag'.""" 
142   
143      try: 
144          return get_dom_text(get_child(node, child_tag)) 
145      except KeyError: 
146          if default is not None: 
147              return default 
148          else: 
149              raise 
150   
151   
152 -def get_child_texts(node, child_tag): 
153      """Return a sequence of text contents of children. 
154   
155      'node' -- A DOM node. 
156   
157      returns -- The list containing all child nodes of 'node' which have 
158      tag 'child_tag'.  Each child must have exactly one child of its own, 
159      which must be a text node.""" 
160   
161      return map(get_dom_text, node.getElementsByTagName(child_tag)) 
162   
163   
164 -def create_dom_text_element(document, tag, text): 
165      """Return a DOM element containing a single text node. 
166   
167      'document' -- The containing DOM document. 
168   
169      'tag' -- The element tag. 
170   
171      'text' -- The text contents of the text node.""" 
172   
173      element = document.createElement(tag) 
174      if text != "": 
175          text_node = document.createTextNode(text) 
176          element.appendChild(text_node) 
177      else: 
178          # Don't create a child node in this case.  For some reason, it 
179          # gets written out with an extraneous newline, and therefore 
180          # when the text is read in, its no longer an empty string. 
181          pass 
182      return element 
183   
184   
185  __dom_implementation = xml.dom.minidom.getDOMImplementation() 
186   
187 -def create_dom_document(public_id, document_element_tag): 
188      """Create a DOM document. 
189   
190      'public_id' -- The (partial) public ID for the DTD. 
191   
192      'document_element_tag' -- The tag of the main document element. 
193   
194      returns -- A DOM document node.""" 
195   
196      public_id = make_public_id(public_id) 
197      system_id = make_system_id(public_id.lower() + ".dtd") 
198      # Create the document type for the XML document. 
199      document_type = __dom_implementation.createDocumentType( 
200          qualifiedName=document_element_tag, 
201          publicId=public_id, 
202          systemId=system_id 
203          ) 
204      # Create a new DOM document. 
205      document = __dom_implementation.\ 
206                      createDocument(namespaceURI=None, 
207                                     qualifiedName=document_element_tag, 
208                                     doctype=document_type) 
209      return document 
210   
211   
212  __hyphen_regex = re.compile("(--+)") 
213   
214 -def __hyphen_replacement(match): 
215      return "-" + " -" * (len(match.group(0)) - 1) 
216   
217   
218 -def sanitize_text_for_comment(text): 
219      """Return 'text' modified so that it is valid for an XML comment.""" 
220   
221      # A comment cannot contain two or more hyphens in a row. 
222      text = __hyphen_regex.sub(__hyphen_replacement, text) 
223   
224      return text 
225   
226   
227  ######################################################################## 
228  # Local Variables: 
229  # mode: python 
230  # indent-tabs-mode: nil 
231  # fill-column: 72 
232  # End: 
233