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

Source Code for Module lxml.sax

  1  """ 
  2  SAX-based adapter to copy trees from/to the Python standard library. 
  3   
  4  Use the `ElementTreeContentHandler` class to build an ElementTree from 
  5  SAX events. 
  6   
  7  Use the `ElementTreeProducer` class or the `saxify()` function to fire 
  8  the SAX events of an ElementTree against a SAX ContentHandler. 
  9   
 10  See http://codespeak.net/lxml/sax.html 
 11  """ 
 12   
 13  from xml.sax.handler import ContentHandler 
 14  from lxml import etree 
 15  from lxml.etree import ElementTree, SubElement 
 16  from lxml.etree import Comment, ProcessingInstruction 
 17   
18 -class SaxError(etree.LxmlError):
19 """General SAX error. 20 """ 21 pass
22
23 -def _getNsTag(tag):
24 if tag[0] == '{': 25 return tuple(tag[1:].split('}', 1)) 26 else: 27 return (None, tag)
28 29
30 -class ElementTreeContentHandler(ContentHandler):
31 """Build an lxml ElementTree from SAX events. 32 """
33 - def __init__(self, makeelement=None):
34 self._root = None 35 self._root_siblings = [] 36 self._element_stack = [] 37 self._default_ns = None 38 self._ns_mapping = { None : [None] } 39 self._new_mappings = {} 40 if makeelement is None: 41 makeelement = etree.Element 42 self._makeelement = makeelement
43
44 - def _get_etree(self):
45 "Contains the generated ElementTree after parsing is finished." 46 return ElementTree(self._root)
47 48 etree = property(_get_etree, doc=_get_etree.__doc__) 49
50 - def setDocumentLocator(self, locator):
51 pass
52
53 - def startDocument(self):
54 pass
55
56 - def endDocument(self):
57 pass
58
59 - def startPrefixMapping(self, prefix, uri):
60 self._new_mappings[prefix] = uri 61 try: 62 self._ns_mapping[prefix].append(uri) 63 except KeyError: 64 self._ns_mapping[prefix] = [uri] 65 if prefix is None: 66 self._default_ns = uri
67
68 - def endPrefixMapping(self, prefix):
69 ns_uri_list = self._ns_mapping[prefix] 70 ns_uri_list.pop() 71 if prefix is None: 72 self._default_ns = ns_uri_list[-1]
73
74 - def _buildTag(self, ns_name_tuple):
75 ns_uri, local_name = ns_name_tuple 76 if ns_uri: 77 el_tag = "{%s}%s" % ns_name_tuple 78 elif self._default_ns: 79 el_tag = "{%s}%s" % (self._default_ns, local_name) 80 else: 81 el_tag = local_name 82 return el_tag
83
84 - def startElementNS(self, ns_name, qname, attributes=None):
85 el_name = self._buildTag(ns_name) 86 if attributes: 87 attrs = {} 88 try: 89 iter_attributes = attributes.iteritems() 90 except AttributeError: 91 iter_attributes = attributes.items() 92 93 for name_tuple, value in iter_attributes: 94 if name_tuple[0]: 95 attr_name = "{%s}%s" % name_tuple 96 else: 97 attr_name = name_tuple[1] 98 attrs[attr_name] = value 99 else: 100 attrs = None 101 102 element_stack = self._element_stack 103 if self._root is None: 104 element = self._root = \ 105 self._makeelement(el_name, attrs, self._new_mappings) 106 if self._root_siblings and hasattr(element, 'addprevious'): 107 for sibling in self._root_siblings: 108 element.addprevious(sibling) 109 del self._root_siblings[:] 110 else: 111 element = SubElement(element_stack[-1], el_name, 112 attrs, self._new_mappings) 113 element_stack.append(element) 114 115 self._new_mappings.clear()
116
117 - def processingInstruction(self, target, data):
118 pi = ProcessingInstruction(target, data) 119 if self._root is None: 120 self._root_siblings.append(pi) 121 else: 122 self._element_stack[-1].append(pi)
123
124 - def endElementNS(self, ns_name, qname):
125 element = self._element_stack.pop() 126 el_tag = self._buildTag(ns_name) 127 if el_tag != element.tag: 128 raise SaxError("Unexpected element closed: " + el_tag)
129
130 - def startElement(self, name, attributes=None):
131 if attributes: 132 attributes = dict( 133 [((None, k), v) for k, v in attributes.items()] 134 ) 135 self.startElementNS((None, name), name, attributes)
136
137 - def endElement(self, name):
138 self.endElementNS((None, name), name)
139
140 - def characters(self, data):
141 last_element = self._element_stack[-1] 142 try: 143 # if there already is a child element, we must append to its tail 144 last_element = last_element[-1] 145 last_element.tail = (last_element.tail or '') + data 146 except IndexError: 147 # otherwise: append to the text 148 last_element.text = (last_element.text or '') + data
149 150 ignorableWhitespace = characters
151 152
153 -class ElementTreeProducer(object):
154 """Produces SAX events for an element and children. 155 """
156 - def __init__(self, element_or_tree, content_handler):
157 try: 158 element = element_or_tree.getroot() 159 except AttributeError: 160 element = element_or_tree 161 self._element = element 162 self._content_handler = content_handler 163 from xml.sax.xmlreader import AttributesNSImpl as attr_class 164 self._attr_class = attr_class 165 self._empty_attributes = attr_class({}, {})
166
167 - def saxify(self):
168 self._content_handler.startDocument() 169 170 element = self._element 171 if hasattr(element, 'getprevious'): 172 siblings = [] 173 sibling = element.getprevious() 174 while getattr(sibling, 'tag', None) is ProcessingInstruction: 175 siblings.append(sibling) 176 sibling = sibling.getprevious() 177 for sibling in siblings[::-1]: 178 self._recursive_saxify(sibling, {}) 179 180 self._recursive_saxify(element, {}) 181 182 if hasattr(element, 'getnext'): 183 sibling = element.getnext() 184 while getattr(sibling, 'tag', None) is ProcessingInstruction: 185 self._recursive_saxify(sibling, {}) 186 sibling = sibling.getnext() 187 188 self._content_handler.endDocument()
189
190 - def _recursive_saxify(self, element, prefixes):
191 content_handler = self._content_handler 192 tag = element.tag 193 if tag is Comment or tag is ProcessingInstruction: 194 if tag is ProcessingInstruction: 195 content_handler.processingInstruction( 196 element.target, element.text) 197 if element.tail: 198 content_handler.characters(element.tail) 199 return 200 201 new_prefixes = [] 202 build_qname = self._build_qname 203 attribs = element.items() 204 if attribs: 205 attr_values = {} 206 attr_qnames = {} 207 for attr_ns_name, value in attribs: 208 attr_ns_tuple = _getNsTag(attr_ns_name) 209 attr_values[attr_ns_tuple] = value 210 attr_qnames[attr_ns_tuple] = build_qname( 211 attr_ns_tuple[0], attr_ns_tuple[1], prefixes, new_prefixes) 212 sax_attributes = self._attr_class(attr_values, attr_qnames) 213 else: 214 sax_attributes = self._empty_attributes 215 216 ns_uri, local_name = _getNsTag(tag) 217 qname = build_qname(ns_uri, local_name, prefixes, new_prefixes) 218 219 for prefix, uri in new_prefixes: 220 content_handler.startPrefixMapping(prefix, uri) 221 content_handler.startElementNS((ns_uri, local_name), 222 qname, sax_attributes) 223 if element.text: 224 content_handler.characters(element.text) 225 for child in element: 226 self._recursive_saxify(child, prefixes) 227 content_handler.endElementNS((ns_uri, local_name), qname) 228 for prefix, uri in new_prefixes: 229 content_handler.endPrefixMapping(prefix) 230 if element.tail: 231 content_handler.characters(element.tail)
232
233 - def _build_qname(self, ns_uri, local_name, prefixes, new_prefixes):
234 if ns_uri is None: 235 return local_name 236 try: 237 prefix = prefixes[ns_uri] 238 except KeyError: 239 prefix = prefixes[ns_uri] = 'ns%02d' % len(prefixes) 240 new_prefixes.append( (prefix, ns_uri) ) 241 return prefix + ':' + local_name
242
243 -def saxify(element_or_tree, content_handler):
244 """One-shot helper to generate SAX events from an XML tree and fire 245 them against a SAX ContentHandler. 246 """ 247 return ElementTreeProducer(element_or_tree, content_handler).saxify()
248