« back to all changes in this revision
Viewing changes to elementtree/ElementTree.py
- browse files at revision 473
- compare with another revision
- download diff
- download tarball
- view history from revision 473
- Committer: Martin Pool
- Date: 2005-05-11 06:20:05 UTC
- Revision ID: mbp@sourcefrog.net-20050511062005-297af3451635dae0
- Don't lose first line of command help!
- files added:
- TODO
- contrib
- contrib/bash
- contrib/zsh
- files removed:
- doc/faq.txt
- files modified:
- .bzrignore
added
removed
elementtree/ElementTree.py
1
1
#
2
2
# ElementTree
3
# $Id: ElementTree.py 1862 2004-06-18 07:31:02Z Fredrik $
3
# $Id: ElementTree.py 2326 2005-03-17 07:45:21Z fredrik $
4
4
#
5
5
# light-weight XML support for Python 1.5.2 and later.
6
6
#
7
# this is a stripped-down version of Secret Labs' effDOM library (part
8
# of xmlToolkit). compared to effDOM, this implementation has:
9
#
10
# - no support for observers
11
# - no html-specific extensions (e.g. entity preload)
12
# - no custom entities, doctypes, etc
13
# - no accelerator module
14
#
15
7
# history:
16
8
# 2001-10-20 fl created (from various sources)
17
9
# 2001-11-01 fl return root from parse method
38
30
# 2004-03-28 fl added XMLID helper
39
31
# 2004-06-02 fl added default support to findtext
40
32
# 2004-06-08 fl fixed encoding of non-ascii element/attribute names
33
# 2004-08-23 fl take advantage of post-2.1 expat features
34
# 2005-02-01 fl added iterparse implementation
35
# 2005-03-02 fl fixed iterparse support for pre-2.2 versions
41
36
#
42
# Copyright (c) 1999-2004 by Fredrik Lundh. All rights reserved.
37
# Copyright (c) 1999-2005 by Fredrik Lundh. All rights reserved.
43
38
#
44
39
# fredrik@pythonware.com
45
40
# http://www.pythonware.com
47
42
# --------------------------------------------------------------------
48
43
# The ElementTree toolkit is
49
44
#
50
# Copyright (c) 1999-2004 by Fredrik Lundh
45
# Copyright (c) 1999-2005 by Fredrik Lundh
51
46
#
52
47
# By obtaining, using, and/or copying this software and/or its
53
48
# associated documentation, you agree that you have read, understood,
78
73
"dump",
79
74
"Element", "ElementTree",
80
75
"fromstring",
81
"iselement",
76
"iselement", "iterparse",
82
77
"parse",
83
78
"PI", "ProcessingInstruction",
84
79
"QName",
143
138
# TODO: add support for custom namespace resolvers/default namespaces
144
139
# TODO: add improved support for incremental parsing
145
140
146
VERSION = "1.2"
141
VERSION = "1.2.6"
147
142
148
143
##
149
144
# Internal element class. This class defines the Element interface,
701
696
for k, v in xmlns_items:
702
697
file.write(" %s=\"%s\"" % (_encode(k, encoding),
703
698
_escape_attrib(v, encoding)))
704
if node.text or node:
699
if node.text or len(node):
705
700
file.write(">")
706
701
if node.text:
707
702
file.write(_escape_cdata(node.text, encoding))
865
860
return tree
866
861
867
862
##
863
# Parses an XML document into an element tree incrementally, and reports
864
# what's going on to the user.
865
#
866
# @param source A filename or file object containing XML data.
867
# @param events A list of events to report back. If omitted, only "end"
868
# events are reported.
869
# @return A (event, elem) iterator.
870
871
class iterparse:
872
873
def __init__(self, source, events=None):
874
if not hasattr(source, "read"):
875
source = open(source, "rb")
876
self._file = source
877
self._events = []
878
self._index = 0
879
self.root = self._root = None
880
self._parser = XMLTreeBuilder()
881
# wire up the parser for event reporting
882
parser = self._parser._parser
883
append = self._events.append
884
if events is None:
885
events = ["end"]
886
for event in events:
887
if event == "start":
888
try:
889
parser.ordered_attributes = 1
890
parser.specified_attributes = 1
891
def handler(tag, attrib_in, event=event, append=append,
892
start=self._parser._start_list):
893
append((event, start(tag, attrib_in)))
894
parser.StartElementHandler = handler
895
except AttributeError:
896
def handler(tag, attrib_in, event=event, append=append,
897
start=self._parser._start):
898
append((event, start(tag, attrib_in)))
899
parser.StartElementHandler = handler
900
elif event == "end":
901
def handler(tag, event=event, append=append,
902
end=self._parser._end):
903
append((event, end(tag)))
904
parser.EndElementHandler = handler
905
elif event == "start-ns":
906
def handler(prefix, uri, event=event, append=append):
907
try:
908
uri = _encode(uri, "ascii")
909
except UnicodeError:
910
pass
911
append((event, (prefix or "", uri)))
912
parser.StartNamespaceDeclHandler = handler
913
elif event == "end-ns":
914
def handler(prefix, event=event, append=append):
915
append((event, None))
916
parser.EndNamespaceDeclHandler = handler
917
918
def next(self):
919
while 1:
920
try:
921
item = self._events[self._index]
922
except IndexError:
923
if self._parser is None:
924
self.root = self._root
925
try:
926
raise StopIteration
927
except NameError:
928
raise IndexError
929
# load event buffer
930
del self._events[:]
931
self._index = 0
932
data = self._file.read(16384)
933
if data:
934
self._parser.feed(data)
935
else:
936
self._root = self._parser.close()
937
self._parser = None
938
else:
939
self._index = self._index + 1
940
return item
941
942
try:
943
iter
944
def __iter__(self):
945
return self
946
except NameError:
947
def __getitem__(self, index):
948
return self.next()
949
950
##
868
951
# Parses an XML document from a string constant. This function can
869
952
# be used to embed "XML literals" in Python code.
870
953
#
1025
1108
class XMLTreeBuilder:
1026
1109
1027
1110
def __init__(self, html=0, target=None):
1028
from xml.parsers import expat
1111
try:
1112
from xml.parsers import expat
1113
except ImportError:
1114
raise ImportError(
1115
"No module named expat; use SimpleXMLTreeBuilder instead"
1116
)
1029
1117
self._parser = parser = expat.ParserCreate(None, "}")
1030
1118
if target is None:
1031
1119
target = TreeBuilder()
1032
1120
self._target = target
1033
1121
self._names = {} # name memo cache
1034
parser.DefaultHandler = self._default
1122
# callbacks
1123
parser.DefaultHandlerExpand = self._default
1035
1124
parser.StartElementHandler = self._start
1036
1125
parser.EndElementHandler = self._end
1037
1126
parser.CharacterDataHandler = self._data
1127
# let expat do the buffering, if supported
1128
try:
1129
self._parser.buffer_text = 1
1130
except AttributeError:
1131
pass
1132
# use new-style attribute handling, if supported
1133
try:
1134
self._parser.ordered_attributes = 1
1135
self._parser.specified_attributes = 1
1136
parser.StartElementHandler = self._start_list
1137
except AttributeError:
1138
pass
1038
1139
encoding = None
1039
1140
if not parser.returns_unicode:
1040
1141
encoding = "utf-8"
1045
1146
def _fixtext(self, text):
1046
1147
# convert text string to ascii, if possible
1047
1148
try:
1048
return str(text) # what if the default encoding is changed?
1149
return _encode(text, "ascii")
1049
1150
except UnicodeError:
1050
1151
return text
1051
1152
1068
1169
attrib[fixname(key)] = self._fixtext(value)
1069
1170
return self._target.start(tag, attrib)
1070
1171
1172
def _start_list(self, tag, attrib_in):
1173
fixname = self._fixname
1174
tag = fixname(tag)
1175
attrib = {}
1176
if attrib_in:
1177
for i in range(0, len(attrib_in), 2):
1178
attrib[fixname(attrib_in[i])] = self._fixtext(attrib_in[i+1])
1179
return self._target.start(tag, attrib)
1180
1071
1181
def _data(self, text):
1072
1182
return self._target.data(self._fixtext(text))
1073
1183
Loggerhead is a web-based interface for Breezy
Version: 2.0.1