| 1 |
import io |
| 2 |
import unittest |
| 3 |
import xml.sax |
| 4 |
|
| 5 |
from xml.sax.xmlreader import AttributesImpl |
| 6 |
from xml.sax.handler import feature_external_ges |
| 7 |
from xml.dom import pulldom |
| 8 |
|
| 9 |
from test.support import findfile |
| 10 |
|
| 11 |
|
| 12 |
tstfile = findfile("test.xml", subdir="xmltestdata") |
| 13 |
|
| 14 |
# A handy XML snippet, containing attributes, a namespace prefix, and a |
| 15 |
# self-closing tag: |
| 16 |
SMALL_SAMPLE = """<?xml version="1.0"?> |
| 17 |
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:xdc="http://www.xml.com/books"> |
| 18 |
<!-- A comment --> |
| 19 |
<title>Introduction to XSL</title> |
| 20 |
<hr/> |
| 21 |
<p><xdc:author xdc:attrib="prefixed attribute" attrib="other attrib">A. Namespace</xdc:author></p> |
| 22 |
</html>""" |
| 23 |
|
| 24 |
|
| 25 |
class PullDOMTestCase(unittest.TestCase): |
| 26 |
|
| 27 |
def test_parse(self): |
| 28 |
"""Minimal test of DOMEventStream.parse()""" |
| 29 |
|
| 30 |
# This just tests that parsing from a stream works. Actual parser |
| 31 |
# semantics are tested using parseString with a more focused XML |
| 32 |
# fragment. |
| 33 |
|
| 34 |
# Test with a filename: |
| 35 |
handler = pulldom.parse(tstfile) |
| 36 |
self.addCleanup(handler.stream.close) |
| 37 |
list(handler) |
| 38 |
|
| 39 |
# Test with a file object: |
| 40 |
with open(tstfile, "rb") as fin: |
| 41 |
list(pulldom.parse(fin)) |
| 42 |
|
| 43 |
def test_parse_semantics(self): |
| 44 |
"""Test DOMEventStream parsing semantics.""" |
| 45 |
|
| 46 |
items = pulldom.parseString(SMALL_SAMPLE) |
| 47 |
evt, node = next(items) |
| 48 |
# Just check the node is a Document: |
| 49 |
self.assertHasAttr(node, "createElement") |
| 50 |
self.assertEqual(pulldom.START_DOCUMENT, evt) |
| 51 |
evt, node = next(items) |
| 52 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 53 |
self.assertEqual("html", node.tagName) |
| 54 |
self.assertEqual(2, len(node.attributes)) |
| 55 |
self.assertEqual(node.attributes.getNamedItem("xmlns:xdc").value, |
| 56 |
"http://www.xml.com/books") |
| 57 |
evt, node = next(items) |
| 58 |
self.assertEqual(pulldom.CHARACTERS, evt) # Line break |
| 59 |
evt, node = next(items) |
| 60 |
# XXX - A comment should be reported here! |
| 61 |
# self.assertEqual(pulldom.COMMENT, evt) |
| 62 |
# Line break after swallowed comment: |
| 63 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 64 |
evt, node = next(items) |
| 65 |
self.assertEqual("title", node.tagName) |
| 66 |
title_node = node |
| 67 |
evt, node = next(items) |
| 68 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 69 |
self.assertEqual("Introduction to XSL", node.data) |
| 70 |
evt, node = next(items) |
| 71 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 72 |
self.assertEqual("title", node.tagName) |
| 73 |
self.assertTrue(title_node is node) |
| 74 |
evt, node = next(items) |
| 75 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 76 |
evt, node = next(items) |
| 77 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 78 |
self.assertEqual("hr", node.tagName) |
| 79 |
evt, node = next(items) |
| 80 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 81 |
self.assertEqual("hr", node.tagName) |
| 82 |
evt, node = next(items) |
| 83 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 84 |
evt, node = next(items) |
| 85 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 86 |
self.assertEqual("p", node.tagName) |
| 87 |
evt, node = next(items) |
| 88 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 89 |
self.assertEqual("xdc:author", node.tagName) |
| 90 |
evt, node = next(items) |
| 91 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 92 |
evt, node = next(items) |
| 93 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 94 |
self.assertEqual("xdc:author", node.tagName) |
| 95 |
evt, node = next(items) |
| 96 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 97 |
evt, node = next(items) |
| 98 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 99 |
evt, node = next(items) |
| 100 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 101 |
# XXX No END_DOCUMENT item is ever obtained: |
| 102 |
#evt, node = next(items) |
| 103 |
#self.assertEqual(pulldom.END_DOCUMENT, evt) |
| 104 |
|
| 105 |
def test_expandItem(self): |
| 106 |
"""Ensure expandItem works as expected.""" |
| 107 |
items = pulldom.parseString(SMALL_SAMPLE) |
| 108 |
# Loop through the nodes until we get to a "title" start tag: |
| 109 |
for evt, item in items: |
| 110 |
if evt == pulldom.START_ELEMENT and item.tagName == "title": |
| 111 |
items.expandNode(item) |
| 112 |
self.assertEqual(1, len(item.childNodes)) |
| 113 |
break |
| 114 |
else: |
| 115 |
self.fail("No \"title\" element detected in SMALL_SAMPLE!") |
| 116 |
# Loop until we get to the next start-element: |
| 117 |
for evt, node in items: |
| 118 |
if evt == pulldom.START_ELEMENT: |
| 119 |
break |
| 120 |
self.assertEqual("hr", node.tagName, |
| 121 |
"expandNode did not leave DOMEventStream in the correct state.") |
| 122 |
# Attempt to expand a standalone element: |
| 123 |
items.expandNode(node) |
| 124 |
self.assertEqual(next(items)[0], pulldom.CHARACTERS) |
| 125 |
evt, node = next(items) |
| 126 |
self.assertEqual(node.tagName, "p") |
| 127 |
items.expandNode(node) |
| 128 |
next(items) # Skip character data |
| 129 |
evt, node = next(items) |
| 130 |
self.assertEqual(node.tagName, "html") |
| 131 |
with self.assertRaises(StopIteration): |
| 132 |
next(items) |
| 133 |
items.clear() |
| 134 |
self.assertIsNone(items.parser) |
| 135 |
self.assertIsNone(items.stream) |
| 136 |
|
| 137 |
@unittest.expectedFailure |
| 138 |
def test_comment(self): |
| 139 |
"""PullDOM does not receive "comment" events.""" |
| 140 |
items = pulldom.parseString(SMALL_SAMPLE) |
| 141 |
for evt, _ in items: |
| 142 |
if evt == pulldom.COMMENT: |
| 143 |
break |
| 144 |
else: |
| 145 |
self.fail("No comment was encountered") |
| 146 |
|
| 147 |
@unittest.expectedFailure |
| 148 |
def test_end_document(self): |
| 149 |
"""PullDOM does not receive "end-document" events.""" |
| 150 |
items = pulldom.parseString(SMALL_SAMPLE) |
| 151 |
# Read all of the nodes up to and including </html>: |
| 152 |
for evt, node in items: |
| 153 |
if evt == pulldom.END_ELEMENT and node.tagName == "html": |
| 154 |
break |
| 155 |
try: |
| 156 |
# Assert that the next node is END_DOCUMENT: |
| 157 |
evt, node = next(items) |
| 158 |
self.assertEqual(pulldom.END_DOCUMENT, evt) |
| 159 |
except StopIteration: |
| 160 |
self.fail( |
| 161 |
"Ran out of events, but should have received END_DOCUMENT") |
| 162 |
|
| 163 |
def test_external_ges_default(self): |
| 164 |
parser = pulldom.parseString(SMALL_SAMPLE) |
| 165 |
saxparser = parser.parser |
| 166 |
ges = saxparser.getFeature(feature_external_ges) |
| 167 |
self.assertEqual(ges, False) |
| 168 |
|
| 169 |
|
| 170 |
class ThoroughTestCase(unittest.TestCase): |
| 171 |
"""Test the hard-to-reach parts of pulldom.""" |
| 172 |
|
| 173 |
def test_thorough_parse(self): |
| 174 |
"""Test some of the hard-to-reach parts of PullDOM.""" |
| 175 |
self._test_thorough(pulldom.parse(None, parser=SAXExerciser())) |
| 176 |
|
| 177 |
@unittest.expectedFailure |
| 178 |
def test_sax2dom_fail(self): |
| 179 |
"""SAX2DOM can"t handle a PI before the root element.""" |
| 180 |
pd = SAX2DOMTestHelper(None, SAXExerciser(), 12) |
| 181 |
self._test_thorough(pd) |
| 182 |
|
| 183 |
def test_thorough_sax2dom(self): |
| 184 |
"""Test some of the hard-to-reach parts of SAX2DOM.""" |
| 185 |
pd = SAX2DOMTestHelper(None, SAX2DOMExerciser(), 12) |
| 186 |
self._test_thorough(pd, False) |
| 187 |
|
| 188 |
def _test_thorough(self, pd, before_root=True): |
| 189 |
"""Test some of the hard-to-reach parts of the parser, using a mock |
| 190 |
parser.""" |
| 191 |
|
| 192 |
evt, node = next(pd) |
| 193 |
self.assertEqual(pulldom.START_DOCUMENT, evt) |
| 194 |
# Just check the node is a Document: |
| 195 |
self.assertHasAttr(node, "createElement") |
| 196 |
|
| 197 |
if before_root: |
| 198 |
evt, node = next(pd) |
| 199 |
self.assertEqual(pulldom.COMMENT, evt) |
| 200 |
self.assertEqual("a comment", node.data) |
| 201 |
evt, node = next(pd) |
| 202 |
self.assertEqual(pulldom.PROCESSING_INSTRUCTION, evt) |
| 203 |
self.assertEqual("target", node.target) |
| 204 |
self.assertEqual("data", node.data) |
| 205 |
|
| 206 |
evt, node = next(pd) |
| 207 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 208 |
self.assertEqual("html", node.tagName) |
| 209 |
|
| 210 |
evt, node = next(pd) |
| 211 |
self.assertEqual(pulldom.COMMENT, evt) |
| 212 |
self.assertEqual("a comment", node.data) |
| 213 |
evt, node = next(pd) |
| 214 |
self.assertEqual(pulldom.PROCESSING_INSTRUCTION, evt) |
| 215 |
self.assertEqual("target", node.target) |
| 216 |
self.assertEqual("data", node.data) |
| 217 |
|
| 218 |
evt, node = next(pd) |
| 219 |
self.assertEqual(pulldom.START_ELEMENT, evt) |
| 220 |
self.assertEqual("p", node.tagName) |
| 221 |
|
| 222 |
evt, node = next(pd) |
| 223 |
self.assertEqual(pulldom.CHARACTERS, evt) |
| 224 |
self.assertEqual("text", node.data) |
| 225 |
evt, node = next(pd) |
| 226 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 227 |
self.assertEqual("p", node.tagName) |
| 228 |
evt, node = next(pd) |
| 229 |
self.assertEqual(pulldom.END_ELEMENT, evt) |
| 230 |
self.assertEqual("html", node.tagName) |
| 231 |
evt, node = next(pd) |
| 232 |
self.assertEqual(pulldom.END_DOCUMENT, evt) |
| 233 |
|
| 234 |
|
| 235 |
class SAXExerciser(object): |
| 236 |
"""A fake sax parser that calls some of the harder-to-reach sax methods to |
| 237 |
ensure it emits the correct events""" |
| 238 |
|
| 239 |
def setContentHandler(self, handler): |
| 240 |
self._handler = handler |
| 241 |
|
| 242 |
def parse(self, _): |
| 243 |
h = self._handler |
| 244 |
h.startDocument() |
| 245 |
|
| 246 |
# The next two items ensure that items preceding the first |
| 247 |
# start_element are properly stored and emitted: |
| 248 |
h.comment("a comment") |
| 249 |
h.processingInstruction("target", "data") |
| 250 |
|
| 251 |
h.startElement("html", AttributesImpl({})) |
| 252 |
|
| 253 |
h.comment("a comment") |
| 254 |
h.processingInstruction("target", "data") |
| 255 |
|
| 256 |
h.startElement("p", AttributesImpl({"class": "paraclass"})) |
| 257 |
h.characters("text") |
| 258 |
h.endElement("p") |
| 259 |
h.endElement("html") |
| 260 |
h.endDocument() |
| 261 |
|
| 262 |
def stub(self, *args, **kwargs): |
| 263 |
"""Stub method. Does nothing.""" |
| 264 |
pass |
| 265 |
setProperty = stub |
| 266 |
setFeature = stub |
| 267 |
|
| 268 |
|
| 269 |
class SAX2DOMExerciser(SAXExerciser): |
| 270 |
"""The same as SAXExerciser, but without the processing instruction and |
| 271 |
comment before the root element, because S2D can"t handle it""" |
| 272 |
|
| 273 |
def parse(self, _): |
| 274 |
h = self._handler |
| 275 |
h.startDocument() |
| 276 |
h.startElement("html", AttributesImpl({})) |
| 277 |
h.comment("a comment") |
| 278 |
h.processingInstruction("target", "data") |
| 279 |
h.startElement("p", AttributesImpl({"class": "paraclass"})) |
| 280 |
h.characters("text") |
| 281 |
h.endElement("p") |
| 282 |
h.endElement("html") |
| 283 |
h.endDocument() |
| 284 |
|
| 285 |
|
| 286 |
class SAX2DOMTestHelper(pulldom.DOMEventStream): |
| 287 |
"""Allows us to drive SAX2DOM from a DOMEventStream.""" |
| 288 |
|
| 289 |
def reset(self): |
| 290 |
self.pulldom = pulldom.SAX2DOM() |
| 291 |
# This content handler relies on namespace support |
| 292 |
self.parser.setFeature(xml.sax.handler.feature_namespaces, 1) |
| 293 |
self.parser.setContentHandler(self.pulldom) |
| 294 |
|
| 295 |
|
| 296 |
class SAX2DOMTestCase(unittest.TestCase): |
| 297 |
|
| 298 |
def confirm(self, test, testname="Test"): |
| 299 |
self.assertTrue(test, testname) |
| 300 |
|
| 301 |
def test_basic(self): |
| 302 |
"""Ensure SAX2DOM can parse from a stream.""" |
| 303 |
with io.StringIO(SMALL_SAMPLE) as fin: |
| 304 |
sd = SAX2DOMTestHelper(fin, xml.sax.make_parser(), |
| 305 |
len(SMALL_SAMPLE)) |
| 306 |
for evt, node in sd: |
| 307 |
if evt == pulldom.START_ELEMENT and node.tagName == "html": |
| 308 |
break |
| 309 |
# Because the buffer is the same length as the XML, all the |
| 310 |
# nodes should have been parsed and added: |
| 311 |
self.assertGreater(len(node.childNodes), 0) |
| 312 |
|
| 313 |
def testSAX2DOM(self): |
| 314 |
"""Ensure SAX2DOM expands nodes as expected.""" |
| 315 |
sax2dom = pulldom.SAX2DOM() |
| 316 |
sax2dom.startDocument() |
| 317 |
sax2dom.startElement("doc", {}) |
| 318 |
sax2dom.characters("text") |
| 319 |
sax2dom.startElement("subelm", {}) |
| 320 |
sax2dom.characters("text") |
| 321 |
sax2dom.endElement("subelm") |
| 322 |
sax2dom.characters("text") |
| 323 |
sax2dom.endElement("doc") |
| 324 |
sax2dom.endDocument() |
| 325 |
|
| 326 |
doc = sax2dom.document |
| 327 |
root = doc.documentElement |
| 328 |
(text1, elm1, text2) = root.childNodes |
| 329 |
text3 = elm1.childNodes[0] |
| 330 |
|
| 331 |
self.assertIsNone(text1.previousSibling) |
| 332 |
self.assertIs(text1.nextSibling, elm1) |
| 333 |
self.assertIs(elm1.previousSibling, text1) |
| 334 |
self.assertIs(elm1.nextSibling, text2) |
| 335 |
self.assertIs(text2.previousSibling, elm1) |
| 336 |
self.assertIsNone(text2.nextSibling) |
| 337 |
self.assertIsNone(text3.previousSibling) |
| 338 |
self.assertIsNone(text3.nextSibling) |
| 339 |
|
| 340 |
self.assertIs(root.parentNode, doc) |
| 341 |
self.assertIs(text1.parentNode, root) |
| 342 |
self.assertIs(elm1.parentNode, root) |
| 343 |
self.assertIs(text2.parentNode, root) |
| 344 |
self.assertIs(text3.parentNode, elm1) |
| 345 |
doc.unlink() |
| 346 |
|
| 347 |
|
| 348 |
if __name__ == "__main__": |
| 349 |
unittest.main() |