Cheap hack so that elementtree support can be tested in the short term

jgraham · jgraham · commit 3ccfea2af0ee · 2006-12-31T00:21:14.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%40393
diff --git a/tests/test_parser_etree.py b/tests/test_parser_etree.py
@@ -0,0 +1,103 @@
+import sys
+import os
+import glob
+import StringIO
+import unittest
+import new
+
+# XXX Allow us to import the sibling module
+os.chdir(os.path.split(os.path.abspath(__file__))[0])
+sys.path.insert(0, os.path.abspath(os.path.join(os.pardir, "src")))
+
+import parser
+import treebuilders
+
+def parseTestcase(testString):
+    testString = testString.split("\n")
+    try:
+        if testString[0] != "#data":
+            print testString
+        assert testString[0] == "#data"
+    except:
+        raise
+    input = []
+    output = []
+    errors = []
+    currentList = input
+    for line in testString:
+        if line and not (line.startswith("#errors") or
+          line.startswith("#document") or line.startswith("#data")):
+            if currentList is output:
+                if line.startswith("|"):
+                    currentList.append(line[2:])
+                else:
+                    currentList.append(line)
+            else:
+                currentList.append(line)
+        elif line == "#errors":
+            currentList = errors
+        elif line == "#document":
+            currentList = output
+    return "\n".join(input), "\n".join(output), errors
+
+def convertTreeDump(treedump):
+    """convert the output of str(document) to the format used in the testcases"""
+    treedump = treedump.split("\n")[1:]
+    rv = []
+    for line in treedump:
+        if line.startswith("|"):
+            rv.append(line[3:])
+        else:
+            rv.append(line)
+    return "\n".join(rv)
+
+class TestCase(unittest.TestCase):
+    def runParserTest(self, input, output, errors):
+        #XXX - move this out into the setup function
+        #concatenate all consecutive character tokens into a single token
+        from treebuilders import etree
+        treebuilder = etree.TreeBuilder
+
+        p = parser.HTMLParser(tree = treebuilder)
+        document = p.parse(StringIO.StringIO(input))
+        errorMsg = "\n".join(["\n\nExpected:", output, "\nRecieved:",
+          convertTreeDump(p.tree.testSerializer(document))])
+        self.assertEquals(output, 
+                          convertTreeDump(p.tree.testSerializer(document)),
+                          errorMsg)
+        #errorMsg2 = "\n".join(["\n\nInput errors:\n" + "\n".join(errors),
+        #  "Actual errors:\n" + "\n".join(p.errors)])
+        #self.assertEquals(len(p.errors), len(errors), errorMsg2)
+
+def test_parser():
+    for filename in glob.glob('tree-construction/*.dat'):
+        f = open(filename)
+        tests = f.read().split("#data\n")
+        for test in tests:
+            if test == "":
+                continue
+            test = "#data\n" + test
+            input, output, errors = parseTestcase(test)
+            yield TestCase.runParserTest, input, output, errors
+
+def buildTestSuite():
+    tests = 0
+    for func, input, output, errors in test_parser():
+        tests += 1
+        testName = 'test%d' % tests
+        testFunc = lambda self, method=func, input=input, output=output, \
+            errors=errors: method(self, input, output, errors)
+        testFunc.__doc__ = 'Parser %s: %s' % (testName, input)
+        instanceMethod = new.instancemethod(testFunc, None, TestCase)
+        setattr(TestCase, testName, instanceMethod)
+    return unittest.TestLoader().loadTestsFromTestCase(TestCase)
+
+def main():
+    buildTestSuite()
+    unittest.main()
+
+if __name__ == "__main__":
+    # XXX Allow us to import the sibling module
+    os.chdir(os.path.split(os.path.abspath(__file__))[0])
+    sys.path.insert(0, os.path.abspath(os.path.join(os.pardir, "src")))
+    main()
