Changes to parser.py to make the command line interface more sane, type parser.py -h for usage. Also rolls in a fix to the import in treebuilders.base so that both parser and unittests should work and a few miscellaneous changes there that should be in a different patch

jgraham · jgraham · commit 8c8eb7e24ac3 · 2006-12-29T09:29:36.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%40370
diff --git a/parse.py b/parse.py
@@ -1,13 +1,11 @@
+"""usage: %prog [options] filename
+
+Parse a document to a DOMlite tree, with optional profiling
 """
- Usage:
-   python parse.py tests/sites/web-apps.htm > outputfile
-     To parse the file web-apps.htm and get a tree.
-   
-   python parse.py tests/sites/web-apps.htm x > outputfile
-     To parse the file web-apps.htm and get a profile.
-"""
+
 import sys
 import os
+from optparse import OptionParser
 
 from src import parser
 
@@ -23,30 +21,48 @@ def convertTreeDump(treedump):
             rv.append(line)
     return "\n".join(rv)
 
-if __name__ == "__main__":
+def parse():
+    optParser = getOptParser()
+    opts,args = optParser.parse_args()
+
     p = parser.HTMLParser()
-    if len(sys.argv) > 1:
-        x = sys.argv[1]
-        if len(sys.argv) > 2:
-            import hotshot
-            import hotshot.stats
-            prof = hotshot.Profile('stats.prof')
-            prof.runcall(p.parse, x, False)
-            prof.close()
-            stats = hotshot.stats.load('stats.prof')
-            stats.strip_dirs()
-            stats.sort_stats('time')
-            stats.print_stats()
-        else:
-            from time import time
-            t = time()
-            document = p.parse(x)
-            t = time() - t
-            t2 = time()
-            print convertTreeDump(document.printTree())
-            t2 = time() - t2
-            print "\n\nDuration:", t, "\nTree dump duration:", t2
+    f = open(args[0])
+    if opts.profile:
+        import hotshot
+        import hotshot.stats
+        prof = hotshot.Profile('stats.prof')
+        prof.runcall(p.parse, f, False)
+        prof.close()
+        #XXX - We should use a temp file here
+        stats = hotshot.stats.load('stats.prof')
+        stats.strip_dirs()
+        stats.sort_stats('time')
+        stats.print_stats()
+    elif opts.time:
+        import time
+        t0 = time.time()
+        document = p.parse(f)
+        t1 = time.time()
+        print convertTreeDump(document.printTree())
+        t2 = time.time()
+        print "\n\nRun took: %fs (plus %fs to print the output)"%(t1-t0, t2-t1)
     else:
-        print """Pass one argument to parse the document and two to get an
-              indication on what's going on.
-              """
+        document = p.parse(f)
+        print convertTreeDump(document.printTree())
+
+def getOptParser():
+    parser = OptionParser(usage=__doc__)
+    
+    parser.add_option("-p", "--profile", action="store_true", default=False,
+                      dest="profile", help="Use the hotdhot profiler to "
+                      "produce a detailed log of the run")
+    
+    parser.add_option("-t", "--time",
+                      action="store_true", default=False, dest="time", 
+                      help="Time the run using time.time (may not be accurate on all platforms, especially for short runs)")
+    
+    return parser
+
+if __name__ == "__main__":
+    print os.path.abspath(os.curdir)
+    parse()
diff --git a/src/parser.py b/src/parser.py
@@ -59,10 +59,6 @@ def parse(self, stream, innerHTML=False):
         # assertations
         self.innerHTML = innerHTML
 
-        # Flag indicationg special insertion mode from elements misnested inside
-        # a table
-        self.insertFromTable = False
-
         self.tokenizer = tokenizer.HTMLTokenizer(stream)
 
         # XXX This is temporary for the moment so there isn't any other
@@ -84,7 +80,7 @@ def parse(self, stream, innerHTML=False):
         # When the loop finishes it's EOF
         self.phase.processEOF()
 
-        return self.tree.document
+        return self.tree.getDocument()
 
     def parseError(self):
         if self.strict:
diff --git a/src/treebuilders/base.py b/src/treebuilders/base.py
@@ -1,10 +1,16 @@
 import sys
 import os
 
-# XXX someone please fix this up! And make sure it doesn't break Windows.
-from src.constants import contentModelFlags, spaceCharacters
-from src.constants import scopingElements, formattingElements, specialElements
-from src.constants import headingElements, tableInsertModeElements
+#Insert the parent directory of the current file into the path
+_curDir = os.path.abspath(os.curdir)
+os.chdir(os.path.dirname(__file__))
+sys.path.insert(0, os.path.abspath(os.pardir))
+os.chdir(_curDir)
+del _curDir
+
+from constants import contentModelFlags, spaceCharacters
+from constants import scopingElements, formattingElements, specialElements
+from constants import headingElements, tableInsertModeElements
 
 # The scope markers are inserted when entering buttons, object elements,
 # marquees, table cells, and table captions, and are used to prevent formatting
@@ -224,3 +230,7 @@ def generateImpliedEndTags(self, exclude=None):
             # we should keep it in.
             # self.processEndTag(name)
             self.generateImpliedEndTags(exclude)
+
+    def getDocument(self):
+        "Return the final tree"
+        return self.document
diff --git a/tests/test_tokenizer.py b/tests/test_tokenizer.py
@@ -51,7 +51,7 @@ def processParseError(self, token):
 
     def processAtheistParseError(self, token):
         """This error is not an error"""
-        self.outputTokens.append(u"AtheistParseError")
+        pass
 
 def concatenateCharacterTokens(tokens):
     outputTokens = []
@@ -118,8 +118,4 @@ def main():
     unittest.main()
 
 if __name__ == "__main__":
-    #Allow us to import the parent module
-    os.chdir(os.path.split(os.path.abspath(__file__))[0])
-    sys.path.insert(0, os.path.abspath(os.pardir))
-
     main()
