add an easy way to count parse errors in parser.py; use that in parse.py; strip trailing spaces in parse.py

annevk · annevk · commit 930e14538194 · 2006-12-30T19:55:58.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%40382
diff --git a/parse.py b/parse.py
@@ -51,18 +51,19 @@ def parse():
     else:
         document = p.parse(f)
         print convertTreeDump(document.printTree())
+        print "\nParse errors:", len(p.errors)
 
 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", 
+                      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__":
diff --git a/src/parser.py b/src/parser.py
@@ -35,6 +35,7 @@ class HTMLParser(object):
     def __init__(self, strict = False, tree=DOMlite.TreeBuilder):
         # Raise an exception on the first error encountered
         self.strict = strict
+        self.errors = []
 
         self.tree = tree()
 
@@ -87,7 +88,7 @@ def parse(self, stream, innerHTML=False):
             elif type == "EndTag":
                 method(token["name"])
             elif type == "ParseError":
-                self.parseError()
+                self.parseError(token["data"])
             else:
                 self.atheistParseError()
 
@@ -96,7 +97,8 @@ def parse(self, stream, innerHTML=False):
 
         return self.tree.getDocument()
 
-    def parseError(self):
+    def parseError(self, data=None):
+        self.errors.append(data)
         if self.strict:
             raise ParseError
 
