quote base reference URL, with tests

git-svn-id: https://linkchecker.svn.sourceforge.net/svnroot/linkchecker/trunk/linkchecker@3402 e7d03fd6-7b0d-0410-9947-9c21f3af8025

ChangeLog

@@ -23,6 +23,10 @@
     Type: bugfix
     Changed: linkchecker
 
+  * Norm the base reference URL.
+    Type: bugfix
+    Changed: linkcheck/checker/urlbase.py
+
 4.1 "Tsotsi" (released 29.5.2006)
 
   * Wait for spawned threads to finish before shutdown. Gets rid

linkcheck/checker/tests/data/http.html

@@ -29,6 +29,9 @@ Just some HTTP links
 <img style="@font-face {src:url(http.html)};background-image:url(http.html)"
  title="CSS urls">
 
+<!-- object with codebase -->
+<object classid="clsid:12345" codebase="http://imadoofus.org/foo/ #a=1,2,3">
+
 <!-- <a href=http://nocheckin> no check because of comment -->
 <a href=http://localhost:8001/">no beginning quote</a>
 <a href="http://localhost:8001/>no ending quote</a>

linkcheck/checker/tests/data/http.html.result

@@ -122,6 +122,13 @@ name html entities
 warning Base URL is not properly normed. Normed URL is http://localhost:8001/?quoted=%%FC.
 valid
 
+url clsid:12345
+cache key clsid:12345
+real url clsid:12345
+baseurl http://imadoofus.org/foo/%%20#a%%3D1%%2C2%%2C3
+warning Clsid URL ignored.
+valid
+
 url http://localhost:8001/ (cached)
 cache key http://localhost:8001/
 real url http://localhost:8001/

linkcheck/checker/urlbase.py

@@ -662,6 +662,7 @@ class UrlBase (object):
                 base_ref = codebase
             else:
                 base_ref = h.base_ref
+            base_ref = linkcheck.url.url_norm(base_ref)[0]
             url_data = linkcheck.checker.get_url_from(url,
                   self.recursion_level+1, self.aggregate, parent_url=self.url,
                   base_ref=base_ref, line=line, column=column, name=name,