From jjlee at codespeak.net  Fri Jan  5 23:12:36 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Fri,  5 Jan 2007 23:12:36 +0100 (CET)
Subject: [wwwsearch-commits] r36173 - wwwsearch/ClientForm/trunk
Message-ID: <20070105221236.EEFF810079@code0.codespeak.net>

Author: jjlee
Date: Fri Jan  5 23:12:34 2007
New Revision: 36173

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
Log:
For backwards compatibility, make ParseError derive from the exceptions that used to be raised during parsing

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Fri Jan  5 23:12:34 2007
@@ -103,6 +103,19 @@
        htmlentitydefs, re, random
 from cStringIO import StringIO
 
+import sgmllib
+# monkeypatch to fix http://www.python.org/sf/803422 :-(
+sgmllib.charref = re.compile("&#(x?[0-9a-fA-F]+)[^0-9a-fA-F]")
+
+# HTMLParser.HTMLParser is recent, so live without it if it's not available
+# (also, sgmllib.SGMLParser is much more tolerant of bad HTML)
+try:
+    import HTMLParser
+except ImportError:
+    HAVE_MODULE_HTMLPARSER = False
+else:
+    HAVE_MODULE_HTMLPARSER = True
+
 try:
     import warnings
 except ImportError:
@@ -426,8 +439,16 @@
 
 class ItemCountError(ValueError): pass
 
-
-class ParseError(Exception): pass
+# for backwards compatibility, ParseError derives from exceptions that were
+# raised by versions of ClientForm <= 0.2.5
+if HAVE_MODULE_HTMLPARSER:
+    class ParseError(sgmllib.SGMLParser,
+                     HTMLParser.HTMLParser,
+                     ):
+        pass
+else:
+    class ParseError(sgmllib.SGMLParser):
+        pass
 
 
 class _AbstractFormParser:
@@ -748,11 +769,7 @@
     def unknown_charref(self, ref): self.handle_data("&#%s;" % ref)
 
 
-# HTMLParser.HTMLParser is recent, so live without it if it's not available
-# (also, htmllib.HTMLParser is much more tolerant of bad HTML)
-try:
-    import HTMLParser
-except ImportError:
+if not HAVE_MODULE_HTMLPARSER:
     class XHTMLCompatibleFormParser:
         def __init__(self, entitydefs=None, encoding=DEFAULT_ENCODING):
             raise ValueError("HTMLParser could not be imported")
@@ -807,9 +824,6 @@
         def unescape_attrs_if_required(self, attrs):
             return attrs  # ditto
 
-import sgmllib
-# monkeypatch to fix http://www.python.org/sf/803422 :-(
-sgmllib.charref = re.compile("&#(x?[0-9a-fA-F]+)[^0-9a-fA-F]")
 
 class _AbstractSgmllibParser(_AbstractFormParser):
 

From jjlee at codespeak.net  Fri Jan  5 23:51:53 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Fri,  5 Jan 2007 23:51:53 +0100 (CET)
Subject: [wwwsearch-commits] r36174 - wwwsearch/ClientForm/trunk
Message-ID: <20070105225153.194C610079@code0.codespeak.net>

Author: jjlee
Date: Fri Jan  5 23:51:52 2007
New Revision: 36174

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
   wwwsearch/ClientForm/trunk/test.py
Log:
Derive ParseError from the right class

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Fri Jan  5 23:51:52 2007
@@ -442,8 +442,8 @@
 # for backwards compatibility, ParseError derives from exceptions that were
 # raised by versions of ClientForm <= 0.2.5
 if HAVE_MODULE_HTMLPARSER:
-    class ParseError(sgmllib.SGMLParser,
-                     HTMLParser.HTMLParser,
+    class ParseError(sgmllib.SGMLParseError,
+                     HTMLParser.HTMLParseError,
                      ):
         pass
 else:

Modified: wwwsearch/ClientForm/trunk/test.py
==============================================================================
--- wwwsearch/ClientForm/trunk/test.py	(original)
+++ wwwsearch/ClientForm/trunk/test.py	Fri Jan  5 23:51:52 2007
@@ -226,12 +226,14 @@
     def test_failing_parse(self):
         # XXX couldn't provoke an error from BeautifulSoup (!), so this has not
         # been tested with RobuststFormParser
+        import sgmllib
         f = StringIO("<!!!!>")
         base_uri = "http://localhost/"
         self.assertRaises(
             ClientForm.ParseError,
             ClientForm.ParseFile, f, base_uri, backwards_compat=False,
             )
+        self.assert_(issubclass(ClientForm.ParseError, sgmllib.SGMLParseError))
 
     def test_unknown_control(self):
         f = StringIO(

From jjlee at codespeak.net  Sat Jan  6 00:05:49 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat,  6 Jan 2007 00:05:49 +0100 (CET)
Subject: [wwwsearch-commits] r36175 - in wwwsearch/mechanize/trunk:
	mechanize test
Message-ID: <20070105230549.5DE6010079@code0.codespeak.net>

Author: jjlee
Date: Sat Jan  6 00:05:30 2007
New Revision: 36175

Modified:
   wwwsearch/mechanize/trunk/mechanize/__init__.py
   wwwsearch/mechanize/trunk/mechanize/_html.py
   wwwsearch/mechanize/trunk/test/test_forms.doctest
   wwwsearch/mechanize/trunk/test/test_html.doctest
Log:
Add mechanize.ParseError class, document it as part of the mechanize.Factory interface, and raise it from all Factory implementations

Modified: wwwsearch/mechanize/trunk/mechanize/__init__.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/__init__.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/__init__.py	Sat Jan  6 00:05:30 2007
@@ -50,6 +50,7 @@
     'MozillaCookieJar',
     'OpenerDirector',
     'OpenerFactory',
+    'ParseError',
     'ProxyBasicAuthHandler',
     'ProxyDigestAuthHandler',
     'ProxyHandler',
@@ -89,6 +90,7 @@
 # configurable URL-opener interface
 from _useragent import UserAgentBase, UserAgent
 from _html import \
+     ParseError, \
      Link, \
      Factory, DefaultFactory, RobustFactory, \
      FormsFactory, LinksFactory, TitleFactory, \

Modified: wwwsearch/mechanize/trunk/mechanize/_html.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_html.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_html.py	Sat Jan  6 00:05:30 2007
@@ -9,6 +9,7 @@
 """
 
 import re, copy, htmlentitydefs
+import sgmllib, HTMLParser, ClientForm
 
 import _request
 from _headersutil import split_header_words, is_html as _is_html
@@ -17,6 +18,10 @@
 DEFAULT_ENCODING = "latin-1"
 
 
+# the base classe is purely for backwards compatibility
+class ParseError(ClientForm.ParseError): pass
+
+
 class CachingGeneratorFunction(object):
     """Caching wrapper around a no-arguments iterable."""
 
@@ -131,37 +136,41 @@
         base_url = self._base_url
         p = self.link_parser_class(response, encoding=encoding)
 
-        for token in p.tags(*(self.urltags.keys()+["base"])):
-            if token.type == "endtag":
-                continue
-            if token.data == "base":
-                base_href = dict(token.attrs).get("href")
-                if base_href is not None:
-                    base_url = base_href
-                continue
-            attrs = dict(token.attrs)
-            tag = token.data
-            name = attrs.get("name")
-            text = None
-            # XXX use attr_encoding for ref'd doc if that doc does not provide
-            #  one by other means
-            #attr_encoding = attrs.get("charset")
-            url = attrs.get(self.urltags[tag])  # XXX is "" a valid URL?
-            if not url:
-                # Probably an <A NAME="blah"> link or <AREA NOHREF...>.
-                # For our purposes a link is something with a URL, so ignore
-                # this.
-                continue
-
-            url = _rfc3986.clean_url(url, encoding)
-            if tag == "a":
-                if token.type != "startendtag":
-                    # hmm, this'd break if end tag is missing
-                    text = p.get_compressed_text(("endtag", tag))
-                # but this doesn't work for eg. <a href="blah"><b>Andy</b></a>
-                #text = p.get_compressed_text()
+        try:
+            for token in p.tags(*(self.urltags.keys()+["base"])):
+                if token.type == "endtag":
+                    continue
+                if token.data == "base":
+                    base_href = dict(token.attrs).get("href")
+                    if base_href is not None:
+                        base_url = base_href
+                    continue
+                attrs = dict(token.attrs)
+                tag = token.data
+                name = attrs.get("name")
+                text = None
+                # XXX use attr_encoding for ref'd doc if that doc does not
+                #  provide one by other means
+                #attr_encoding = attrs.get("charset")
+                url = attrs.get(self.urltags[tag])  # XXX is "" a valid URL?
+                if not url:
+                    # Probably an <A NAME="blah"> link or <AREA NOHREF...>.
+                    # For our purposes a link is something with a URL, so
+                    # ignore this.
+                    continue
 
-            yield Link(base_url, url, text, tag, token.attrs)
+                url = _rfc3986.clean_url(url, encoding)
+                if tag == "a":
+                    if token.type != "startendtag":
+                        # hmm, this'd break if end tag is missing
+                        text = p.get_compressed_text(("endtag", tag))
+                    # but this doesn't work for eg.
+                    # <a href="blah"><b>Andy</b></a>
+                    #text = p.get_compressed_text()
+
+                yield Link(base_url, url, text, tag, token.attrs)
+        except sgmllib.SGMLParseError, exc:
+            raise ParseError(exc)
 
 class FormsFactory:
 
@@ -202,16 +211,19 @@
     def forms(self):
         import ClientForm
         encoding = self.encoding
-        forms = ClientForm.ParseResponseEx(
-            self._response,
-            select_default=self.select_default,
-            form_parser_class=self.form_parser_class,
-            request_class=self.request_class,
-            encoding=encoding,
-            _urljoin=_rfc3986.urljoin,
-            _urlparse=_rfc3986.urlsplit,
-            _urlunparse=_rfc3986.urlunsplit,
-            )
+        try:
+            forms = ClientForm.ParseResponseEx(
+                self._response,
+                select_default=self.select_default,
+                form_parser_class=self.form_parser_class,
+                request_class=self.request_class,
+                encoding=encoding,
+                _urljoin=_rfc3986.urljoin,
+                _urlparse=_rfc3986.urlsplit,
+                _urlunparse=_rfc3986.urlunsplit,
+                )
+        except ClientForm.ParseError, exc:
+            raise ParseError(exc)
         self.global_form = forms[0]
         return forms[1:]
 
@@ -228,11 +240,14 @@
         p = _pullparser.TolerantPullParser(
             self._response, encoding=self._encoding)
         try:
-            p.get_tag("title")
-        except _pullparser.NoMoreTokensError:
-            return None
-        else:
-            return p.get_text()
+            try:
+                p.get_tag("title")
+            except _pullparser.NoMoreTokensError:
+                return None
+            else:
+                return p.get_text()
+        except sgmllib.SGMLParseError, exc:
+            raise ParseError(exc)
 
 
 def unescape(data, entities, encoding):
@@ -420,6 +435,8 @@
 
     Public attributes:
 
+    Note that accessing these attributes may raise ParseError.
+
     encoding: string specifying the encoding of response if it contains a text
      document (this value is left unspecified for documents that do not have
      an encoding, e.g. an image file)
@@ -505,7 +522,10 @@
             return self.global_form
 
     def forms(self):
-        """Return iterable over ClientForm.HTMLForm-like objects."""
+        """Return iterable over ClientForm.HTMLForm-like objects.
+
+        Raises mechanize.ParseError on failure.
+        """
         # this implementation sets .global_form as a side-effect, for benefit
         # of __getattr__ impl
         if self._forms_genf is None:
@@ -520,7 +540,10 @@
         return self._forms_genf()
 
     def links(self):
-        """Return iterable over mechanize.Link-like objects."""
+        """Return iterable over mechanize.Link-like objects.
+
+        Raises mechanize.ParseError on failure.
+        """
         if self._links_genf is None:
             try:
                 self._links_genf = CachingGeneratorFunction(

Modified: wwwsearch/mechanize/trunk/test/test_forms.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_forms.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_forms.doctest	Sat Jan  6 00:05:30 2007
@@ -41,7 +41,7 @@
 already been .read().  Fixed by calling Factory.set_response() on
 error.
 
->>> import mechanize, sgmllib
+>>> import mechanize
 >>> br = mechanize.Browser()
 >>> r = mechanize._response.test_html_response("""\
 ... <form>
@@ -52,8 +52,8 @@
 >>> br.set_response(r)
 >>> try:
 ...     br.select_form(nr=0)
-... except sgmllib.SGMLParseError:
+... except mechanize.ParseError:
 ...     pass
 >>> br.select_form(nr=0)  # doctest: +IGNORE_EXCEPTION_DETAIL
 Traceback (most recent call last):
-SGMLParseError: expected name token
+ParseError: expected name token

Modified: wwwsearch/mechanize/trunk/test/test_html.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_html.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_html.doctest	Sat Jan  6 00:05:30 2007
@@ -1,3 +1,122 @@
+>>> import mechanize
+>>> from mechanize._response import test_html_response
+>>> from mechanize._html import LinksFactory, FormsFactory, TitleFactory, \
+... MechanizeBs, \
+... RobustLinksFactory,  RobustFormsFactory, RobustTitleFactory
+
+mechanize.ParseError should be raised on parsing erroneous HTML.
+
+For backwards compatibility, mechanize.ParseError derives from
+exception classes that mechanize used to raise, prior to version
+0.1.6.
+
+>>> import sgmllib
+>>> import HTMLParser
+>>> import ClientForm
+>>> issubclass(mechanize.ParseError, sgmllib.SGMLParseError)
+True
+>>> issubclass(mechanize.ParseError, HTMLParser.HTMLParseError)
+True
+>>> issubclass(mechanize.ParseError, ClientForm.ParseError)
+True
+
+>>> def create_response(error=True):
+...     extra = ""
+...     if error:
+...         extra = "<!!!>"
+...     html = """\
+... <html>
+... <head>
+...     <title>Title</title>
+...     %s
+... </head>
+... <body>
+...     <p>Hello world
+... </body>
+... </html>
+... """ % extra
+...     return test_html_response(html)
+
+>>> f = LinksFactory()
+>>> f.set_response(create_response(), "http://example.com", "latin-1")
+>>> list(f.links())  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+>>> f = FormsFactory()
+>>> f.set_response(create_response(), "latin-1")
+>>> list(f.forms())  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+>>> f = TitleFactory()
+>>> f.set_response(create_response(), "latin-1")
+>>> f.title()  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+
+
+Accessing attributes on Factory may also raise ParseError
+
+>>> def factory_getattr(attr_name):
+...    fact = mechanize.DefaultFactory()
+...    fact.set_response(create_response())
+...    getattr(fact, attr_name)
+>>> factory_getattr("title")  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+>>> factory_getattr("global_form")  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+
+
+BeautifulSoup ParseErrors:
+
+XXX If I could come up with examples that break links and forms
+parsing, I'd uncomment these!
+
+>>> def create_soup(html):
+...     r = test_html_response(html)
+...     return MechanizeBs("latin-1", r.read())
+
+#>>> f = RobustLinksFactory()
+#>>> html = """\
+#... <a href="a">
+#... <frame src="b">
+#... <a href="c">
+#... <iframe src="d">
+#... </a>
+#... </area>
+#... </frame>
+#... """
+#>>> f.set_soup(create_soup(html), "http://example.com", "latin-1")
+#>>> list(f.links())  # doctest: +IGNORE_EXCEPTION_DETAIL
+#Traceback (most recent call last):
+#ParseError:
+
+>>> html = """\
+... <table>
+... <tr><td>
+... <input name='broken'>
+... </td>
+... </form>
+... </tr>
+... </form>
+... """
+>>> f = RobustFormsFactory()
+>>> f.set_response(create_response(), "latin-1")
+>>> list(f.forms())  # doctest: +IGNORE_EXCEPTION_DETAIL
+Traceback (most recent call last):
+ParseError:
+
+#>>> f = RobustTitleFactory()
+#>>> f.set_soup(create_soup(""), "latin-1")
+#>>> f.title()  # doctest: +IGNORE_EXCEPTION_DETAIL
+#Traceback (most recent call last):
+#ParseError:
+
+
+
+Utility class for caching forms etc.
+
 >>> from mechanize._html import CachingGeneratorFunction
 
 >>> i = [1]

From jjlee at codespeak.net  Sun Jan  7 17:13:04 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 17:13:04 +0100 (CET)
Subject: [wwwsearch-commits] r36193 - wwwsearch/mechanize/trunk/test
Message-ID: <20070107161304.7D2981005A@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 17:13:02 2007
New Revision: 36193

Modified:
   wwwsearch/mechanize/trunk/test/test_browser.py
Log:
Fix spurious test_reload_read_incomplete test failure

Modified: wwwsearch/mechanize/trunk/test/test_browser.py
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_browser.py	(original)
+++ wwwsearch/mechanize/trunk/test/test_browser.py	Sun Jan  7 17:13:02 2007
@@ -256,12 +256,14 @@
         br = Browser()
         data = "<html><head><title></title></head><body>%s</body></html>"
         data = data % ("The quick brown fox jumps over the lazy dog."*100)
-        r = test_response(data, [("content-type", "text/html")])
-        br.add_handler(make_mock_handler()([("http_open", r)]))
+        class Handler(mechanize.BaseHandler):
+            def http_open(self, requst):
+                return test_response(data, [("content-type", "text/html")])
+        br.add_handler(Handler())
 
         # .reload() on .back() if the whole response hasn't already been read
         # (.read_incomplete is True)
-        r = br.open(r.geturl())
+        r = br.open("http://example.com")
         r.read(10)
         br.open('http://www.example.com/blah')
         self.failIf(br.reloaded)

From jjlee at codespeak.net  Sun Jan  7 17:15:36 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 17:15:36 +0100 (CET)
Subject: [wwwsearch-commits] r36194 - wwwsearch/mechanize/trunk/test
Message-ID: <20070107161536.E9CAD10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 17:15:35 2007
New Revision: 36194

Modified:
   wwwsearch/mechanize/trunk/test/test_response.doctest
Log:
Add test for read_complete bug fixed in rev 36082

Modified: wwwsearch/mechanize/trunk/test/test_response.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_response.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_response.doctest	Sun Jan  7 17:15:35 2007
@@ -3,6 +3,7 @@
 .reload() the response if not.
 
 I've noted here the various cases where .read_complete may be set.
+
 >>> text = "To err is human, to moo, bovine.\n"*10
 >>> def get_wrapper():
 ...     import cStringIO
@@ -94,3 +95,18 @@
 True
 >>> wr2.read_complete
 True
+
+
+Fix from -r36082: .read() after .close() used to break
+.read_complete state
+
+>>> from mechanize._response import test_response
+>>> r = test_response(text)
+>>> junk = r.read(64)
+>>> r.close()
+>>> r.read_complete
+False
+>>> r.read()
+''
+>>> r.read_complete
+False

From jjlee at codespeak.net  Sun Jan  7 17:39:25 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 17:39:25 +0100 (CET)
Subject: [wwwsearch-commits] r36196 - wwwsearch/mechanize/trunk/test
Message-ID: <20070107163925.18DAC1006E@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 17:39:22 2007
New Revision: 36196

Modified:
   wwwsearch/mechanize/trunk/test/test_browser.doctest
Log:
Add test for first .back() bug fixed in rev 36082 (returned old response, not the reloaded one)

Modified: wwwsearch/mechanize/trunk/test/test_browser.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_browser.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_browser.doctest	Sun Jan  7 17:39:22 2007
@@ -49,6 +49,24 @@
 True
 
 
+.back() may reload if the complete response was not read.  If so, it
+should return the new response, not the old one
+
+>>> class ReloadCheckBrowser(TestHttpBrowser):
+...     reloaded = False
+...     def reload(self):
+...         self.reloaded = True
+...         return TestHttpBrowser.reload(self)
+>>> br = ReloadCheckBrowser()
+>>> old = br.open("http://example.com")
+>>> junk = br.open("http://example.com/2")
+>>> new = br.back()
+>>> br.reloaded
+True
+>>> new.wrapped is not old.wrapped
+True
+
+
 Warn early about some mistakes setting a response object
 
 >>> import StringIO

From jjlee at codespeak.net  Sun Jan  7 18:31:25 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 18:31:25 +0100 (CET)
Subject: [wwwsearch-commits] r36197 - wwwsearch/mechanize/common
Message-ID: <20070107173125.178EE10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 18:31:23 2007
New Revision: 36197

Modified:
   wwwsearch/mechanize/common/ChangeLog
Log:
Update ChangeLog

Modified: wwwsearch/mechanize/common/ChangeLog
==============================================================================
--- wwwsearch/mechanize/common/ChangeLog	(original)
+++ wwwsearch/mechanize/common/ChangeLog	Sun Jan  7 18:31:23 2007
@@ -1,7 +1,42 @@
 This isn't really in proper GNU ChangeLog format, it just happens to
 look that way.
 
-2006-10-14 John J Lee <jjl at pobox.com>
+2007-01-07 John J Lee <jjl at pobox.com>
+
+	* 0.1.6b release
+	* Add mechanize.ParseError class, document it as part of the
+	  mechanize.Factory interface, and raise it from all Factory
+	  implementations.  This is backwards-compatible, since the new
+	  exception derives from the old exceptions.
+	* Bug fix: Truncation when there is no full .read() before
+	  navigating to the next page, and an old response is read after
+	  navigation.  This happened e.g. with r = br.open();
+	  r.readline(); br.open(url); r.read(); br.back() .
+	* Bug fix: when .back() caused a reload, it was returning the old
+	  response, not the .reload()ed one.
+	* Bug fix: .back() was not returning a copy of the response, which
+	  presumably would cause seek position problems.
+	* Bug fix: base tag without href attribute would override document
+	  URL with a None value, causing a crash (thanks Nathan Eror).
+	* Fix .set_response() to close current response first.
+	* Fix non-idempotent behaviour of Factory.forms() / .links() .
+	  Previously, if for example you got a ParseError during execution
+	  of .forms(), you could call it again and have it not raise an
+	  exception, because it started out where it left off!
+	* Add a missing copy.copy() to RobustFactory .
+	* Fix redirection to 'URIs' that contain characters that are not
+	  allowed in URIs (thanks Riko Wichmann).  Also, Request
+	  constructor now logs a module logging warning about any such bad
+	  URIs.
+	* Add .global_form() method to Browser to support form controls
+	  whose HTML elements are not descendants of any FORM element.
+	* Add a new method .visit_response() .  This creates a new history
+	  entry from a response object, rather than just changing the
+	  current visited response.  This is useful e.g. when you want to
+	  use Browser features in a handler.
+	* Misc minor bug fixes.
+
+2006-10-25 John J Lee <jjl at pobox.com>
 
 	* 0.1.5b release: Update setuptools dependencies to depend on
 	  ClientForm>=0.2.5 (for an important bug fix affecting fragments

From jjlee at codespeak.net  Sun Jan  7 18:38:28 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 18:38:28 +0100 (CET)
Subject: [wwwsearch-commits] r36198 - wwwsearch/ClientForm/common
Message-ID: <20070107173828.7618F10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 18:38:25 2007
New Revision: 36198

Modified:
   wwwsearch/ClientForm/common/ChangeLog
Log:
Update ChangeLog

Modified: wwwsearch/ClientForm/common/ChangeLog
==============================================================================
--- wwwsearch/ClientForm/common/ChangeLog	(original)
+++ wwwsearch/ClientForm/common/ChangeLog	Sun Jan  7 18:38:25 2007
@@ -1,6 +1,20 @@
 This isn't really in proper GNU ChangeLog format, it just happens to
 look that way.
 
+2007-01-07 John J Lee <jjl at pobox.com>
+	* 0.2.6 release:
+	* Don't allow underlying parser errors (e.g. SGMLParseError)
+	  through -- always raise ClientForm.ParseError .  To preserve.
+	  However, any code that distinguishes between
+	  ClientForm.ParseError and these other exceptions will break.
+	* Allow controls to appear outside of forms (the HTML spec allows
+	  this).  This involved adding new functions ParseFileEx and
+	  ParseResponseEx, which return a list that's always one longer
+	  than the return value of their counterparts ParseFile and
+	  ParseResponse.  The new first element in the list of forms is an
+	  HTMLForm representing the collection of all forms that lie
+	  outside of any FORM element.
+
 2006-10-24 John J Lee <jjl at pobox.com>
 	* 0.2.5 release:
 	* Fix fragment bug introduced in 0.2.4 (thanks Dave Marble).  This

From jjlee at codespeak.net  Sun Jan  7 19:01:40 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:01:40 +0100 (CET)
Subject: [wwwsearch-commits] r36199 - wwwsearch/release_scripts
Message-ID: <20070107180140.AB4DF10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:01:38 2007
New Revision: 36199

Modified:
   wwwsearch/release_scripts/cfrelease.py
Log:
Add 2.5 to list of Pythons to test

Modified: wwwsearch/release_scripts/cfrelease.py
==============================================================================
--- wwwsearch/release_scripts/cfrelease.py	(original)
+++ wwwsearch/release_scripts/cfrelease.py	Sun Jan  7 19:01:38 2007
@@ -32,7 +32,7 @@
 
     sdist = release.SDist(NAME, src.fs_path, version, pretend=pretend)
 
-    sdist.test_pythons = [(2,4), (2,3), (2,2), (2,1), (2,0)]
+    sdist.test_pythons = [(2,5), (2,4), (2,3), (2,2), (2,1), (2,0)]
     sdist.add_svn_working_copies([www, common, src])
     def make_readmes(fn, pretend=False):
         readme_html = release.empy(fn, pretend=pretend)

From jjlee at codespeak.net  Sun Jan  7 19:03:16 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:03:16 +0100 (CET)
Subject: [wwwsearch-commits] r36200 -
	wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48
Message-ID: <20070107180316.8F97A10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:03:13 2007
New Revision: 36200

Added:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48/
      - copied from r36199, wwwsearch/ClientForm/trunk/
Modified:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48/setup.cfg
Log:
Tagged ClientForm (trunk -r HEAD, from working copy) release 0.2.6-2007-01-07T18:00:48

Modified: wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48/setup.cfg
==============================================================================
--- wwwsearch/ClientForm/trunk/setup.cfg	(original)
+++ wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48/setup.cfg	Sun Jan  7 19:03:13 2007
@@ -1,3 +0,0 @@
-[egg_info]
-tag_build = .dev
-tag_svn_revision = 1

From jjlee at codespeak.net  Sun Jan  7 19:29:37 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:29:37 +0100 (CET)
Subject: [wwwsearch-commits] r36201 - wwwsearch/ClientForm/trunk
Message-ID: <20070107182937.996B610070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:29:35 2007
New Revision: 36201

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
   wwwsearch/ClientForm/trunk/test.py
Log:
2.0 fixes

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sun Jan  7 19:29:35 2007
@@ -447,8 +447,14 @@
                      ):
         pass
 else:
-    class ParseError(sgmllib.SGMLParser):
-        pass
+    if hasattr(sgmllib, "SGMLParseError"):
+        SGMLLIB_PARSEERROR = sgmllib.SGMLParseError
+        class ParseError(sgmllib.SGMLParseError):
+            pass
+    else:
+        SGMLLIB_PARSEERROR = RuntimeError
+        class ParseError(RuntimeError):
+            pass
 
 
 class _AbstractFormParser:
@@ -858,7 +864,7 @@
     def feed(self, data):
         try:
             sgmllib.SGMLParser.feed(self, data)
-        except sgmllib.SGMLParseError, exc:
+        except SGMLLIB_PARSEERROR, exc:
             raise ParseError(exc)
 
 
@@ -881,7 +887,7 @@
         def feed(self, data):
             try:
                 self.bs_base_class.feed(self, data)
-            except sgmllib.SGMLParseError, exc:
+            except SGMLLIB_PARSEERROR, exc:
                 raise ParseError(exc)
 
 

Modified: wwwsearch/ClientForm/trunk/test.py
==============================================================================
--- wwwsearch/ClientForm/trunk/test.py	(original)
+++ wwwsearch/ClientForm/trunk/test.py	Sun Jan  7 19:29:35 2007
@@ -225,15 +225,19 @@
 
     def test_failing_parse(self):
         # XXX couldn't provoke an error from BeautifulSoup (!), so this has not
-        # been tested with RobuststFormParser
+        # been tested with RobustFormParser
         import sgmllib
-        f = StringIO("<!!!!>")
-        base_uri = "http://localhost/"
-        self.assertRaises(
-            ClientForm.ParseError,
-            ClientForm.ParseFile, f, base_uri, backwards_compat=False,
-            )
-        self.assert_(issubclass(ClientForm.ParseError, sgmllib.SGMLParseError))
+        # Python 2.0 sgmllib raises RuntimeError rather than SGMLParseError,
+        # but seems never to even raise that except as an assertion, from
+        # reading the code...
+        if hasattr(sgmllib, "SGMLParseError"):
+            f = StringIO("<!!!!>")
+            base_uri = "http://localhost/"
+            self.assertRaises(
+                ClientForm.ParseError,
+                ClientForm.ParseFile, f, base_uri, backwards_compat=False,
+                )
+            self.assert_(issubclass(ClientForm.ParseError, sgmllib.SGMLParseError))
 
     def test_unknown_control(self):
         f = StringIO(

From jjlee at codespeak.net  Sun Jan  7 19:30:48 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:30:48 +0100 (CET)
Subject: [wwwsearch-commits] r36202 -
	wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48
Message-ID: <20070107183048.9353B10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:30:47 2007
New Revision: 36202

Removed:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:00:48/
Log:
Remove unused tag

From jjlee at codespeak.net  Sun Jan  7 19:31:02 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:31:02 +0100 (CET)
Subject: [wwwsearch-commits] r36203 -
	wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36
Message-ID: <20070107183102.A798B10075@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:31:01 2007
New Revision: 36203

Added:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36/
      - copied from r36202, wwwsearch/ClientForm/trunk/
Modified:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36/setup.cfg
Log:
Tagged ClientForm (trunk -r HEAD, from working copy) release 0.2.6-2007-01-07T18:28:36

Modified: wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36/setup.cfg
==============================================================================
--- wwwsearch/ClientForm/trunk/setup.cfg	(original)
+++ wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36/setup.cfg	Sun Jan  7 19:31:01 2007
@@ -1,3 +0,0 @@
-[egg_info]
-tag_build = .dev
-tag_svn_revision = 1

From jjlee at codespeak.net  Sun Jan  7 19:32:16 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:32:16 +0100 (CET)
Subject: [wwwsearch-commits] r36204 - wwwsearch/ClientForm/trunk
Message-ID: <20070107183216.5739210070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:32:15 2007
New Revision: 36204

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
Log:
2.0 fixes

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sun Jan  7 19:32:15 2007
@@ -442,6 +442,7 @@
 # for backwards compatibility, ParseError derives from exceptions that were
 # raised by versions of ClientForm <= 0.2.5
 if HAVE_MODULE_HTMLPARSER:
+    SGMLLIB_PARSEERROR = sgmllib.SGMLParseError
     class ParseError(sgmllib.SGMLParseError,
                      HTMLParser.HTMLParseError,
                      ):

From jjlee at codespeak.net  Sun Jan  7 19:32:37 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:32:37 +0100 (CET)
Subject: [wwwsearch-commits] r36205 -
	wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36
Message-ID: <20070107183237.9DC2510075@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:32:36 2007
New Revision: 36205

Removed:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:28:36/
Log:
Remove unused tag

From jjlee at codespeak.net  Sun Jan  7 19:32:46 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 19:32:46 +0100 (CET)
Subject: [wwwsearch-commits] r36206 -
	wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:30:20
Message-ID: <20070107183246.0AC2C10070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 19:32:44 2007
New Revision: 36206

Added:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:30:20/
      - copied from r36205, wwwsearch/ClientForm/trunk/
Modified:
   wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:30:20/setup.cfg
Log:
Tagged ClientForm (trunk -r HEAD, from working copy) release 0.2.6-2007-01-07T18:30:20

Modified: wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:30:20/setup.cfg
==============================================================================
--- wwwsearch/ClientForm/trunk/setup.cfg	(original)
+++ wwwsearch/ClientForm/tag/release/0.2.6-2007-01-07T18:30:20/setup.cfg	Sun Jan  7 19:32:44 2007
@@ -1,3 +0,0 @@
-[egg_info]
-tag_build = .dev
-tag_svn_revision = 1

From jjlee at codespeak.net  Sun Jan  7 20:21:02 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 20:21:02 +0100 (CET)
Subject: [wwwsearch-commits] r36207 - wwwsearch/ClientForm/trunk
Message-ID: <20070107192102.EE31C10076@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 20:21:00 2007
New Revision: 36207

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
   wwwsearch/ClientForm/trunk/README.html.in
   wwwsearch/ClientForm/trunk/setup.py
Log:
Increment version

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sun Jan  7 20:21:00 2007
@@ -125,7 +125,7 @@
     def deprecation(message):
         warnings.warn(message, DeprecationWarning, stacklevel=2)
 
-VERSION = "0.2.6"
+VERSION = "0.2.7"
 
 CHUNK = 1024  # size of chunks fed to parser, in bytes
 

Modified: wwwsearch/ClientForm/trunk/README.html.in
==============================================================================
--- wwwsearch/ClientForm/trunk/README.html.in	(original)
+++ wwwsearch/ClientForm/trunk/README.html.in	Sun Jan  7 20:21:00 2007
@@ -182,7 +182,7 @@
 methods are still there, but some have been deprecated and a few added).
 
 <ul>
-@{version = "0.2.6"}
+@{version = "0.2.7"}
 <li><a href="./src/ClientForm-@(version).tar.gz">ClientForm-@(version).tar.gz</a>
 <li><a href="./src/ClientForm-@(version).zip">ClientForm-@(version).zip</a>
 <li><a href="./src/ChangeLog.txt">Change Log</a> (included in distribution)

Modified: wwwsearch/ClientForm/trunk/setup.py
==============================================================================
--- wwwsearch/ClientForm/trunk/setup.py	(original)
+++ wwwsearch/ClientForm/trunk/setup.py	Sun Jan  7 20:21:00 2007
@@ -15,7 +15,7 @@
 import re
 #VERSION_MATCH = re.search(r'VERSION = "(.*)"', open("ClientForm.py").read())
 #VERSION = VERSION_MATCH.group(1)
-VERSION = '0.2.6'
+VERSION = '0.2.7'
 INSTALL_REQUIRES = []
 NAME = "ClientForm"
 PACKAGE = False

From jjlee at codespeak.net  Sun Jan  7 20:22:41 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 20:22:41 +0100 (CET)
Subject: [wwwsearch-commits] r36208 -
	wwwsearch/mechanize/tag/release/0.1.6b-2007-01-07T19:20:13
Message-ID: <20070107192241.4D02410076@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 20:22:39 2007
New Revision: 36208

Added:
   wwwsearch/mechanize/tag/release/0.1.6b-2007-01-07T19:20:13/
      - copied from r36207, wwwsearch/mechanize/trunk/
Modified:
   wwwsearch/mechanize/tag/release/0.1.6b-2007-01-07T19:20:13/setup.cfg
Log:
Tagged mechanize (trunk -r HEAD, from working copy) release 0.1.6b-2007-01-07T19:20:13

Modified: wwwsearch/mechanize/tag/release/0.1.6b-2007-01-07T19:20:13/setup.cfg
==============================================================================
--- wwwsearch/mechanize/trunk/setup.cfg	(original)
+++ wwwsearch/mechanize/tag/release/0.1.6b-2007-01-07T19:20:13/setup.cfg	Sun Jan  7 20:22:39 2007
@@ -1,3 +0,0 @@
-[egg_info]
-tag_build = .dev
-tag_svn_revision = 1

From jjlee at codespeak.net  Sun Jan  7 22:44:11 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 22:44:11 +0100 (CET)
Subject: [wwwsearch-commits] r36209 - in wwwsearch/mechanize/trunk: . attic
	docs-in-progress examples mechanize test test-tools
Message-ID: <20070107214411.3F00310070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 22:43:57 2007
New Revision: 36209

Modified:
   wwwsearch/mechanize/trunk/COPYING.txt   (props changed)
   wwwsearch/mechanize/trunk/INSTALL.txt   (props changed)
   wwwsearch/mechanize/trunk/MANIFEST.in   (props changed)
   wwwsearch/mechanize/trunk/README.html.in   (props changed)
   wwwsearch/mechanize/trunk/attic/BSDDBCookieJar.py   (props changed)
   wwwsearch/mechanize/trunk/attic/MSIEDBCookieJar.py   (props changed)
   wwwsearch/mechanize/trunk/doc.html.in   (props changed)
   wwwsearch/mechanize/trunk/docs-in-progress/doc.rst   (props changed)
   wwwsearch/mechanize/trunk/docs-in-progress/doc2.rst   (props changed)
   wwwsearch/mechanize/trunk/examples/hack21.py   (props changed)
   wwwsearch/mechanize/trunk/examples/pypi.py   (props changed)
   wwwsearch/mechanize/trunk/ez_setup.py   (props changed)
   wwwsearch/mechanize/trunk/functional_tests.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/__init__.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_auth.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_beautifulsoup.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_clientcookie.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_debug.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_gzip.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_headersutil.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_html.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_http.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_lwpcookiejar.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_mechanize.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_mozillacookiejar.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_msiecookiejar.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_opener.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_pullparser.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_request.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_response.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_rfc3986.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_seek.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_upgrade.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_urllib2.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_useragent.py   (props changed)
   wwwsearch/mechanize/trunk/mechanize/_util.py   (props changed)
   wwwsearch/mechanize/trunk/setup.cfg   (props changed)
   wwwsearch/mechanize/trunk/setup.py   (props changed)
   wwwsearch/mechanize/trunk/test-tools/doctest.py   (props changed)
   wwwsearch/mechanize/trunk/test-tools/linecache_copy.py   (props changed)
   wwwsearch/mechanize/trunk/test.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_browser.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_browser.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_cookies.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_date.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_forms.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_headers.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_history.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_html.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_html.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_opener.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_password_manager.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_pullparser.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_request.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_response.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_response.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_rfc3986.doctest   (props changed)
   wwwsearch/mechanize/trunk/test/test_urllib2.py   (props changed)
   wwwsearch/mechanize/trunk/test/test_useragent.py   (props changed)
Log:
Set svn:eol-style to native for all text files

From jjlee at codespeak.net  Sun Jan  7 22:48:37 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 22:48:37 +0100 (CET)
Subject: [wwwsearch-commits] r36210 - wwwsearch/mechanize/common
Message-ID: <20070107214837.BDFA410070@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 22:48:36 2007
New Revision: 36210

Modified:
   wwwsearch/mechanize/common/ChangeLog   (props changed)
Log:
Set ChangeLog svn:eol-style native

From jjlee at codespeak.net  Sun Jan  7 22:50:36 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun,  7 Jan 2007 22:50:36 +0100 (CET)
Subject: [wwwsearch-commits] r36212 - wwwsearch/ClientForm/common
Message-ID: <20070107215036.262F010075@code0.codespeak.net>

Author: jjlee
Date: Sun Jan  7 22:50:34 2007
New Revision: 36212

Modified:
   wwwsearch/ClientForm/common/ChangeLog   (props changed)
Log:
Set ChangeLog svn:eol-style native

From jjlee at codespeak.net  Wed Jan 10 23:44:28 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Wed, 10 Jan 2007 23:44:28 +0100 (CET)
Subject: [wwwsearch-commits] r36445 - wwwsearch/mechanize/trunk/mechanize
Message-ID: <20070110224428.6C2D910089@code0.codespeak.net>

Author: jjlee
Date: Wed Jan 10 23:44:27 2007
New Revision: 36445

Modified:
   wwwsearch/mechanize/trunk/mechanize/_response.py
Log:
Fix MRO error caused by double-wrapping of error response (I think the actual exception only showed up with Python 2.5, but the underlying problem was there since this bit of code was added)

Modified: wwwsearch/mechanize/trunk/mechanize/_response.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_response.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_response.py	Wed Jan 10 23:44:27 2007
@@ -413,7 +413,8 @@
 
     Accepts responses from both mechanize and urllib2 handlers.
     """
-    if isinstance(response, urllib2.HTTPError):
+    if (isinstance(response, urllib2.HTTPError) and
+        not isinstance(response, response_seek_wrapper)):
         class httperror_seek_wrapper(response_seek_wrapper, response.__class__):
             # this only derives from HTTPError in order to be a subclass --
             # the HTTPError behaviour comes from delegation

From jjlee at codespeak.net  Thu Jan 11 22:20:11 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Thu, 11 Jan 2007 22:20:11 +0100 (CET)
Subject: [wwwsearch-commits] r36532 - wwwsearch/mechanize/trunk
Message-ID: <20070111212011.7A83210061@code0.codespeak.net>

Author: jjlee
Date: Thu Jan 11 22:20:08 2007
New Revision: 36532

Modified:
   wwwsearch/mechanize/trunk/functional_tests.py
Log:
Add a functional test for 302 + 404

Modified: wwwsearch/mechanize/trunk/functional_tests.py
==============================================================================
--- wwwsearch/mechanize/trunk/functional_tests.py	(original)
+++ wwwsearch/mechanize/trunk/functional_tests.py	Thu Jan 11 22:20:08 2007
@@ -45,6 +45,16 @@
         self.browser.open('/mechanize/')
         self.assertEqual(self.browser.title(), 'mechanize')
 
+    def test_302_and_404(self):
+        # the combination of 302 (caused by use of "sf.net") and 404 has caused
+        # problems in the past due to accidental double-wrapping of the error
+        # response
+        import urllib2
+        self.assertRaises(
+            urllib2.HTTPError,
+            self.browser.open, "http://wwwsearch.sf.net/doesnotexist"
+            )
+
     def test_reread(self):
         r = self.browser.open('http://wwwsearch.sourceforge.net/')
         data = r.read()

From jjlee at codespeak.net  Thu Jan 11 22:29:20 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Thu, 11 Jan 2007 22:29:20 +0100 (CET)
Subject: [wwwsearch-commits] r36535 - wwwsearch/mechanize/trunk/test
Message-ID: <20070111212920.2059A10061@code0.codespeak.net>

Author: jjlee
Date: Thu Jan 11 22:28:43 2007
New Revision: 36535

Modified:
   wwwsearch/mechanize/trunk/test/test_opener.py
Log:
Add missing import in tests (caused failure on Windows)

Modified: wwwsearch/mechanize/trunk/test/test_opener.py
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_opener.py	(original)
+++ wwwsearch/mechanize/trunk/test/test_opener.py	Thu Jan 11 22:28:43 2007
@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-import os, math
+import os, math, stat
 from unittest import TestCase
 
 import mechanize

From jjlee at codespeak.net  Sat Jan 20 01:09:30 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 20 Jan 2007 01:09:30 +0100 (CET)
Subject: [wwwsearch-commits] r37045 - wwwsearch/ClientForm/trunk
Message-ID: <20070120000930.B762910077@code0.codespeak.net>

Author: jjlee
Date: Sat Jan 20 01:09:28 2007
New Revision: 37045

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
   wwwsearch/ClientForm/trunk/test.py
Log:
Fix entity ref double-decoding bug (dmoews at fastmail.fm, bshanks at ucsd.edu)

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sat Jan 20 01:09:28 2007
@@ -488,7 +488,7 @@
         debug("%s", attrs)
         for key, value in attrs:
             if key == "href":
-                self.base = value
+                self.base = self.unescape_attr_if_required(value)
 
     def end_body(self):
         debug("")
@@ -508,14 +508,14 @@
         d = {}
         for key, value in attrs:
             if key == "name":
-                name = value
+                name = self.unescape_attr_if_required(value)
             elif key == "action":
-                action = value
+                action = self.unescape_attr_if_required(value)
             elif key == "method":
-                method = value.upper()
+                method = self.unescape_attr_if_required(value.upper())
             elif key == "enctype":
-                enctype = value.lower()
-            d[key] = value
+                enctype = self.unescape_attr_if_required(value.lower())
+            d[key] = self.unescape_attr_if_required(value)
         controls = []
         self._current_form = (name, action, method, enctype), d, controls
 
@@ -536,7 +536,7 @@
             raise ParseError("SELECT inside TEXTAREA")
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
 
         self._select = d
         self._add_label(d)
@@ -561,7 +561,7 @@
             raise ParseError("OPTGROUP outside of SELECT")
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
 
         self._optgroup = d
 
@@ -580,7 +580,7 @@
 
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
 
         self._option = {}
         self._option.update(d)
@@ -619,7 +619,7 @@
             raise ParseError("TEXTAREA inside SELECT")
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
         self._add_label(d)
 
         self._textarea = d
@@ -641,7 +641,7 @@
             self.end_label()
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
         taken = bool(d.get("for"))  # empty id is invalid
         d["__text"] = ""
         d["__taken"] = taken
@@ -709,7 +709,7 @@
         d = {}
         d["type"] = "submit"  # default
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
         controls = self._current_form[2]
 
         type = d["type"]
@@ -727,7 +727,7 @@
         d = {}
         d["type"] = "text"  # default
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
         controls = self._current_form[2]
 
         type = d["type"]
@@ -739,7 +739,7 @@
         debug("%s", attrs)
         d = {}
         for key, val in attrs:
-            d[key] = val
+            d[key] = self.unescape_attr_if_required(val)
         controls = self._current_form[2]
 
         self._add_label(d)
@@ -1113,9 +1113,6 @@
             action = base_uri
         else:
             action = _urljoin(base_uri, action)
-        action = fp.unescape_attr_if_required(action)
-        name = fp.unescape_attr_if_required(name)
-        attrs = fp.unescape_attrs_if_required(attrs)
         # would be nice to make HTMLForm class (form builder) pluggable
         form = HTMLForm(
             action, method, enctype, name, attrs, request_class,
@@ -1124,8 +1121,6 @@
         form._urlunparse = _urlunparse
         for ii in range(len(controls)):
             type, name, attrs = controls[ii]
-            attrs = fp.unescape_attrs_if_required(attrs)
-            name = fp.unescape_attr_if_required(name)
             # index=ii*10 allows ImageControl to return multiple ordered pairs
             form.new_control(type, name, attrs, select_default=select_default,
                              index=ii*10)

Modified: wwwsearch/ClientForm/trunk/test.py
==============================================================================
--- wwwsearch/ClientForm/trunk/test.py	(original)
+++ wwwsearch/ClientForm/trunk/test.py	Sat Jan 20 01:09:28 2007
@@ -135,19 +135,38 @@
 
     def test_unescape_parsing(self):
         file = StringIO(
-"""<form action="&amp;&mdash;&#x2014;&#8212;">
-<textarea name="name&amp;&mdash;&#x2014;&#8212;">val&amp;&mdash;&#x2014;&#8212;</textarea>
+"""<form action="&amp;amp;&mdash;&#x2014;&#8212;">
+<textarea name="name&amp;amp;&mdash;&#x2014;&#8212;">val&amp;amp;&mdash;&#x2014;&#8212;</textarea>
 </form>
 """)  #"
         forms = ClientForm.ParseFile(file, "http://localhost/",
                                      backwards_compat=False, encoding="utf-8")
         form = forms[0]
-        test_string = "&"+(u"\u2014".encode('utf8')*3)
+        test_string = "&amp;"+(u"\u2014".encode('utf8')*3)
         self.assertEqual(form.action, "http://localhost/"+test_string)
         control = form.find_control(type="textarea", nr=0)
         self.assertEqual(control.value, "val"+test_string)
         self.assertEqual(control.name, "name"+test_string)
 
+    def test_unescape_parsing_select(self):
+        f = StringIO("""\
+<form>
+<select name="a">
+    <option>1&amp;amp;&mdash;&#x2014;&#8212;</option>
+    <option value="2&amp;amp;&mdash;&#x2014;&#8212;">2&amp;amp;&mdash;&#x2014;&#8212;</option>
+</select>
+</form>
+""")  #"
+        forms = ClientForm.ParseFileEx(
+            f, "http://localhost/", encoding="utf-8")
+        form = forms[1]
+        test_string = "&amp;"+(u"\u2014".encode('utf8')*3)
+        control = form.find_control(nr=0)
+        for ii in range(len(control.items)):
+            item = control.items[ii]
+            self.assertEqual(item.name, str(ii+1)+test_string)
+            # XXX label
+
     def test_unescape_parsing_data(self):
         file = StringIO(
 """\
@@ -403,10 +422,7 @@
         entity_ctl = form.find_control(type="textarea", nr=2)
         self.assertEqual(entity_ctl.name, '"ta"')
         self.assertEqual(entity_ctl.attrs["id"], "foo&amp;bar")
-
-        # sgmllib gets this wrong (unescapes twice) (module HTMLParser does
-        # it right)
-        #self.assertEqual(entity_ctl.value, "Hello testers &amp; users!")
+        self.assertEqual(entity_ctl.value, "Hello testers &amp; users!")
 
     def testSelect(self):
         file = StringIO(

From jjlee at codespeak.net  Sat Jan 20 01:13:52 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 20 Jan 2007 01:13:52 +0100 (CET)
Subject: [wwwsearch-commits] r37046 - wwwsearch/ClientForm/trunk
Message-ID: <20070120001352.C141710077@code0.codespeak.net>

Author: jjlee
Date: Sat Jan 20 01:13:51 2007
New Revision: 37046

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
Log:
Fix debug prints

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sat Jan 20 01:13:51 2007
@@ -71,6 +71,7 @@
 
 try:
     import logging
+    import inspect
 except ImportError:
     def debug(msg, *args, **kwds):
         pass
@@ -82,11 +83,7 @@
         if OPTIMIZATION_HACK:
             return
 
-        try:
-            raise Exception()
-        except:
-            caller_name = (
-                sys.exc_info()[2].tb_frame.f_back.f_back.f_code.co_name)
+        caller_name = inspect.stack()[1][3]
         extended_msg = '%%s %s' % msg
         extended_args = (caller_name,)+args
         debug = _logger.debug(extended_msg, *extended_args, **kwds)

From jjlee at codespeak.net  Sat Jan 20 01:15:05 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 20 Jan 2007 01:15:05 +0100 (CET)
Subject: [wwwsearch-commits] r37047 - wwwsearch/ClientForm/trunk
Message-ID: <20070120001505.EF9B010077@code0.codespeak.net>

Author: jjlee
Date: Sat Jan 20 01:15:02 2007
New Revision: 37047

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
Log:
Note some bugs to fix

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sat Jan 20 01:15:02 2007
@@ -27,6 +27,7 @@
 """
 
 # XXX
+# SELECT outside of FORM loses last OPTION
 # add an __all__
 # Remove parser testing hack
 # safeUrl()-ize action
@@ -1644,6 +1645,8 @@
         return res
 
     def __repr__(self):
+        # XXX appending the attrs without distinguishing them from name and id
+        # is silly
         attrs = [("name", self.name), ("id", self.id)]+self.attrs.items()
         return "<%s %s>" % (
             self.__class__.__name__,

From jjlee at codespeak.net  Sun Jan 21 00:02:57 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun, 21 Jan 2007 00:02:57 +0100 (CET)
Subject: [wwwsearch-commits] r37081 - wwwsearch/mechanize/trunk/test
Message-ID: <20070120230257.7FDB41007F@code0.codespeak.net>

Author: jjlee
Date: Sun Jan 21 00:02:56 2007
New Revision: 37081

Modified:
   wwwsearch/mechanize/trunk/test/test_response.doctest
Log:
Add some tests for upgrade_response

Modified: wwwsearch/mechanize/trunk/test/test_response.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_response.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_response.doctest	Sun Jan 21 00:02:56 2007
@@ -110,3 +110,108 @@
 ''
 >>> r.read_complete
 False
+
+
+
+Tests for the truly horrendous upgrade_response()
+
+>>> def is_response(r):
+...     names = "get_data read readline readlines close seek code msg".split()
+...     for name in names:
+...         if not hasattr(r, name):
+...             return False
+...     return r.get_data() == "test data"
+
+>>> from cStringIO import StringIO
+>>> from mechanize._response import upgrade_response, make_headers, \
+...     make_response, closeable_response, seek_wrapper
+>>> data="test data"; url="http://example.com/"; code=200; msg="OK"
+
+Normal response (closeable_response wrapped with seek_wrapper): return a copy
+
+>>> r1 = make_response(data, [], url, code, msg)
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+>>> r1 is not r2
+True
+>>> r1.wrapped is r2.wrapped
+True
+
+closeable_response with no seek_wrapper: wrap with seek_wrapper
+
+>>> r1 = closeable_response(StringIO(data), make_headers([]), url, code, msg)
+>>> is_response(r1)
+False
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+>>> r1 is not r2
+True
+>>> r1 is r2.wrapped
+True
+
+urllib2.addinfourl: extract .fp and wrap it with closeable_response
+and seek_wrapper
+
+>>> import urllib2
+>>> r1= urllib2.addinfourl(StringIO(data), make_headers([]), url)
+>>> is_response(r1)
+False
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+>>> r1 is not r2
+True
+>>> r1 is not r2.wrapped
+True
+>>> r1.fp is r2.wrapped.fp
+True
+
+addinfourl with code, msg
+
+>>> r1= urllib2.addinfourl(StringIO(data), make_headers([]), url)
+>>> r1.code = 206
+>>> r1.msg = "cool"
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+>>> r2.code == r1.code
+True
+>>> r2.msg == r1.msg
+True
+
+addinfourl with seek wrapper: cached data is not lost
+
+>>> r1= urllib2.addinfourl(StringIO(data), make_headers([]), url)
+>>> r1 = seek_wrapper(r1)
+>>> r1.read(4)
+'test'
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+
+addinfourl wrapped with HTTPError -- remains an HTTPError of the same
+subclass (through horrible trickery)
+
+>>> hdrs = make_headers([])
+>>> r1 = urllib2.addinfourl(StringIO(data), hdrs, url)
+>>> class MyHTTPError(urllib2.HTTPError): pass
+>>> r1 = MyHTTPError(url, code, msg, hdrs, r1)
+>>> is_response(r1)
+False
+>>> r2 = upgrade_response(r1)
+>>> is_response(r2)
+True
+>>> isinstance(r2, MyHTTPError)
+True
+
+The trickery does not cause double-wrapping
+
+>>> r3 = upgrade_response(r2)
+>>> is_response(r3)
+True
+>>> r3 is not r2
+True
+>>> r3.wrapped is r2.wrapped
+True

From jjlee at codespeak.net  Thu Feb  8 02:40:53 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Thu,  8 Feb 2007 02:40:53 +0100 (CET)
Subject: [wwwsearch-commits] r38132 - wwwsearch/mechanize/trunk/mechanize
Message-ID: <20070208014053.8A51910083@code0.codespeak.net>

Author: jjlee
Date: Thu Feb  8 02:40:50 2007
New Revision: 38132

Modified:
   wwwsearch/mechanize/trunk/mechanize/_request.py
Log:
Fix HTTP proxy support: r29110 meant that Request.get_selector() didn't take into account the change to .__r_host (tgates at xplantechnology.com)

Modified: wwwsearch/mechanize/trunk/mechanize/_request.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_request.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_request.py	Thu Feb  8 02:40:50 2007
@@ -49,9 +49,7 @@
         self.origin_req_host = origin_req_host
 
     def get_selector(self):
-        if self.selector is None:
-            self.selector, self.__r_selector = urllib.splittag(self.__r_host)
-        return self.selector
+        return urllib.splittag(self.__r_host)[0]
 
     def get_origin_req_host(self):
         return self.origin_req_host

From jjlee at codespeak.net  Sat Mar 24 17:10:09 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 24 Mar 2007 17:10:09 +0100 (CET)
Subject: [wwwsearch-commits] r41240 - wwwsearch/mechanize/trunk
Message-ID: <20070324161009.0F35A10072@code0.codespeak.net>

Author: jjlee
Date: Sat Mar 24 17:10:07 2007
New Revision: 41240

Modified:
   wwwsearch/mechanize/trunk/test.py
Log:
Add -u switch to skip unittests (and only run the doctests)

Modified: wwwsearch/mechanize/trunk/test.py
==============================================================================
--- wwwsearch/mechanize/trunk/test.py	(original)
+++ wwwsearch/mechanize/trunk/test.py	Sat Mar 24 17:10:07 2007
@@ -144,6 +144,9 @@
     run_doctests = "-d" not in sys.argv
     if not run_doctests:
         sys.argv.remove("-d")
+    run_unittests = "-u" not in sys.argv
+    if not run_unittests:
+        sys.argv.remove("-u")
 
     # import local copy of Python 2.5 doctest
     assert os.path.isdir("test")
@@ -165,6 +168,11 @@
 
     import mechanize
 
+    class DefaultResult:
+        def wasSuccessful(self):
+            return True
+    result = DefaultResult()
+
     if run_doctests:
         # run .doctest files needing special support
         common_globs = {"mechanize": mechanize}
@@ -202,15 +210,16 @@
         doctest.testmod(_pullparser)
         doctest.testmod(_http)
 
-    # run vanilla unittest tests
-    import unittest
-    test_path = os.path.join(os.path.dirname(sys.argv[0]), "test")
-    sys.path.insert(0, test_path)
-    test_runner = None
-    if use_cgitb:
-        test_runner = CgitbTextTestRunner()
-    prog = TestProgram(MODULE_NAMES, testRunner=test_runner)
-    result = prog.runTests()
+    if run_unittests:
+        # run vanilla unittest tests
+        import unittest
+        test_path = os.path.join(os.path.dirname(sys.argv[0]), "test")
+        sys.path.insert(0, test_path)
+        test_runner = None
+        if use_cgitb:
+            test_runner = CgitbTextTestRunner()
+        prog = TestProgram(MODULE_NAMES, testRunner=test_runner)
+        result = prog.runTests()
 
     if run_coverage:
         # HTML coverage report

From jjlee at codespeak.net  Sat Mar 24 17:12:19 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 24 Mar 2007 17:12:19 +0100 (CET)
Subject: [wwwsearch-commits] r41241 - wwwsearch/mechanize/trunk/mechanize
Message-ID: <20070324161219.B134C10071@code0.codespeak.net>

Author: jjlee
Date: Sat Mar 24 17:12:18 2007
New Revision: 41241

Modified:
   wwwsearch/mechanize/trunk/mechanize/_response.py
Log:
Remove a type test and add an assertion

Modified: wwwsearch/mechanize/trunk/mechanize/_response.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_response.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_response.py	Sat Mar 24 17:12:18 2007
@@ -414,7 +414,7 @@
     Accepts responses from both mechanize and urllib2 handlers.
     """
     if (isinstance(response, urllib2.HTTPError) and
-        not isinstance(response, response_seek_wrapper)):
+        not hasattr(response, "seek")):
         class httperror_seek_wrapper(response_seek_wrapper, response.__class__):
             # this only derives from HTTPError in order to be a subclass --
             # the HTTPError behaviour comes from delegation
@@ -440,6 +440,7 @@
     if hasattr(response, "closeable_response"):
         if not hasattr(response, "seek"):
             response = wrapper_class(response)
+        assert hasattr(response, "get_data")
         return copy.copy(response)
 
     # a urllib2 handler constructed the response, i.e. the response is an

From jjlee at codespeak.net  Sat Mar 24 19:18:52 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 24 Mar 2007 19:18:52 +0100 (CET)
Subject: [wwwsearch-commits] r41252 - in wwwsearch/mechanize/trunk:
	mechanize test
Message-ID: <20070324181852.631AE10071@code0.codespeak.net>

Author: jjlee
Date: Sat Mar 24 19:18:50 2007
New Revision: 41252

Modified:
   wwwsearch/mechanize/trunk/mechanize/_http.py
   wwwsearch/mechanize/trunk/mechanize/_opener.py
   wwwsearch/mechanize/trunk/mechanize/_urllib2.py
   wwwsearch/mechanize/trunk/test/test_browser.py
   wwwsearch/mechanize/trunk/test/test_urllib2.py
Log:
Remove most ugly part of mock objects used in tests, and fix a multiple-response-wrapping bug that got revealed as a result

Modified: wwwsearch/mechanize/trunk/mechanize/_http.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_http.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_http.py	Sat Mar 24 19:18:50 2007
@@ -578,6 +578,23 @@
     https_response = http_response
 
 
+class HTTPDefaultErrorHandler(BaseHandler):
+    def http_error_default(self, req, fp, code, msg, hdrs):
+        # why these error methods took the code, msg, headers args in the first
+        # place rather than a response object, I don't know, but to avoid
+        # multiple wrapping, we're discarding them
+
+        if isinstance(fp, urllib2.HTTPError):
+            response = fp
+        else:
+            response = urllib2.HTTPError(
+                req.get_full_url(), code, msg, hdrs, fp)
+        assert code == response.code
+        assert msg == response.msg
+        assert hdrs == response.hdrs
+        raise response
+
+
 class AbstractHTTPHandler(BaseHandler):
 
     def __init__(self, debuglevel=0):

Modified: wwwsearch/mechanize/trunk/mechanize/_opener.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_opener.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_opener.py	Sat Mar 24 19:18:50 2007
@@ -304,7 +304,7 @@
         urllib2.ProxyHandler,
         urllib2.UnknownHandler,
         _http.HTTPHandler,  # derived from new AbstractHTTPHandler
-        urllib2.HTTPDefaultErrorHandler,
+        _http.HTTPDefaultErrorHandler,
         _http.HTTPRedirectHandler,  # bugfixed
         urllib2.FTPHandler,
         urllib2.FileHandler,

Modified: wwwsearch/mechanize/trunk/mechanize/_urllib2.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_urllib2.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_urllib2.py	Sat Mar 24 19:18:50 2007
@@ -28,7 +28,6 @@
 # ...from urllib2...
 from urllib2 import \
      BaseHandler, \
-     HTTPDefaultErrorHandler, \
      UnknownHandler, \
      FTPHandler, \
      CacheFTPHandler, \
@@ -37,6 +36,7 @@
 # ...and from mechanize
 from _http import \
      HTTPHandler, \
+     HTTPDefaultErrorHandler, \
      HTTPRedirectHandler, \
      HTTPEquivProcessor, \
      HTTPCookieProcessor, \

Modified: wwwsearch/mechanize/trunk/test/test_browser.py
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_browser.py	(original)
+++ wwwsearch/mechanize/trunk/test/test_browser.py	Sat Mar 24 19:18:50 2007
@@ -32,9 +32,6 @@
         self.fp = StringIO.StringIO(data)
         if info is None: info = {}
         self._info = MockHeaders(info)
-        self.source = "%d%d" % (id(self), random.randint(0, sys.maxint-1))
-        # otherwise we can't test for "same_response" in test_history
-        self.read_complete = True
     def info(self): return self._info
     def geturl(self): return self.url
     def read(self, size=-1): return self.fp.read(size)
@@ -43,12 +40,6 @@
         self.fp.seek(0)
     def close(self): pass
     def get_data(self): pass
-    def __getstate__(self):
-        state = self.__dict__
-        state['source'] = self.source
-        return state
-    def __setstate__(self, state):
-        self.__dict__ = state
 
 def make_mock_handler(response_class=MockResponse):
     class MockHandler:
@@ -189,12 +180,21 @@
 
     def test_history(self):
         import mechanize
+        from mechanize import _response
 
         def same_response(ra, rb):
-            return ra.source == rb.source
+            return ra.wrapped is rb.wrapped
 
-        b = TestBrowser()
-        b.add_handler(make_mock_handler()([("http_open", None)]))
+        class Handler(mechanize.BaseHandler):
+            def http_open(self, request):
+                r = _response.test_response(url=request.get_full_url())
+                # these tests aren't interested in auto-.reload() behaviour of
+                # .back(), so read the response to prevent that happening
+                r.get_data()
+                return r
+
+        b = TestBrowser2()
+        b.add_handler(Handler())
         self.assertRaises(mechanize.BrowserStateError, b.back)
         r1 = b.open("http://example.com/")
         self.assertRaises(mechanize.BrowserStateError, b.back)
@@ -204,6 +204,7 @@
         r4 = b.open("http://example.com/spam")
         self.assert_(same_response(b.back(), r3))
         self.assert_(same_response(b.back(), r1))
+        self.assertEquals(b.geturl(), "http://example.com/")
         self.assertRaises(mechanize.BrowserStateError, b.back)
         # reloading does a real HTTP fetch rather than using history cache
         r5 = b.reload()
@@ -221,13 +222,17 @@
         self.assertRaises(mechanize.BrowserStateError, b.back, 2)
         r8 = b.open("/spam")
 
-        # even if we get a HTTPError, history, .response() and .request should
+        # even if we get an HTTPError, history, .response() and .request should
         # still get updated
-        error = urllib2.HTTPError("http://example.com/bad", 503, "Oops",
-                                  MockHeaders(), StringIO.StringIO())
-        b.add_handler(make_mock_handler()([("https_open", error)]))
+        class Handler2(mechanize.BaseHandler):
+            def https_open(self, request):
+                r = urllib2.HTTPError(
+                    "https://example.com/bad", 503, "Oops",
+                    MockHeaders(), StringIO.StringIO())
+                return r
+        b.add_handler(Handler2())
         self.assertRaises(urllib2.HTTPError, b.open, "https://example.com/badreq")
-        self.assertEqual(b.response().geturl(), error.geturl())
+        self.assertEqual(b.response().geturl(), "https://example.com/bad")
         self.assertEqual(b.request.get_full_url(), "https://example.com/badreq")
         self.assert_(same_response(b.back(), r8))
 

Modified: wwwsearch/mechanize/trunk/test/test_urllib2.py
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_urllib2.py	(original)
+++ wwwsearch/mechanize/trunk/test/test_urllib2.py	Sat Mar 24 19:18:50 2007
@@ -33,6 +33,10 @@
 ## l = getLogger("mechanize")
 ## l.setLevel(DEBUG)
 
+class AlwaysEqual:
+    def __cmp__(self, other):
+        return 0
+
 class MockOpener:
     addheaders = []
     def open(self, req, data=None):
@@ -259,6 +263,20 @@
                 if expected_args[j] is not Unknown:
                     self.assert_(args[j] == expected_args[j])
 
+    def test_http_error_raised(self):
+        # should get an HTTPError if an HTTP handler raises a non-200 response
+        # XXX it worries me that this is the only test that excercises the else
+        # branch in HTTPDefaultErrorHandler
+        from mechanize import _response
+        o = mechanize.OpenerDirector()
+        o.add_handler(mechanize.HTTPErrorProcessor())
+        o.add_handler(mechanize.HTTPDefaultErrorHandler())
+        class HTTPHandler(AbstractHTTPHandler):
+            def http_open(self, req):
+                return _response.test_response(code=302)
+        o.add_handler(HTTPHandler())
+        self.assertRaises(mechanize.HTTPError, o.open, "http://example.com/")
+
     def test_processors(self):
         # *_request / *_response methods get called appropriately
         o = OpenerDirector()
@@ -656,21 +674,52 @@
         self.assert_(req.unredirected_hdrs["Referer"] == ref)
 
     def test_errors(self):
+        from mechanize import _response
         h = HTTPErrorProcessor()
         o = h.parent = MockOpener()
 
-        url = "http://example.com/"
-        req = Request(url)
+        req = Request("http://example.com")
         # 200 OK is passed through
-        r = MockResponse(200, "OK", {}, "", url)
+        r = _response.test_response()
         newr = h.http_response(req, r)
         self.assert_(r is newr)
         self.assert_(not hasattr(o, "proto"))  # o.error not called
         # anything else calls o.error (and MockOpener returns None, here)
-        r = MockResponse(201, "Created", {}, "", url)
+        r = _response.test_response(code=201, msg="Created")
         self.assert_(h.http_response(req, r) is None)
         self.assert_(o.proto == "http")  # o.error called
-        self.assert_(o.args == (req, r, 201, "Created", {}))
+        self.assert_(o.args == (req, r, 201, "Created", AlwaysEqual()))
+
+    def test_raise_http_errors(self):
+        # HTTPDefaultErrorHandler should raise HTTPError if no error handler
+        # handled the error response
+        from mechanize import _response
+        h = mechanize.HTTPDefaultErrorHandler()
+
+        url = "http://example.com"; code = 500; msg = "Error"
+        request = mechanize.Request(url)
+        response = _response.test_response(url=url, code=code, msg=msg)
+
+        # case 1. it's not an HTTPError
+        try:
+            h.http_error_default(
+                request, response, code, msg, response.info())
+        except mechanize.HTTPError, exc:
+            self.assert_(exc is not response)
+            self.assert_(exc.fp is response)
+        else:
+            self.assert_(False)
+
+        # case 2. response object is already an HTTPError, so just re-raise it
+        error = mechanize.HTTPError(
+            url, code, msg, "fake headers", response)
+        try:
+            h.http_error_default(
+                request, error, code, msg, error.info())
+        except mechanize.HTTPError, exc:
+            self.assert_(exc is error)
+        else:
+            self.assert_(False)
 
     def test_robots(self):
         # XXX useragent

From jjlee at codespeak.net  Sun Mar 25 17:55:54 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sun, 25 Mar 2007 17:55:54 +0200 (CEST)
Subject: [wwwsearch-commits] r41292 - wwwsearch/ClientForm/trunk
Message-ID: <20070325155554.DD6C810083@code0.codespeak.net>

Author: jjlee
Date: Sun Mar 25 17:55:52 2007
New Revision: 41292

Modified:
   wwwsearch/ClientForm/trunk/ClientForm.py
   wwwsearch/ClientForm/trunk/test.py
Log:
Fix some deprecation bugs: Fix stacklevel (warnings should cite calling code, not library code).  Fix cases where we were accidentally messing with global warning state where we just wanted a utility function to temporarily hide warnings.  Also, fix one case to use backwards_compat=False.

Modified: wwwsearch/ClientForm/trunk/ClientForm.py
==============================================================================
--- wwwsearch/ClientForm/trunk/ClientForm.py	(original)
+++ wwwsearch/ClientForm/trunk/ClientForm.py	Sun Mar 25 17:55:52 2007
@@ -117,11 +117,11 @@
 try:
     import warnings
 except ImportError:
-    def deprecation(message):
+    def deprecation(message, stack_offset=0):
         pass
 else:
-    def deprecation(message):
-        warnings.warn(message, DeprecationWarning, stacklevel=2)
+    def deprecation(message, stack_offset=0):
+        warnings.warn(message, DeprecationWarning, stacklevel=3+stack_offset)
 
 VERSION = "0.2.7"
 
@@ -1081,7 +1081,7 @@
                  _urlunparse=urlparse.urlunparse,
                  ):
     if backwards_compat:
-        deprecation("operating in backwards-compatibility mode")
+        deprecation("operating in backwards-compatibility mode", 1)
     fp = form_parser_class(entitydefs, encoding)
     while 1:
         data = file.read(CHUNK)

Modified: wwwsearch/ClientForm/trunk/test.py
==============================================================================
--- wwwsearch/ClientForm/trunk/test.py	(original)
+++ wwwsearch/ClientForm/trunk/test.py	Sun Mar 25 17:55:52 2007
@@ -50,6 +50,7 @@
         warnings.filterwarnings('ignore', category=DeprecationWarning)
     def reset_deprecations():
         warnings.filterwarnings('default', category=DeprecationWarning)
+        #warnings.resetwarnings()  # XXX probably safer
     def raise_deprecations():
         try:
             registry = ClientForm.__warningregistry__
@@ -324,7 +325,9 @@
             url,
             )
 
+        hide_deprecations()
         forms = ClientForm.ParseResponse(r)
+        reset_deprecations()
         self.assertEqual(len(forms), 1)
         form = forms[0]
         self.assertEqual(form.action, url+"abc")
@@ -750,6 +753,7 @@
         forms = ClientForm.ParseFile(
             StringIO("<form><textarea>\n\nblah\n</textarea></form>"),
             "http://example.com/",
+            backwards_compat=False,
             )
         ctl = forms[0].find_control(type="textarea")
         self.assertEqual(ctl.value, "\r\nblah\r\n")
@@ -810,11 +814,10 @@
 
         def get_control(name, file=file, compat=compat):
             file.seek(0)
-            hide_deprecations()
             forms = ClientForm.ParseFile(file, "http://localhost/",
-                                         backwards_compat=compat)
-            reset_deprecations()
+                                         backwards_compat=False)
             form = forms[0]
+            form.backwards_compat = compat
             return form.find_control(name)
 
         # can't call item_disabled with no args
@@ -2164,10 +2167,8 @@
     def test_deselect_disabled(self):
         def get_new_form(f, compat):
             f.seek(0)
-            hide_deprecations()
             form = ClientForm.ParseFile(f, "http://example.com/",
                                         backwards_compat=False)[0]
-            reset_deprecations()
             form.backwards_compat = compat
             return form
 
@@ -3075,8 +3076,10 @@
     </select>
 </form>
 """)
+        if compat: hide_deprecations()
         form = ClientForm.ParseFile(f, "http://example.com/",
                                     backwards_compat=compat)[0]
+        if compat: reset_deprecations()
         ctl = form.find_control("form.grocery")
         # ordinary case
         self.assertEqual(ctl.get("p", nr=1).id, "3")

From jjlee at codespeak.net  Fri Mar 30 21:27:29 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Fri, 30 Mar 2007 21:27:29 +0200 (CEST)
Subject: [wwwsearch-commits] r41734 - wwwsearch/mechanize/trunk
Message-ID: <20070330192729.7DB901007A@code0.codespeak.net>

Author: jjlee
Date: Fri Mar 30 21:27:25 2007
New Revision: 41734

Modified:
   wwwsearch/mechanize/trunk/doc.html.in
Log:
Fix HTML nits

Modified: wwwsearch/mechanize/trunk/doc.html.in
==============================================================================
--- wwwsearch/mechanize/trunk/doc.html.in	(original)
+++ wwwsearch/mechanize/trunk/doc.html.in	Fri Mar 30 21:27:25 2007
@@ -347,7 +347,9 @@
 single chain of HTTP requests (so, for example, if you use a single
 HTTPRefererProcessor to fetch a series of URLs extracted from a single page,
 <strong>this will break</strong>).  <a
-href="../mechanize/">mechanize.Browser</a> does this properly.</p></dd>
+href="../mechanize/">mechanize.Browser</a> does this properly.</p>
+
+</dl>
 
 @{colorize(r"""
 import mechanize
@@ -360,11 +362,10 @@
 opener.open("http://www.rhubarb.com/")
 """)}
 
-</dl>
 
 
-<a name="seekable">
-<h2>Seekable responses</h2></a>
+<a name="seekable"></a>
+<h2>Seekable responses</h2>
 
 <p>Response objects returned from (or raised as exceptions by)
 <code>mechanize.SeekableResponseOpener</code>, <code>mechanize.UserAgent</code>
@@ -647,7 +648,7 @@
 original request and final response.  mechanize's responses can be
 made to have <code>.seek()</code> and <code>.get_data()</code>
 methods.  It's often useful to use the <code>.get_data()</code> method
-during debugging (see <a href="./doc.html#seekable">).
+during debugging (see <a href="./doc.html#seekable">above</a>).
 
 <p>Also, note <code>HTTPRedirectDebugProcessor</code> (which prints information
 about redirections) and <code>HTTPResponseDebugProcessor</code> (which prints
@@ -873,11 +874,12 @@
   <li>How do I download only part of a response body?
   <p>Just call <code>.read()</code> or <code>.readline()</code> methods on your
      response object as many times as you need.  The <code>.seek()</code>
-     method (which is not always present, see <a ref="./doc.html#seekable">)
-     still works, because mechanize caches read data.
+     method (which is not always present, see <a
+     href="./doc.html#seekable">above</a>) still works, because mechanize
+     caches read data.
   <li>What's the difference between the <code>.load()</code> and
       <code>.revert()</code> methods of <code>CookieJar</code>?
-  <p><code>.load()</code> <emph>appends</emph> cookies from a file.
+  <p><code>.load()</code> <em>appends</em> cookies from a file.
      <code>.revert()</code> discards all existing cookies held by the
      <code>CookieJar</code> first (but it won't lose any existing cookies if
      the loading fails).
@@ -895,7 +897,7 @@
      contains an explanation.  Processors are now a standard part of urllib2
      in Python 2.4.
   <li>How do I use it without urllib2.py?
-  <p>@{colorize(r"""
+  @{colorize(r"""
 from mechanize import CookieJar
 print CookieJar.extract_cookies.__doc__
 print CookieJar.add_cookie_header.__doc__

From jjlee at codespeak.net  Sat Mar 31 01:16:43 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 31 Mar 2007 01:16:43 +0200 (CEST)
Subject: [wwwsearch-commits] r41741 - in wwwsearch/mechanize/trunk: .
	mechanize test
Message-ID: <20070330231643.DD3D71007D@code0.codespeak.net>

Author: jjlee
Date: Sat Mar 31 01:16:32 2007
New Revision: 41741

Modified:
   wwwsearch/mechanize/trunk/doc.html.in
   wwwsearch/mechanize/trunk/mechanize/_auth.py
   wwwsearch/mechanize/trunk/mechanize/_http.py
   wwwsearch/mechanize/trunk/test/test_browser.doctest
Log:
Sub-requests should not usually be visiting, so make it so.  In fact the visible behaviour wasn't really broken here, since .back() skips over None responses (which is odd in itself, but won't be changed until after stable release is out).  However, this patch does change visible behaviour in that it creates new Request objects for sub-requests (e.g. basic auth retries) where previously we just mutated the existing Request object.

Modified: wwwsearch/mechanize/trunk/doc.html.in
==============================================================================
--- wwwsearch/mechanize/trunk/doc.html.in	(original)
+++ wwwsearch/mechanize/trunk/doc.html.in	Sat Mar 31 01:16:32 2007
@@ -444,6 +444,10 @@
 handlers unless you use <code>mechanize.Request</code> in the first place.
 Sorry about that.
 
+<p>Note also that handlers may create new <code>Request</code> instances (for
+example when performing redirects) rather than adding headers to existing
+<code>Request objects</code>.
+
 
 <a name="headers"></a>
 <h2>Adding headers</h2>

Modified: wwwsearch/mechanize/trunk/mechanize/_auth.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_auth.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_auth.py	Sat Mar 31 01:16:32 2007
@@ -11,7 +11,7 @@
 
 """
 
-import re, base64, urlparse, posixpath, md5, sha, sys
+import re, base64, urlparse, posixpath, md5, sha, sys, copy
 
 from urllib2 import BaseHandler
 from urllib import getproxies, unquote, splittype, splituser, splitpasswd, \
@@ -234,8 +234,10 @@
             auth = 'Basic %s' % base64.encodestring(raw).strip()
             if req.headers.get(self.auth_header, None) == auth:
                 return None
-            req.add_header(self.auth_header, auth)
-            return self.parent.open(req)
+            newreq = copy.copy(req)
+            newreq.add_header(self.auth_header, auth)
+            newreq.visit = False
+            return self.parent.open(newreq)
         else:
             return None
 
@@ -325,9 +327,10 @@
             auth_val = 'Digest %s' % auth
             if req.headers.get(self.auth_header, None) == auth_val:
                 return None
-            req.add_unredirected_header(self.auth_header, auth_val)
-            resp = self.parent.open(req)
-            return resp
+            newreq = copy.copy(req)
+            newreq.add_unredirected_header(self.auth_header, auth_val)
+            newreq.visit = False
+            return self.parent.open(newreq)
 
     def get_cnonce(self, nonce):
         # The cnonce-value is an opaque

Modified: wwwsearch/mechanize/trunk/mechanize/_http.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_http.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_http.py	Sat Mar 31 01:16:32 2007
@@ -77,15 +77,13 @@
             # from the user (of urllib2, in this case).  In practice,
             # essentially all clients do redirect in this case, so we do
             # the same.
-            try:
-                visit = req.visit
-            except AttributeError:
-                visit = None
+            # XXX really refresh redirections should be visiting; tricky to
+            #  fix, so this will wait until post-stable release
             return Request(newurl,
                            headers=req.headers,
                            origin_req_host=req.get_origin_req_host(),
                            unverifiable=True,
-                           visit=visit,
+                           visit=False,
                            )
         else:
             raise HTTPError(req.get_full_url(), code, msg, headers, fp)
@@ -446,7 +444,7 @@
     HTTPRefererProcessor to fetch a series of URLs extracted from a single
     page, this will break).
 
-    There's a proper implementation of this in module mechanize.
+    There's a proper implementation of this in mechanize.Browser.
 
     """
     def __init__(self):

Modified: wwwsearch/mechanize/trunk/test/test_browser.doctest
==============================================================================
--- wwwsearch/mechanize/trunk/test/test_browser.doctest	(original)
+++ wwwsearch/mechanize/trunk/test/test_browser.doctest	Sat Mar 31 01:16:32 2007
@@ -145,7 +145,7 @@
 >>> req = Request("http://example.com")
 >>> req.visit = False
 >>> br = TestBrowser2()
->>> hh = MockHTTPHandler(301, "Location: http://example.com/\r\n\r\n")
+>>> hh = MockHTTPHandler(302, "Location: http://example.com/\r\n\r\n")
 >>> br.add_handler(hh)
 >>> br.add_handler(HTTPRedirectHandler())
 >>> def raises(exc_class, fn, *args, **kwds):
@@ -166,6 +166,70 @@
 True
 
 
+...in fact, any redirection (but not refresh), proxy request, basic or
+digest auth request, or robots.txt request should be non-visiting,
+even if .visit is True:
+
+>>> from test_urllib2 import MockPasswordManager
+>>> def test_one_visit(handlers):
+...     br = TestBrowser2()
+...     for handler in handlers: br.add_handler(handler)
+...     req = Request("http://example.com")
+...     req.visit = True
+...     br.open(req)
+...     return br
+>>> def test_state(br):
+...     # XXX the _history._history check is needed because of the weird
+...     # throwing-away of history entries by .back() where response is
+...     # None, which makes the .back() check insufficient to tell if a
+...     # history entry was .add()ed.  I don't want to change this until
+...     # post-stable.
+...     return (
+...         br.response() and
+...         br.request and
+...         len(br._history._history) == 0 and
+...         raises(BrowserStateError, br.back))
+
+>>> hh = MockHTTPHandler(302, "Location: http://example.com/\r\n\r\n")
+>>> br = test_one_visit([hh, HTTPRedirectHandler()])
+>>> test_state(br)
+True
+
+>>> class MockPasswordManager:
+...     def add_password(self, realm, uri, user, password): pass
+...     def find_user_password(self, realm, authuri): return '', ''
+
+>>> ah = mechanize.HTTPBasicAuthHandler(MockPasswordManager())
+>>> hh = MockHTTPHandler(
+...     401, 'WWW-Authenticate: Basic realm="realm"\r\n\r\n')
+>>> test_state(test_one_visit([hh, ah]))
+True
+
+>>> ph = mechanize.ProxyHandler(dict(http="proxy.example.com:3128"))
+>>> ah = mechanize.ProxyBasicAuthHandler(MockPasswordManager())
+>>> hh = MockHTTPHandler(
+...     407, 'Proxy-Authenticate: Basic realm="realm"\r\n\r\n')
+>>> test_state(test_one_visit([ph, hh, ah]))
+True
+
+XXX Can't really fix this one properly without significant changes --
+the refresh should go onto the history *after* the call, but currently
+all redirects, including refreshes, are done by recursive .open()
+calls, which gets the history wrong in this case.  Will have to wait
+until after stable release:
+
+#>>> hh = MockHTTPHandler(
+#...     "refresh", 'Location: http://example.com/\r\n\r\n')
+#>>> br = test_one_visit([hh, HTTPRedirectHandler()])
+#>>> br.response() is not None
+#True
+#>>> br.request is not None
+#True
+#>>> r = br.back()
+
+XXX digest, robots
+
+
 .global_form() is separate from the other forms (partly for backwards-
 compatibility reasons).
 

From jjlee at codespeak.net  Sat Mar 31 01:29:46 2007
From: jjlee at codespeak.net (jjlee at codespeak.net)
Date: Sat, 31 Mar 2007 01:29:46 +0200 (CEST)
Subject: [wwwsearch-commits] r41742 - wwwsearch/mechanize/trunk/mechanize
Message-ID: <20070330232946.C32F21007D@code0.codespeak.net>

Author: jjlee
Date: Sat Mar 31 01:29:28 2007
New Revision: 41742

Modified:
   wwwsearch/mechanize/trunk/mechanize/_http.py
Log:
Note poor choice of exception

Modified: wwwsearch/mechanize/trunk/mechanize/_http.py
==============================================================================
--- wwwsearch/mechanize/trunk/mechanize/_http.py	(original)
+++ wwwsearch/mechanize/trunk/mechanize/_http.py	Sat Mar 31 01:29:28 2007
@@ -427,6 +427,7 @@
             if self.rfp.can_fetch(ua, request.get_full_url()):
                 return request
             else:
+                # XXX This should really have raised URLError.  Too late now...
                 msg = "request disallowed by robots.txt"
                 raise RobotExclusionError(
                     request,