Patches item #2893965, was opened at 2009-11-07 22:34
Message generated for change (Settings changed) made by sf-robot
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=311118&aid=2893965&group_id=11118
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Closed
Resolution: None
Priority: 5
Private: No
Submitted By: Lee (ler762)
Assigned to: Lee (ler762)
Summary: work-around for firefox hanging on blocked javascript pages
Initial Comment:
see https://bugzilla.mozilla.org/show_bug.cgi?id=492459 for problem description of firefox hanging sometimes when Privoxy blocks a page
see actionsfile feedback item 2790091 for discussion of how to work around the FF problem
Adds a new
handle-as-empty-doc-returns-ok
runtime configuration variable that, when set, tells Privoxy to return a "200 OK" status for pages blocked with +handle-as-empty-document instead of
"403 Request blocked by Privoxy"
----------------------------------------------------------------------
>Comment By: SourceForge Robot (sf-robot)
Date: 2009-11-24 02:20
Message:
This Tracker item was closed automatically by the system. It was
previously set to a Pending status, and the original submitter
did not respond within 14 days (the time period specified by
the administrator of this Tracker).
----------------------------------------------------------------------
Comment By: Fabian Keil (fabiankeil)
Date: 2009-11-14 18:23
Message:
Turns out Docbook requires:
</listitem>
+ </varlistentry>
+ <varlistentry>
<term>Effect if set:</term>
<listitem>
but other than that the patch builds.
----------------------------------------------------------------------
Comment By: Fabian Keil (fabiankeil)
Date: 2009-11-09 18:54
Message:
I think the URL to the bug report should be a link.
The title of the Firefox bug report should probably be in <quote> tags.
"+handle-as-empty-document" should probably be either a link or in
<literal> tags. Looks like p-config.sgml isn't consistent there.
Other than that the patch looks fine to me and if necessary, we can always
improve it after it has been committed.
I most likely won't have access to my GNU/Linux test system (where the
Docbook mess is more or less working) until Friday, but if there are any
syntax issues, I can deal with them then and this shouldn't hold you back.
If you feel strongly about this, feel free to enable it by default. While
I still think it's wrong, I agree with you that it shouldn't cause any
actual technical problems and Privoxy isn't advertised to be entirely RFC
compliant anyway.
----------------------------------------------------------------------
Comment By: Lee (ler762)
Date: 2009-11-08 20:42
Message:
Fabian,
The proposed patch for doc/source/p-config.sgml attached. I never got the
docbook stuff working, so I can't tell if the patch is good or not :(
I was talking to a friend about this & they asked if that might be the
reason they couldn't log in to the IEEE web site using Privoxy! I'd really
like this option to be enabled by default.
Thanks,
Lee
----------------------------------------------------------------------
Comment By: Fabian Keil (fabiankeil)
Date: 2009-11-08 19:37
Message:
Looks great.
If the flags are sorted by bit order, I can live with that, too.
I'd prefer the option to be off by default, but it's not that important to
me.
It needs to be documented, no matter if it's on or off by default,
though.
----------------------------------------------------------------------
Comment By: Lee (ler762)
Date: 2009-11-08 18:39
Message:
Please take a look at the -v2 diff.
> think the rsp->status line should be either "200 OK. Request blocked by
> Privoxy" or "200 Request blocked by Privoxy". Firefox should be happy
> either way, but you may want to verify this.
"200 Request blocked by Privoxy". works just fine.
> The RUNTIME_FEATURE_* flags above the feature_flags member should stay
> sorted alphabetically.
They're also in bit order & it seemed like it would be better to have them
in 1,2,4,8,16,... order than alphabetically. But if you really want them
sorted by name let me know..
> The comments should use capitalization where appropriate.
as you can see, I'm not always all that good at "appropriate" :)
> The feature needs to be documented in doc/source/p-config.sgml.
Right. That's one of the reasons I didn't want a run-time flag. Even if
I document this change so users can understand it, how many people read the
docs? Vs. how many people expect things to just work "out-of-the-box"?
I haven't had any Privoxy problems with IE or Seamonkey .. as far as I can
tell, this change only affects Firefox (for the better :). Are you sure
that you want it to default to off?
Thanks
Lee
----------------------------------------------------------------------
Comment By: Fabian Keil (fabiankeil)
Date: 2009-11-08 13:36
Message:
Looks good, there are a couple of minor issues though:
In block_url(), the comment text in the if block should start below the
"/*" line and the else block should use curly braces.
I think the rsp->status line should be either "200 OK. Request blocked by
Privoxy" or "200 Request blocked by Privoxy". Firefox should be happy
either way, but you may want to verify this.
In load_config(), there's superfluous white space on the line after the
break for the handle_as_empty_returns_ok case.
In project.h, the RUNTIME_FEATURE_EMPTY_DOC_RETURNS_OK comment is
inconsistent with the comments above, as it doesn't start with
"configuration_spec::feature_flags".
The RUNTIME_FEATURE_* flags above the feature_flags member should stay
sorted alphabetically.
The currently missing ones you spotted, that are unrelated to
RUNTIME_FEATURE_EMPTY_DOC_RETURNS_OK should be committed separately, so the
RUNTIME_FEATURE_EMPTY_DOC_RETURNS_OK commit can be easily reverted after
Firefox has been fixed for a while. (Assuming the Firefox developers
eventually take a look at the bug report and additionally agree that it
actually is a bug.)
The comments should use capitalization where appropriate.
The feature needs to be documented in doc/source/p-config.sgml.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=311118&aid=2893965&group_id=11118
|