From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp1.ugent.be ([157.193.71.182]) by ewsd; Fri Mar 20 13:40:09 EDT 2020 Received: from localhost (mcheck3.ugent.be [157.193.71.89]) by smtp1.ugent.be (Postfix) with ESMTP id 660C4A0777; Fri, 20 Mar 2020 18:40:07 +0100 (CET) X-Virus-Scanned: by UGent DICT Received: from smtp1.ugent.be ([157.193.71.182]) by localhost (mcheck3.ugent.be [157.193.43.11]) (amavisd-new, port 10024) with ESMTP id o25vHYi2aVKH; Fri, 20 Mar 2020 18:40:07 +0100 (CET) Received: from smtp.irc.ugent.be (cp0242.irc.ugent.be [157.193.189.5]) by smtp1.ugent.be (Postfix) with ESMTP id 67B1AA0756; Fri, 20 Mar 2020 18:40:06 +0100 (CET) Received: from localhost (localhost [127.0.0.1]) by smtp.irc.ugent.be (Postfix) with ESMTP id B74E6113105; Fri, 20 Mar 2020 18:41:19 +0100 (CET) X-Virus-Scanned: amavisd-new at irc.ugent.be Received: from smtp.irc.ugent.be ([127.0.0.1]) by localhost (smtp.irc.ugent.be [127.0.0.1]) (amavisd-new, port 10024) with LMTP id Rx7wQ_s2MYbZ; Fri, 20 Mar 2020 18:41:19 +0100 (CET) Received: from Hugin (178-117-59-139.access.telenet.be [178.117.59.139]) by smtp.irc.ugent.be (Postfix) with ESMTPSA id 63066113115; Fri, 20 Mar 2020 18:41:17 +0100 (CET) Date: Fri, 20 Mar 2020 18:40:02 +0100 From: Jens Staal To: 9front@9front.org Cc: ori@eigenstate.org, knusbaum@sdf.org, kokamoto@hera.eonet.ne.jp Subject: Re: [9front] adding javascript enable to netsurf Message-ID: <20200320174002.6bp7uhzhx6pa4wck@Hugin> References: <2E9D0BC175C6CB97ED70074A091C37A7@eigenstate.org> <77D9629A24752EF2B79C31DC37C2C76C@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <77D9629A24752EF2B79C31DC37C2C76C@gmail.com> X-Miltered: at jchkm3 with ID 5E74FFF6.002 by Joe's j-chkmail (http://helpdesk.ugent.be/email/)! X-j-chkmail-Enveloppe: 5E74FFF6.002 from cp0242.irc.ugent.be/cp0242.irc.ugent.be/157.193.189.5/smtp.irc.ugent.be/ X-j-chkmail-Score: MSGID : 5E74FFF6.002 on smtp1.ugent.be : j-chkmail score : . : R=. U=. O=. B=0.000 -> S=0.000 X-j-chkmail-Status: Ham List-ID: <9front.9front.org> List-Help: X-Glyph: ➈ X-Bullshit: patented shared replication rich-client component On Fri, Mar 20, 2020 at 06:12:41PM +0100, telephil9@gmail.com wrote: > On Fri Mar 20 17:16:38 CET 2020, ori@eigenstate.org wrote: > > > I think the error is due to a limitation of the webfs fetcher. Currently, the fetcher hangs on > > > to webfs handles until they're freed by netsurf, even after the data is > > > read, because after they're closed things like headers are unavailable. > > > This means tha webfs handles can be exhausted, at which point netsurf > > > starts to run into that error. > > > > > > This can definitely be improved in the webfs fetcher, but maybe the > > > relatively low webfs handle limit could also be raised. > > > > > > > Are they ever released, or are they effectively leaked? > > > > Also, searching on Google just.. doesn't work. It seems like we get > > a bogus URL. > > > > I tracked this one down to iconv not being implemented (see posix/src/iconv.c). > For the bogus URL, the reason is that when the form is submitted, its fields are encoded (form_encode_item() in netsurf/content/handlers/html/form.c) using iconv which returns null each time (hence the many =&=&=&=). > Proper fix would be to implement the iconv() function, easy cheat is to return the unmodified item in the form_encode_item() function. > I remember ages ago when I built various netsurf libraries for Plan9, I used the Apache (apr) version of iconv. I don't remember how I came to the conclusion that I needed it.