Re: downloading older development versions
On Mon, Apr 22, 2013 at 01:12:08PM +, Tony Moore wrote: > Between NetSurf #417 (9 Oct 2012) and #461 (13 Oct 2012), it became > impossilbe to log-in to Yahoo webmail because, apparently, redirection > failed. I reported the problem > https://sourceforge.net/tracker/?func=detail&aid=3579166&group_id=51719&atid=464312 > but it persists, as of today, in #1083. > > I would like to determine the version in which the problem first occurs > but I cannot find any way of downloading versions prior to those listed > at http://ci.netsurf-browser.org/builds/riscos/ > > Could someone please suggest how this may be done? It cannot, the CI service only keeps the last 25 builds However: build #417 was at commit 38a9c374a4685c6b06625e63c3e305101e531d81 build #461 was at commit 61a2b8ddc452b915925793f4f538dbb2ca9b5ba5 and git log 38a9c374a4685c6b06625e63c3e305101e531d81..61a2b8ddc452b915925793f4f538dbb2ca9b5ba5 shows all changes between them I have added teh list to the bug > > Tony > > > > > -- Regards Vincent http://www.kyllikki.org/
Re: downloading older development versions
On 23 Apr 2013, Vincent Sanders wrote: > On Mon, Apr 22, 2013 at 01:12:08PM +, Tony Moore wrote: [snip] > > I would like to determine the version in which the problem first > > occurs but I cannot find any way of downloading versions prior to > > those listed at http://ci.netsurf-browser.org/builds/riscos/ > > > > Could someone please suggest how this may be done? > > It cannot, the CI service only keeps the last 25 builds > > However: [snip] > git log > 38a9c374a4685c6b06625e63c3e305101e531d81..61a2b8ddc452b915925793f4f538dbb2ca9b5ba5 > > shows all changes between them > > I have added teh list to the bug Many thanks for the response, and for the log. Unfortunately, I'm none the wiser since none of the log descriptions suggests - to me, a user - that it might be responsible for the problem. I wonder if anyone else, who uses Yahoo Mail, can replicate this: On the log-in screen, enter ID and Password, click 'Sign In' This leads to a screen which says If you are seeing this page, your browser settings prevent you from automatically redirecting to a new URL. Please click here to continue. Notwithstanding the warning, older versions of NetSurf _do_ redirect, and display the webmail page, as expected. The current version (#1083 json, or jsoff) displays a 'Warning from NetSurf' saying 'BadType', and no redirection. For me this is a nuisance, rather than a show-stopper, since I normally use pop/smtp, but I need to remember to load #417, to access the Spam folder, etc. Tony
Failed rendering of background image
The changes to NetSurf making it no longer abort the redraw if one of the images within the rectangle fails to render have brought a significant improvement in usability, but they don't seem to cover all cases. In particular, the old symptoms of white trails left by windows dragged across the page in a certain direction are very visible on this page: http://raphazlab.wordpress.com/ The offending image is the banner, which for some reason is inserted into the CSS like so: background: url('http://raphazlab.files.wordpress.com/2012/09/banner2012.jpg') no-repeat; This appears to cause NetSurf to fail to implement its workaround, and fall back on the old behaviour of aborting redraw. Not really sure if that counts as a bug or not. It may well be intentional. What would a suitable alternative behaviour for background images be? -- __<^>__ / _ _ \ You always find something in the last place you look. ( ( |_| ) ) \_> <_/ === Martin Bazley ==