Sorry to by-pass the correct procedure, but I really can't get past
the log-in procedure.
The whole family of Royal Mail sites, as typified by
all fail to parse. This is almost certainly a fault with the sites
rather than a failure of NetSurf: No RISC OS browser can open them.
> Date: Sun, 21 Jan 2007 21:51:13 +0000 (GMT)
> From: David H Wild <dhwild(a)talktalk.net>
> Subject: [Netsurf-develop] Netsurf failure with Yahoo site
> Netsurf Test Build (18 Jan 2007 01:30)
> I had an odd experience yesterday. Each evening I look at several web
> sites from the history file and one of these is
> I got access to the initial page but, when I clicked on any of the
> links (3 times altogether) Netsurf came up with an error and asked me
> to file a bug report. I didn't have time to do all that yesterday
> evening but used an older version to look at the site with no
> This morning, I tried to reproduce the problem, but everything worked
I have only joined this list today, although I have been using NetSurf
for some time.
I also downloaded "Netsurf Test Build (18 Jan 2007 01:30)" this
morning and I have since had some problems with two websites. In each
case the problem occurs when I use the top-right-hand icon to maximise
the window. In one case a long page is then displayed without any
scroll bars and appears locked until I resize it with the
bottom-right-hand icon, then the scroll bars appear. In the other
case the main frame of three frames appears blank until I refresh the
page or resize it, then the whole page is displayed correctly. This
latter case was fine before and did not exhibit this problem.
David J Worden
Netsurf Test Build (18 Jan 2007 01:30)
I had an odd experience yesterday. Each evening I look at several web sites
from the history file and one of these is
I got access to the initial page but, when I clicked on any of the links (3
times altogether) Netsurf came up with an error and asked me to file a bug
report. I didn't have time to do all that yesterday evening but used an
older version to look at the site with no problems.
This morning, I tried to reproduce the problem, but everything worked
David Wild using RISC OS on broadband
I've just tried looking at someone's Picassa web album in Netsurf, but was
unable to click on any of the images. Looking at the source, it turns out
that all the '/'s in the URLs have been changed to '\x2F's.
Is there any way that Netsurf could be made to decode these URLs?
|\ /| ark Fraser /mfraz2002(a)hotmail.com
| \/ | Somerset /www.mfraz.freeserve.co.uk
| |__________/Acorn SA RISC PC You know what the sig means!
Thanks to John-Mark Bell's response to my earlier request, my version of
nsgtk manages to download html and style sheets from the 'net.
I'm working with the default tarball of trunk/netsurf, download from the SVN
web interface on the 17th (and 12th and 15th).
As far as I can see, my problem now is that layout_document (and hence
html_reformat) requires that the type field of context->data.html.layout is
BOX_BLOCK, BOX_INLINE_BLOCK or BOX_TABLE_CELL, but the call to xml_to_box in
html_convert results in a type of BOX_INLINE_CONTAINER (using
www.netsurf-browser.org as the test page).
This is so fundamental, it can't be a bug, so it must be a misunderstanding
on my part.
Does this say anything to anyone?
Can anyone say if the drobe gallery pages (http://www.drobe.co.uk/gallery/)
appearing ridiculously wide, is down to NetSurf or drobe. If its the former
I'll raise a bug report.
David J. Ruck Phone: +44- (0)7974 108301 Email: druck(a)druck.org.uk
IS there any possibility in the near future of solving the bug which always in my experience crashes Netsurf? If you forward delete with "End" a few times in a form then the programme just quits.
This mail sent through http://www.ukonline.net
Paul Vigay wrote:
> I notice that in the URL bar in NetSurf there's a little,
> mini HTML file icon, so I was wondering if there are any
> plans to add favicon.ico support to NetSurf....? :-)
Yes, this was the intention. The URL bar (and autocomplete) currently shows
the small_xxx icon of the content-type being displayed (similar to the
I added support for decoding .ico files quite a long time ago as it is
basically just an extension of BMP, so all that really needs to be done is
to add support for the meta tag, add support for retrieving/saving them via
urldb (in a similar fashion to thumbnails), and add support for fetching
<url>/favicon.ico (or whatever the file is, I forget.)
Unfortunately, the amount of time I have at the moment and for the
forseable future is hugely limited. This means that unless someone else
wants to jump in and implement it (I can hapily help out and provide
complete details of what needs doing) then it's not going to be done any
This is my first post to this list, apologies if it's the wrong place.
I've been working on a RISC OS like GUI for Linux using the frame buffer
interface and, since it became moderately stable, writing a GDK layer for
GTK+. (An old screenshot can be seen at stoppers.drobe.co.uk.) Now the
example programs seem to display reasonably well, I've been using NetSurf
(nsgtk) as a test program.
The menus, buttons, etc. seem to work alright, but nothing gets displayed in
the main window. I looked at the NetSurf code and noticed a problem. With
curl-7.16.0, the routine curl_multi_add_handle returns with the error code
CURLM_BAD_EASY_HANDLE, because the handle returned from curl_easy_duphandle
in fetch_get_handle doesn't have its "magic" field filled in (done by
Curl_open). This means that no data (either from a local file or http
request) is returned to NetSurf (the throbber just keeps on going until I
I don't believe this is a problem with my version of the GDK, since they are
unrelated libraries, so I'd like to know if I should be using a different
version of libcurl, or reporting a bug to them.
(In the mean time I'll try removing the call to duphandle.)