Re: [-SPAM-] Zap and the Raspberry Pi

Steve – you might be waiting a long time to get responses from everyone. Zap isn't a terribly alive project these days :-(

You don't have to include sources, but if you make modifications then you need to make them available somewhere. If it's just stock Zap, all you'd have to do is link back to our website.

James

On 30 Sep 2012, at 01:14, Steve (ROOL) <srevill@riscosopen.org> wrote:

> On 28 Aug, James Aylett <james@tartarus.org> wrote:
>> I think we're generally happy with this kind of thing. Darren / Christian,
>> any objections?
>
> Bump! :)
>
>> (Has stock Zap been confirmed to run on Raspberry Pi? I know we've had
>> various patches and re-distros floating around for some other RO targets
>> over the last few years.)
>
> It should work OK because we're running the RPi in ARMv5 compatible mode,
> which means it's broadly the same as an Iyonix from Zap's PoV.
>
> Do we need to include sources in the distro or are we OK to just have the
> app and it's bits (ZapChoices, ZapFonts)?
>
> Ta,
>
> Steve
>
> --
> Stephen Revill, Director www.riscosopen.org
> RISC OS Open Ltd
> Sovereign House, 22 Shelley Road,
> Worthing, West Sussex, BN11 1TU

--
James Aylett
Writing: nsfwcorp.com; film: talktorex.co.uk
Also: devfort.com, spacelog.org, wildlifenearyou.com


--
_______________________________________________
zap-technical maillist - zap-technical@zap.tartarus.org
http://lists.tartarus.org/mailman/listinfo/zap-technical

Re: Crashes on GMail

Still crash on OS X using m.gmail.com

Re: Crashes on GMail

In article
<CAGVx7UU2+6_vxJ9+Uk=N-Cij_66xUbbXCiemNsDczC6hxeETPQ@mail.gmail.com>, Ma
Xiaojun <damage3025@gmail.com> wrote:
> I'm using Mac OS X 10.7.5, NetSurf 2.9.
> http://www.netsurf-browser.org/downloads/macosx/

> It seems to be running smooth on a bunch sites.

> I'd like to use GMail, I don't mind it is just mobile view only.
> However, all the following addresses gave me a crash:

> www.gmail.com mail.google.com g.illinois.edu

> The last one is the Google Web App portal of University of Illinois.

http://m.gmail.com seems to go straight to my inbox in 'mobile mode'. The
adoption of the m prefix for mobile sites seems quite popular.
http://m.facebook.com http://m.twitter.com and even http://m.sarva.co.uk

--
Tim Hill
..............................................................
www.timil.com

Re: Crashes on GMail

Ma Xiaojun wrote

> I'm using Mac OS X 10.7.5, NetSurf 2.9.
.......

> I'd like to use GMail, I don't mind it is just mobile view only.
> However, all the following addresses gave me a crash:

> www.gmail.com
> mail.google.com
> g.illinois.edu


No problem on Iyonix 5.18 with 2.9

I link to:- https://mail.google.com/mail/?ui=1
This brings up a page:-
https://mail.google.com/mail/u/0/?ui=2
That says:-
JavaScript must be enabled in order for you to use Gmail in standard
view......
To use Gmail's basic HTML view, which does not require JavaScript,
click here...
https://mail.google.com/mail/u/0/h/11xp6d428d699/?zy=c&f=1

I click on the link and up comes the gmail inbox.


--
John - http://mug.riscos.org/

Re: User feedback requested (esp. RISC OS)

In message <406ffdd652.DaveMeUK@my.inbox.com>
Dave Higton <dave@davehigton.me.uk> wrote:

>In message <989ef9d652.DaveMeUK@my.inbox.com>
> Dave Higton <dave@davehigton.me.uk> wrote:
>
>> In message <52d6e2c396tlsa@netsurf-browser.org>
>> Michael Drake <tlsa@netsurf-browser.org> wrote:
>>
>> >
>> > Hi,
>> >
>> > We have done quite a bit of development since we last issued test builds
>> > and we have also changed our autobuilding infrastructure completely. With
>> > these changes we are now building NetSurf builds with far more up-to-date
>> > versions of 3rd party libraries, and it is now quite trivial to keep them
>> > up-to-date.
>> >
>> > It would help us if people could test the new builds and let us know how
>> > you get on.
>> >
>> >
>> > CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS
>> >
>> > We have at least one case of NetSurf freezing a
>> > RISC OS Iyonix such that it needs to be reset.
>>
>> Sadly, I've also seen it once with the earliest new build that I can see,
>> 2012-09-11_14-52-47, as well as 3 times with 2012-09-28_19-54-37. At least
>> it hasn't eaten the hard drive :-)
>>
>> There doesn't seem to be any special thing that causes it to stiff the
>> Iyonix.
>
>In response to another question: no fetching of news or mail was going
>on at the time. The only network activity unrelated to Netsurf would
>probably be ShareFS's normal broadcasting. NewsHound and AntiSpam were
>on the icon bar but merely quiescent.

Checked a few minutes ago with 2012-09-30_10-23-17 and again got a
freeze. This time the only app also on the icon bar was Zap. The
freeze occurred a few seconds after rendering a page from the BBC
News web site.

Iyonix, 5.18, 512 MiB.

Dave

____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth

Re: Zap and the Raspberry Pi

Please can everyone replying take the "Re: [-SPAM-] " prefix out of the
subject line, as it being matched by various spam filters.

---druck

--
Email: druck@druck.org.uk
Phone: +44-(0)7974 108301


--
_______________________________________________
zap-technical maillist - zap-technical@zap.tartarus.org
http://lists.tartarus.org/mailman/listinfo/zap-technical

Re: User feedback requested (esp. RISC OS)

In article <52d74f974dchrisg@care4free.net>,
Chris Gransden <chrisg@care4free.net> wrote:
> In article <52d74e4138tlsa@netsurf-browser.org>,
> Michael Drake <tlsa@netsurf-browser.org> wrote:

> > Interesting. What do you mean by the latest version? The latest
> > NetSurf from our Git repo?

> I'm using the latest version of NetSurf and libraries from the Git repo.

Sounds promising. What versions of 3rd party libraries are you using?

Please could you follow these steps:

1. Run your version of NetSurf.
2. Visit URL: about:testament
3. Save the page.
4. Quit NetSurf.
5. Double click !NetSurf.OpenScrap
6. Open the Log file in the directory that opens.
7. Copy the line that says something like:

(0.30000) content/fetchers/curl.c fetch_curl_register 167:
curl_version libcurl/7.20.1 OpenSSL/0.9.8n zlib/1.2.3.3 c-ares/1.6.0

It should be near the top, line 5.
8. Copy the saved testament and line from the Log file into an e-mail.

Cheers,

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: User feedback requested (esp. RISC OS)

In message <52d74d441achrisg@care4free.net>
Chris Gransden <chrisg@care4free.net> wrote:

> On a Pandaboard with RISC OS 5.19 using
> http://ci.netsurf-browser.org/builds/riscos/netsurf-2012-09-30_10-23-17.zip
> it's possible to trigger a freeze every time by loading any web page and
> pressing return several times in a !Nettle
> (http://www.riscos.info/packages/NetworkDetails.html#Nettle) window.
>
> I've been using the latest version of netsurf built with the autobulder
> (revision 5854 and gcc 4.6.4) locally for some time. I haven't had any
> lockups at all with this.

FYI, gcc 4.6.4 is a development snapshot so might contain problems producing
wrong binaries.

John.
--
John Tytgat
joty@netsurf-browser.org

Re: User feedback requested (esp. RISC OS)

In article <52d74e4138tlsa@netsurf-browser.org>,
Michael Drake <tlsa@netsurf-browser.org> wrote:

> Interesting. What do you mean by the latest version? The latest NetSurf
> from our Git repo?

I'm using the latest version of NetSurf and libraries from the Git repo.

Chris.

Re: User feedback requested (esp. RISC OS)

front end: RISC OS
OS version: 5.18
problem: freeze while scrolling through
http://www.independent.co.uk/
download file: netsurf-2012-09-28_19-54-37/zip

Note that POPstar was not running - so no clash with other
fetchers. So far freezes seem to happen after just
the slightest scrolling down the page; sometimes after
quite lengthy and varied browsing, sometimes, as in this
case, on the first page opened (so doubtful if any gc can
have taken place). This apparent randomness gives me the
hunch that it has something to do with buffers being
allocated where they should not be, but maybe such a comment
is unhelpful without any evidence.

Sometimes pressing the reset button works - sometimes I have
to toggle the on/off switch at the back.

--
Gavin Wraith (gavin@wra1th.plus.com)
Home page: http://www.wra1th.plus.com/

Re: User feedback requested (esp. RISC OS)

In article <52d74d441achrisg@care4free.net>,
Chris Gransden <chrisg@care4free.net> wrote:

> I've been using the latest version of netsurf built with the autobulder
> (revision 5854 and gcc 4.6.4) locally for some time. I haven't had any
> lockups at all with this.

Interesting. What do you mean by the latest version? The latest NetSurf
from our Git repo?

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: User feedback requested (esp. RISC OS)

In article <52d6e2c396tlsa@netsurf-browser.org>,
Michael Drake <tlsa@netsurf-browser.org> wrote:

> Hi,

> We have done quite a bit of development since we last issued test builds
> and we have also changed our autobuilding infrastructure completely. With
> these changes we are now building NetSurf builds with far more up-to-date
> versions of 3rd party libraries, and it is now quite trivial to keep them
> up-to-date.

> It would help us if people could test the new builds and let us know how
> you get on.


> CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS

> We have at least one case of NetSurf freezing a
> RISC OS Iyonix such that it needs to be reset.

On a Pandaboard with RISC OS 5.19 using
http://ci.netsurf-browser.org/builds/riscos/netsurf-2012-09-30_10-23-17.zip
it's possible to trigger a freeze every time by loading any web page and
pressing return several times in a !Nettle
(http://www.riscos.info/packages/NetworkDetails.html#Nettle) window.

I've been using the latest version of netsurf built with the autobulder
(revision 5854 and gcc 4.6.4) locally for some time. I haven't had any
lockups at all with this.

Chris.

--

Re: [-SPAM-] Re: Zap and the Raspberry Pi

On 28 Aug, James Aylett <james@tartarus.org> wrote:
> I think we're generally happy with this kind of thing. Darren / Christian,
> any objections?

Bump! :)

> (Has stock Zap been confirmed to run on Raspberry Pi? I know we've had
> various patches and re-distros floating around for some other RO targets
> over the last few years.)

It should work OK because we're running the RPi in ARMv5 compatible mode,
which means it's broadly the same as an Iyonix from Zap's PoV.

Do we need to include sources in the distro or are we OK to just have the
app and it's bits (ZapChoices, ZapFonts)?

Ta,

Steve

--
Stephen Revill, Director www.riscosopen.org
RISC OS Open Ltd
Sovereign House, 22 Shelley Road,
Worthing, West Sussex, BN11 1TU

--
_______________________________________________
zap-technical maillist - zap-technical@zap.tartarus.org
http://lists.tartarus.org/mailman/listinfo/zap-technical

Re: Crashes on GMail

I just tested NetSurf on Ubuntu 12.04.
I used the "netsurf" package found in system repository.
http://packages.ubuntu.com/netsurf
It's 2.8 and GMail works.
University of Illinois portal gives a undesired error message; it
should show a login.

Then I tried 2.8 on my Mac OS X 10.7.5, but nothing changed.

So I wonder there is some problem in the Mac OS X port.

Crashes on GMail

I'm using Mac OS X 10.7.5, NetSurf 2.9.
http://www.netsurf-browser.org/downloads/macosx/

It seems to be running smooth on a bunch sites.

I'd like to use GMail, I don't mind it is just mobile view only.
However, all the following addresses gave me a crash:

www.gmail.com
mail.google.com
g.illinois.edu

The last one is the Google Web App portal of University of Illinois.

Re: User feedback requested (esp. RISC OS)

On 29 Sep 2012 Michael Drake wrote:

> It would help us if people could test the new builds and let us know how
> you get on.

A heading with font size=6 comes out massively smaller than it did in
earlier versions.

I haven't had any lockups yet.

RO6.14
--
Richard Porter http://www.minijem.plus.com/
mailto:ricp@minijem.plus.com
I don't want a "user experience" - I just want stuff that works.

Re: User feedback requested (esp. RISC OS)

In article <52d6f5c6cdtlsa@netsurf-browser.org>,
Michael Drake <tlsa@netsurf-browser.org> wrote:
> > Yes, ditto here - twice. Both times with Messenger Pro 6.06 and
> > !NetFetch 3.65.
> > Iyonix, RO5.18.

> This freeze has been reported on RISC OS 5.18 and 5.19.

> Can anyone reproduce it with RISC OS 4 or RISC OS 6?

Well, I've been out all evening and only just got back in but I'll give it
a try and see out it goes - RO4.39

--
Stuart Winsor

Only plain text for emails
http://www.asciiribbon.org

Re: User feedback requested (esp. RISC OS)

In message <989ef9d652.DaveMeUK@my.inbox.com>
Dave Higton <dave@davehigton.me.uk> wrote:

> In message <52d6e2c396tlsa@netsurf-browser.org>
> Michael Drake <tlsa@netsurf-browser.org> wrote:
>
> >
> > Hi,
> >
> > We have done quite a bit of development since we last issued test builds
> > and we have also changed our autobuilding infrastructure completely. With
> > these changes we are now building NetSurf builds with far more up-to-date
> > versions of 3rd party libraries, and it is now quite trivial to keep them
> > up-to-date.
> >
> > It would help us if people could test the new builds and let us know how
> > you get on.
> >
> >
> > CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS
> >
> > We have at least one case of NetSurf freezing a
> > RISC OS Iyonix such that it needs to be reset.
>
> Sadly, I've also seen it once with the earliest new build that I can see,
> 2012-09-11_14-52-47, as well as 3 times with 2012-09-28_19-54-37. At least
> it hasn't eaten the hard drive :-)
>
> There doesn't seem to be any special thing that causes it to stiff the
> Iyonix.

In response to another question: no fetching of news or mail was going
on at the time. The only network activity unrelated to Netsurf would
probably be ShareFS's normal broadcasting. NewsHound and AntiSpam were
on the icon bar but merely quiescent.

Dave

____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth

Re: User feedback requested (esp. RISC OS)

The following bytes were arranged on 29 Sep 2012 by Michael Drake :

> It would help us if people could test the new builds and let us know how
> you get on.

...And now for a more mundane, and tiresomely repetitive regression...

http://www.andyfanton.com

"Serious error". Every time, without fail. Same computer and archive
as in my last report.

The log (same every time) says:

"render/html.c", line 1269: html_object_callback: Assertion failed: 0

2.9 works.

Instant crashes with this particular website have been reported on three
separate previous occasions:

http://sourceforge.net/tracker/index.php?func=detail&aid=3182729&group_id=51719&atid=464312

http://sourceforge.net/tracker/index.php?func=detail&aid=3238323&group_id=51719&atid=464312

http://sourceforge.net/tracker/index.php?func=detail&aid=3376796&group_id=51719&atid=464312

Each time, the bug was fixed, only for another version a few months
down the line to introduce a different bug which also caused NetSurf to
crash on sight of that site. It appears this cycle has now gone round
for a fourth time.

I've never had problems of this scale with any other site. Whatever
andyfanton.com does, NetSurf really doesn't like it!

In the interests of preventing this regression occurring a fifth time,
perhaps 'Jenkins' could be amended to make sure NetSurf can cope with a
fetch and display of that specific page? Stopping regressions what
automated testing is for, right?

--
__<^>__
/ _ _ \ You always find something in the last place you look.
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

Re: User feedback requested (esp. RISC OS)

In message <52d6e2c396tlsa@netsurf-browser.org>
Michael Drake <tlsa@netsurf-browser.org> wrote:

>
>Hi,
>
>We have done quite a bit of development since we last issued test builds
>and we have also changed our autobuilding infrastructure completely. With
>these changes we are now building NetSurf builds with far more up-to-date
>versions of 3rd party libraries, and it is now quite trivial to keep them
>up-to-date.
>
>It would help us if people could test the new builds and let us know how
>you get on.
>
>
> CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS
>
> We have at least one case of NetSurf freezing a
> RISC OS Iyonix such that it needs to be reset.

Sadly, I've also seen it once with the earliest new build that I can
see, 2012-09-11_14-52-47, as well as 3 times with 2012-09-28_19-54-37.
At least it hasn't eaten the hard drive :-)

There doesn't seem to be any special thing that causes it to stiff
the Iyonix.

I've reverted to r1571.

Iyonix, RO 5.18, 512 MiB.

Dave

____________________________________________________________
FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family!
Visit http://www.inbox.com/photosharing to find out more!

Re: User feedback requested (esp. RISC OS)

In message <52d6e2c396tlsa@netsurf-browser.org>
Michael Drake <tlsa@netsurf-browser.org> wrote:

>
>Hi,
>
>We have done quite a bit of development since we last issued test builds
>and we have also changed our autobuilding infrastructure completely. With
>these changes we are now building NetSurf builds with far more up-to-date
>versions of 3rd party libraries, and it is now quite trivial to keep them
>up-to-date.
>
>It would help us if people could test the new builds and let us know how
>you get on.
>
>
> CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS
>
> We have at least one case of NetSurf freezing a
> RISC OS Iyonix such that it needs to be reset.

I've seen the same three times now with 2012-09-28_19-54-37. First
occurrence was during loading of this version for the first time;
other occurrences have been during browsing. This Iyonix does not
normally freeze like that at all (mouse cursor froze, no response
to NumLock).

Iyonix, RO 5.18 (16-Jan-12), 512 MiB.

Dave

____________________________________________________________
FREE 3D MARINE AQUARIUM SCREENSAVER - Watch dolphins, sharks & orcas on your desktop!
Check it out at http://www.inbox.com/marineaquarium

Re: User feedback requested (esp. RISC OS)

In article <021eead652.SuperAl@albenito.eclipse2k.freeserve.co.uk>,

> Yes, ditto here - twice. Both times with Messenger Pro 6.06 and
> !NetFetch 3.65.
> Iyonix, RO5.18.

This freeze has been reported on RISC OS 5.18 and 5.19.

Can anyone reproduce it with RISC OS 4 or RISC OS 6?

Cheers,

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: User feedback requested (esp. RISC OS)

The following bytes were arranged on 29 Sep 2012 by Michael Drake :

> It would help us if people could test the new builds and let us know how
> you get on.

I've tried everything suggested in this thread to make this RO5.19
ARMini crash, including fetching email (though with AntiSpam 1.63, not
Hermes) and logging into the ROOL site. It stubbornly refused to crash
for about two hours, until suddenly the mouse pointer froze after
clicking on a Google Groups link (from a list of Google search results).

Given that I've also had problems with the previous development version
(from the old autobuilder) freezing the ARMini (not so much the Iyonix
or RiscPC), I don't think you can read too much into this.

> + which front end (e.g. RISC OS, Amiga, etc)

RISC OS.

> + which version of the OS

5.19 (16-May-12)

> + whether there was a problem

Infinite loop with interrupts disabled.

> > what was happening when it went wrong

Nothing in particular, especially not reported triggers like NewsHound.

> + what the download file name of the NetSurf version you tried was

2012-09-28_19-54-37

--
__<^>__ "Your pet, our passion." - Purina
/ _ _ \ "Your potential, our passion." - Microsoft, a few months later
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

Re: User feedback requested (esp. RISC OS)

Peter Young, on 29 Sep, wrote:

> On 29 Sep 2012 Peter Young <pnyoung@ormail.co.uk> wrote:
>
> > On 29 Sep 2012 Paul Sprangers <Paul@sprie.nl> wrote:
>
> > > Hi all,
>
> > > Version 3.0 (development) seems to work fine (and fast) on my RISC OS
> > > 5.18 computer. Keep up the good work (just add java script).
>
> > Also here, ARMini, RISC OS 5.19. Glad to see that test builds are
> > available again, many thanks.
>
> Might have spoken too soon. I sent this message using Hermes, and when
> NewsHound kicked in, the machine froze. The message evidently went, but is
> still in Messenger Pro's queue. Going back to NetSurf 2.9 for the time
> being, I regret.

NetSurf 3.0 (Development) has stiffed my ARMini twice now trying to get onto
ROOL's site, https://www.riscosopen.org/content/ and that had nothing to do
with mail or usenet fetching. Sometimes that site has been reachable, just
guessing is https perhaps an issue, other sites seem OK.

I cannot browse the web at all on VRPC on my Mac with 3.0, resolving times
out after 30000ms, 2.9 is fine.

I will see if I can drum up more info later but right now it is time for Dr
Who.

--
David Pitt

Re: User feedback requested (esp. RISC OS)

In message <52d6e2c396tlsa@netsurf-browser.org>
Michael Drake <tlsa@netsurf-browser.org> wrote:

>
> Hi,
>
> We have done quite a bit of development since we last issued test builds
> and we have also changed our autobuilding infrastructure completely. With
> these changes we are now building NetSurf builds with far more up-to-date
> versions of 3rd party libraries, and it is now quite trivial to keep them
> up-to-date.
>
> It would help us if people could test the new builds and let us know how
> you get on.
>
>
> CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS
>
> We have at least one case of NetSurf freezing a
> RISC OS Iyonix such that it needs to be reset.
>
>
> New builds are available from:
>
> http://ci.netsurf-browser.org/ (Use the latest one available)
>
>
> If you try one of the new builds, please let us know:
>
> + which front end (e.g. RISC OS, Amiga, etc)
> + which version of the OS
> + whether there was a problem
> > what was happening when it went wrong
> + what the download file name of the NetSurf version you tried was
>
> Cheers,
>
I'm using the 3.0 development build 2012-09-28_19-54-37 on RPCEmu
0.8.9/RISC OS 4.02 on a Windows 7 (64-bit) PC. Got a run of 'Base
stylesheet failed to load' errors, but quitting and reloading Netsurf
3.0 restored normal service. Seems fine so far, and nippier than 2.9.
Good work!

George

--
george greenfield

Re: User feedback requested (esp. RISC OS)

On 29 Sep 2012 Michael Drake <tlsa@netsurf-browser.org> wrote:

> In article <46b9e6d652.pnyoung@pnyoung.ormail.co.uk>,
> Peter Young <pnyoung@ormail.co.uk> wrote:

>> Might have spoken too soon. I sent this message using Hermes, and when
>> NewsHound kicked in, the machine froze.

> Thanks, can you say what versions of Hermes and NewsHound you have?

Sorry, should have said.

Hermes 2.65 (11 April, 2012)
NewsHound 1.42 ( Can't find the date, as it removes itself from the
icon bar too quickly!)

Both in NetFetch 3.65 (26 April, 2012)

With best wishes,

Peter.

--
Peter \ / zfc Ta \ Prestbury, Cheltenham, Glos. GL52
and \/ __ __ \ England.
family / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
/ \__/ \_/ | \| \__/ \______________ pnyoung@ormail.co.uk

Re: User feedback requested (esp. RISC OS)

In message <52d6e2c396tlsa@netsurf-browser.org>
Michael Drake <tlsa@netsurf-browser.org> wrote:


> Hi,

> We have done quite a bit of development since we last issued test builds
> and we have also changed our autobuilding infrastructure completely. With
> these changes we are now building NetSurf builds with far more up-to-date
> versions of 3rd party libraries, and it is now quite trivial to keep them
> up-to-date.

> It would help us if people could test the new builds and let us know how
> you get on.


> CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS

> We have at least one case of NetSurf freezing a
> RISC OS Iyonix such that it needs to be reset.

Yes, ditto here - twice. Both times with Messenger Pro 6.06 and
!NetFetch 3.65.

Iyonix, RO5.18.


--
Allan Bennett

Re: User feedback requested (esp. RISC OS)

In article <46b9e6d652.pnyoung@pnyoung.ormail.co.uk>,
Peter Young <pnyoung@ormail.co.uk> wrote:

> Might have spoken too soon. I sent this message using Hermes, and when
> NewsHound kicked in, the machine froze.

Thanks, can you say what versions of Hermes and NewsHound you have?

Cheers,

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: User feedback requested (esp. RISC OS)

In article <2999e6d652.wra1th@wra1th.plus.com>,
Gavin Wraith <gavin@wra1th.plus.com> wrote:

> front end: RISC OS
> OS version: 5.18
> problem: freeze while scrolling through

Is there any chance this happened at the same time as other network
activity? For example while fetching mail?

If so, please can you say what other software and what versions?

Cheers,

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: User feedback requested (esp. RISC OS)

On 29 Sep 2012 Peter Young <pnyoung@ormail.co.uk> wrote:

> On 29 Sep 2012 Paul Sprangers <Paul@sprie.nl> wrote:

>> Hi all,

>> Version 3.0 (development) seems to work fine (and fast) on my RISC OS 5.18
>> computer. Keep up the good work (just add java script).

> Also here, ARMini, RISC OS 5.19. Glad to see that test builds are
> available again, many thanks.

Might have spoken too soon. I sent this message using Hermes, and when
NewsHound kicked in, the machine froze. The message evidently went,
but is still in Messenger Pro's queue. Going back to NetSurf 2.9 for
the time being, I regret.

With best wishes,

Peter.

--
Peter \ / zfc Ta \ Prestbury, Cheltenham, Glos. GL52
and \/ __ __ \ England.
family / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
/ \__/ \_/ | \| \__/ \______________ pnyoung@ormail.co.uk

Re: User feedback requested (esp. RISC OS)

front end: RISC OS
OS version: 5.18
problem: freeze while scrolling through
https://www.riscosopen.org/forum/forums/1/topics/1321
download file: netsurf-2012-09-28_19-54-37/zip

It certainly feels faster than 2.9.
--
Gavin Wraith (gavin@wra1th.plus.com)
Home page: http://www.wra1th.plus.com/

Re: User feedback requested (esp. RISC OS)

[snip]


> It would help us if people could test the new builds and let us know how
> you get on.

I have just downloaded and briefly used Netsurf from archive
<netsurf-2012-09-28_19-54-37/zip>.

So far it has been in use for about 35 minutes, visiting several sites
without problem.

--

Brian Jordan
Virtual RPC-AdjustSA
RISC OS 6.20

Re: User feedback requested (esp. RISC OS)

On 29 Sep 2012 Paul Sprangers <Paul@sprie.nl> wrote:

> Hi all,

> Version 3.0 (development) seems to work fine (and fast) on my RISC OS 5.18
> computer. Keep up the good work (just add java script).

Also here, ARMini, RISC OS 5.19. Glad to see that test builds are
available again, many thanks.

With best wishes,

Peter.

--
Peter \ / zfc Ta \ Prestbury, Cheltenham, Glos. GL52
and \/ __ __ \ England.
family / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
/ \__/ \_/ | \| \__/ \______________ pnyoung@ormail.co.uk

Re: User feedback requested (esp. RISC OS)

Hi all,

Version 3.0 (development) seems to work fine (and fast) on my RISC OS 5.18
computer. Keep up the good work (just add java script).

Kind regards,
Paul Sprangers

User feedback requested (esp. RISC OS)

Hi,

We have done quite a bit of development since we last issued test builds
and we have also changed our autobuilding infrastructure completely. With
these changes we are now building NetSurf builds with far more up-to-date
versions of 3rd party libraries, and it is now quite trivial to keep them
up-to-date.

It would help us if people could test the new builds and let us know how
you get on.


CAUTION: SAVE ALL WORK BEFORE RUNNING THESE BUILDS

We have at least one case of NetSurf freezing a
RISC OS Iyonix such that it needs to be reset.


New builds are available from:

http://ci.netsurf-browser.org/ (Use the latest one available)


If you try one of the new builds, please let us know:

+ which front end (e.g. RISC OS, Amiga, etc)
+ which version of the OS
+ whether there was a problem
> what was happening when it went wrong
+ what the download file name of the NetSurf version you tried was

Cheers,

--

Michael Drake (tlsa) http://www.netsurf-browser.org/

Re: can't get to ROOL site

(Try again, using a proper email reader this time...)

The following bytes were arranged on 26 Sep 2012 by Jim Nagel :

> Jim Nagel was stymied by:
> >> Warning from Netsurf
> >> "error setting certificate verify locations:"
> >> (Nothing after the colon)
> >> What could the explification be?

[snip]

> Perhaps the possible cure could be footnoted to the error message for
> future patients.

Personally I'd prefer the bug being fixed.

--
__<^>__ Red sky in the morning: Shepherd's warning
/ _ _ \ Red sky at night: Shepherd's delight
( ( |_| ) ) Mince and potatoes: Shepherd's pie
\_> <_/ ======================= Martin Bazley ==========================

[Rpcemu] Some RPCEmu observations

When my Iyonix stopped working earlier this week, I decided to move as
many things as possible to RPCEmu on my Linux box. I have now managed to
resurrect the Iyo (PSU problem - again), but I intend to keep the RPCEmu
setup up to date, just in case.
While working with the emulator on a daily basis, I noticed a few
things:

- Using a RiscPC keyboard, the ` and ' keys are switched around.
Question: Which part of the source code do I have to change to fix
this?

- Ctrl+End is not a good choice to call up the menu.
I have used this key combination with StrongED for many years and
not being able to use it is more than a minor nuisance.
Question: Which part of the source code do I have to change to fix
this?

- Top bit set characters in filenames can be problematic.
I know, you shouldn't use any, but spaces tend to creep in almost
unnoticed. RISC OS fakes them in filenames by coding them to 160 and
RPCEmu uses the ISO-8859-1 character set while most Linux distros
will be on UTF-8 these days. This can cause problems if you want to
access such a file from Linux.
Question: Is there a way around this?

- Servers running inside the emulator cannot be reached and active FTP
doesn't work.
Not surprising. Would be nice if there was a way around this (maybe
some more iptables magic? need to study this...).

- Networking is unreliable with RISC OS 5.19.
Spurious errors occurred during an AntiSpam session of around 40
messages. The session had to be restarted several times and each time
processing a different message would produce another kind of
nonsensical error. In the end I gave up and switched to RISC OS 4.39,
which worked like a charm.
This is probably more of an RO 5.19 problem than something to do with
RPCEmu, but I thought I would also mention it here for completeness.

Also, is there any chance of the serial port and mouse scrollwheel ever
being made to work?

Details of RPCEmu used: version 0.8.9, recompiler, SA110, 256M.

Regards,
Frank


_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu

Re: can't get to ROOL site

Jim Nagel was stymied by:
>> Warning from Netsurf
>> "error setting certificate verify locations:"
>> (Nothing after the colon)
>> What could the explification be?

Martin Bazley prescribed on 25 Sep:
> * Open Choices window
> * Click anywhere
> * Click 'Set'
> * Restart

> It's a longstanding known bug when using default (i.e. not loaded from
> file) choices.

Precise remedy. Problem cured. Thanks, Dr Martin!


> You haven't fiddled with your boot sequence recently, have you?

Not exactly "fiddled". I haven't used this Iyonix for some while, so
recently upgraded it to 5.18 and then freshly installed Netsurf 2.9 on
it. Have been using 2.9 on RiscPC for yonks -- which explains why it
didn't show this baffling "certificate verify location" thing.

Perhaps the possible cure could be footnoted to the error message for
future patients.

--
Jim Nagel www.archivemag.co.uk
See you at London? www.riscoslondonshow.co.uk October 27

Re: Changes at QVC

In article <0769efd452.pnyoung@pnyoung.ormail.co.uk>,
Peter Young <pnyoung@ormail.co.uk> wrote:
> On 25 Sep 2012 Dave Symes <dave@triffid.co.uk> wrote:

> > Oh dear, Fay is not a happy person at the moment, apparently QVC have
> > re-designed their front page, and as she uses NetSurf (r13571) to view
> > the stuff, suddenly she can't see menus and such like that a couple of
> > days ago she could.

> > I have to assume it's something to do with more javascript, but way
> > outside my paygrade so to speak.

> > Could someone with knowledge please have a quick look at...
> > http://www.qvcuk.com

> > It's mainly the menus down the lefthand side.

> I don't have knowledge, but in so far as my knowledge goes, I've done
> a bit of furkling. Here, the menus on the left show briefly on
> loading, but are then overwritten by a blank pane. Looking at the
> source, (F8) it seems to teem with JavaScript indeed; StrongED reports
> "79 hits on 49 lines". I also looked on their FAQ page, on which is:

> <quote>

> To enjoy most of our content your browser must also be set to the
> following options:

> JavaScript enabled
> ActiveX enabled (Internet Explorer only)
> Pop-ups allowed from www.qvcuk.com

> <unquote>

[Snippy]

> With best wishes,

> Peter.

At the moment I've got her to use the status bar... That is slowly scroll
the mouse pointer down where the menu should display, looking carefully at
the status bar on the bottom of the NS window, as the pointer scrolls
down, the links are shown and she can double click 'em.

Dave

--

Dave Triffid

Re: Changes at QVC

On 25 Sep 2012 Dave Symes <dave@triffid.co.uk> wrote:

> Oh dear,
> Fay is not a happy person at the moment, apparently QVC have re-designed
> their front page, and as she uses NetSurf (r13571) to view the stuff,
> suddenly she can't see menus and such like that a couple of days ago she
> could.

> I have to assume it's something to do with more javascript, but way
> outside my paygrade so to speak.

> Could someone with knowledge please have a quick look at...
> http://www.qvcuk.com

> It's mainly the menus down the lefthand side.

I don't have knowledge, but in so far as my knowledge goes, I've done
a bit of furkling. Here, the menus on the left show briefly on
loading, but are then overwritten by a blank pane. Looking at the
source, (F8) it seems to teem with JavaScript indeed; StrongED reports
"79 hits on 49 lines". I also looked on their FAQ page, on which is:

<quote>

To enjoy most of our content your browser must also be set to the
following options:

JavaScript enabled
ActiveX enabled (Internet Explorer only)
Pop-ups allowed from www.qvcuk.com

<unquote>

So it looks as if it's a Dark Side only site, and that they have done
the dreaded "In the interests of product improvement we reserve the
right to foul it up at any time".

There is a feedback option on the page, which of course doesn't work,
and also a Contact Us option at the bottom, which does, but I fear
that trying to persuade them that they've sinned might activate the GM
again :-)

Some sites have a "mobile" option, which sometimes works with NetSurf,
but I can't see such a one there.

With best wishes,

Peter.

--
Peter \ / zfc Ta \ Prestbury, Cheltenham, Glos. GL52
and \/ __ __ \ England.
family / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
/ \__/ \_/ | \| \__/ \______________ pnyoung@ormail.co.uk

Re: can't get to ROOL site

The following bytes were arranged on 25 Sep 2012 by Jim Nagel :

> Warning from Netsurf
> "error setting certificate verify locations:"
> (Nothing after the colon)
>
> What could the explification be?

* Open Choices window

* Click anywhere

* Click 'Set'

* Restart

It's a longstanding known bug when using default (i.e. not loaded from
file) choices. You haven't fiddled with your boot sequence recently,
have you?

--
__<^>__ "Start off every day with a smile and get it over with."
/ _ _ \ - W.C. Fields
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

Re: can't get to ROOL site

On 25 Sep, Jim Nagel wrote in message
<9e25d3d452.jim@nails.abbeypress.net>:

> On the Netsurf 2.9 home page, clicking on the link to ROOL produces a
> cryptic error:
> Warning from Netsurf
> "error setting certificate verify locations:"
> (Nothing after the colon)
>
> No idea what this means. The guys at ROOL say their SSL certificate is
> unchanged and valid until the end of 2013. Besides, this link is calling
> http, not https. (Using an Iyonix here, OS 5.18.)

ROOL's site serves everything securely, IIRC.

> What could the explification be?
>
> However, I can go to a page further in at the ROOL site, which I found
> from a Google search --
> http://www.riscosopen.org/content/downloads/module-zipfiles -- and from
> the "tab" panel there to other pages at ROOL.
>
> Curiously, Netsurf 2.9 on the RiscPC (OS 4.39) has no problem.

Have you tried cleaning out (or at least temporarily hiding) *all* of
NetSurf's data inside !Boot on the affected system? There's some in Choices
and some in Scrap.

--
Steve Fryatt - Leeds, England

http://www.stevefryatt.org.uk/

Changes at QVC

Oh dear,
Fay is not a happy person at the moment, apparently QVC have re-designed
their front page, and as she uses NetSurf (r13571) to view the stuff,
suddenly she can't see menus and such like that a couple of days ago she
could.

I have to assume it's something to do with more javascript, but way
outside my paygrade so to speak.

Could someone with knowledge please have a quick look at...
http://www.qvcuk.com

It's mainly the menus down the lefthand side.

Thanks
Dave

--

Dave Triffid

Re: can't get to ROOL site

On 25 Sep, Peter Young wrote:
> On 25 Sep 2012 Jim Nagel <netsurf@abbeypress.co.uk> wrote:

[snip]

>> The guys at ROOL say their SSL certificate is unchanged and valid
>> until the end of 2013. Besides, this link is calling http, not https.
>> (Using an Iyonix here, OS 5.18.)

> Works fine here with the same NetSurf and RISC OS 5.19, ARMini.

Also works here with both 2.9 and 3.0 r13571 on Iyonix with 5.18.
--
Brian Howlett
----------------------------------------------
All electronic components run on smoke.
If you let the smoke out, they stop working...

Re: can't get to ROOL site

On 25 Sep 2012 Jim Nagel <netsurf@abbeypress.co.uk> wrote:

> On the Netsurf 2.9 home page, clicking on the link to ROOL produces a
> cryptic error:
> Warning from Netsurf
> "error setting certificate verify locations:"
> (Nothing after the colon)

> No idea what this means. The guys at ROOL say their SSL certificate
> is unchanged and valid until the end of 2013. Besides, this link is
> calling http, not https. (Using an Iyonix here, OS 5.18.)

Works fine here with the same NetSurf and RISC OS 5.19, ARMini.

With best wishes,

Peter.

--
Peter \ / zfc Ta \ Prestbury, Cheltenham, Glos. GL52
and \/ __ __ \ England.
family / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
/ \__/ \_/ | \| \__/ \______________ pnyoung@ormail.co.uk

can't get to ROOL site

On the Netsurf 2.9 home page, clicking on the link to ROOL produces a
cryptic error:
Warning from Netsurf
"error setting certificate verify locations:"
(Nothing after the colon)

No idea what this means. The guys at ROOL say their SSL certificate
is unchanged and valid until the end of 2013. Besides, this link is
calling http, not https. (Using an Iyonix here, OS 5.18.)

What could the explification be?

However, I can go to a page further in at the ROOL site, which I found
from a Google search --
http://www.riscosopen.org/content/downloads/module-zipfiles -- and
from the "tab" panel there to other pages at ROOL.

Curiously, Netsurf 2.9 on the RiscPC (OS 4.39) has no problem.

--
Jim Nagel www.archivemag.co.uk
See you at London? www.riscoslondonshow.co.uk October 27

[GeSHi-devel] [Release] Three new language files

Hi,

I would like to release three language files (from the era of the mainframes).

- JCL (Job Control Language) with additional support for Mainframe utilities such as DFSORT, ICETOOL and IDCAMS
- EZT (Easytrieve)
- ISPF Panel Definition Language

The files can be downloaded from the following link: http://thecodeisclear.in/wp-content/plugins/download-monitor/download.php?id=26

I want to validate it, but langcheck.php seems to hang (on localhost). I have tested the language files and the highlighting looks good. Samples and more info in this post (http://thecodeisclear.in/geshi-languages-mainframe/). What are the next steps to have these as part of the next release?

Special thanks to Carne, Benny for answering my queries through this list.

Thanks,
Ramesh

Re: [gccsdk] Clear screen under Task Window

In message <001501cd97b6$ea09b840$be1d28c0$@tenox.tc>
<tenox@tenox.tc> wrote:

> So is there any way to clear the screen output of Edit window from a "text
> mode" application?
>
> I found that when I double click on the binary produced by gcc it opens a
> small roughly 80x24 window where VDU codes do work. But they do not seem to
> work inside Edit. Is there any other way?

Perhaps GraphTask @ http://www.armclub.org.uk/free is what you're looking
for.

John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

The following bytes were arranged on 21 Sep 2012 by Theo Markettos :

> The wget commit log is here:
> http://git.savannah.gnu.org/cgit/wget.git/log/
> if you want to poke through it to find out if the change was deliberate or a
> bug (it's pretty busy).

Sadly I couldn't locate the code which spat out the 200 OK message, but
I did find this illuminating commit message for the equivalent FTP code:

http://git.savannah.gnu.org/cgit/wget.git/commit/?id=ef113e51d8c80b16902042de7279ef634e7fdc2a

So it looks very like a deliberate change. But like I said, it probably
never crossed his mind that anybody might be relying on the distinction
between output/no output.

> I think the right answer here is either to use the return value, which as
> you say has concurrency issues under RISC OS, or to use curl. IME wget is
> more of a UI tool and curl is more of a 'plumbing' tool. Matching the UI
> output (as you're doing from wget) is akin to telling a GUI tool to click on
> the 4th menu item down - it's fragile and not likely to be stable. (For
> starters, what happens in translated versions of wget? I have no idea).

From a brief browse it seems as if curl has similar problems (one
command-line solution advocated using grep to second-guess the
diagnostic output, which is exactly the approach I want to avoid),
unless one uses libcurl, which is completely out in BASIC.

(It does remind me of an idea I've been playing with for a while, some
form of automated translation of C library headers into RISC OS modules
so they could both be used from other languages and shared between C
programs without the need to faff about with !SOManager. I'm convinced
it should be theoretically possible, given a set of fixed, sensible
calling standards and nailing down the format of C's data types,
particularly structs, but it's not happening any time soon. But hey, I
see one of my modules for this year is entitled "Compilers", so you
never know...)

--
__<^>__
/ _ _ \ It is written that Geeks shall inherit the Earth.
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

On Fri, Sep 21, 2012 at 08:51:05PM +0100, Martin Bazley wrote:
> I think I may need to introduce you to the concept of an "illustrative
> example"...

My apologies. It's been a very long week.

The wget commit log is here:
http://git.savannah.gnu.org/cgit/wget.git/log/
if you want to poke through it to find out if the change was deliberate or a
bug (it's pretty busy).

I think the right answer here is either to use the return value, which as
you say has concurrency issues under RISC OS, or to use curl. IME wget is
more of a UI tool and curl is more of a 'plumbing' tool. Matching the UI
output (as you're doing from wget) is akin to telling a GUI tool to click on
the 4th menu item down - it's fragile and not likely to be stable. (For
starters, what happens in translated versions of wget? I have no idea).

But yes, it does raise the issue of per-process return codes. I'm not sure
of the best way to implement that. DDEUtils perhaps? (like per-process
current directories)

Theo

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

The following bytes were arranged on 21 Sep 2012 by Theo Markettos :

> Can you give a specific command line that's failing? I just used the
> autobuilder to download source to Debian's wget 1.13.4-3 and built it on
> Linux, trying:
>
> $ src/wget http://www.somethi.ng/valid.html
> --2012-09-21 17:44:01-- http://www.somethi.ng/valid.html
> Resolving www.somethi.ng (www.somethi.ng)... failed: Name or service not known.

I think I may need to introduce you to the concept of an "illustrative
example"...

And just for the avoidance of doubt:

*wget -nv --spider http://www.riscos.com
2012-09-21 20:44:51 URL: http://www.riscos.com/ 200 OK

*wget -nv --spider http://www.riscos.com/wibble.html
http://www.riscos.com/wibble.html:
Remote file does not exist -- broken link!!!

With the earlier version of wget, the first command produces no output
but the second does. The program has changed since then - probably by
someone who thought it was just an insignificant change which would have
no effect whatsoever - so that the two outcomes can no longer be easily
distinguished.

I'm leaning towards shunting all the wget calls into an Obey file and
giving up on the idea of using a TaskWindow altogether, so that I can be
sure nothing else can poke the value of Sys$ReturnCode. That would
still solve the WimpSlot problem and it doesn't multitask in the public
release anyway, so it's not like it's a downgrade.

--
__<^>__ "Your pet, our passion." - Purina
/ _ _ \ "Your potential, our passion." - Microsoft, a few months later
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

On Thu, Sep 20, 2012 at 11:45:52PM +0200, John Tytgat wrote:
> Could very well be a bug in that particular version (or Debian patches
> if any). Have you tried mining the wget ML/IRC and/or manual mentioned
> at www.gnu.org/software/wget/ ?

Can you give a specific command line that's failing? I just used the
autobuilder to download source to Debian's wget 1.13.4-3 and built it on
Linux, trying:

$ src/wget http://www.somethi.ng/valid.html
--2012-09-21 17:44:01-- http://www.somethi.ng/valid.html
Resolving www.somethi.ng (www.somethi.ng)... failed: Name or service not known.
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv --spider http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv --spider http://not.val.id/at.all
wget: unable to resolve host address `not.val.id'
$ src/wget -nv --spider http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv --spider http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv --spider http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'
$ src/wget -nv --spider http://www.somethi.ng/valid.html
wget: unable to resolve host address `www.somethi.ng'


I'm puzzled that it got '200 OK', because that's being returned by the
server. You don't have any kind of proxy in the way - for example your ISP
isn't doing transparent proxying or deep packet inspection or anything?
Try
*wget -v --save-headers http://www.riscos.com
and look at the index/html file. I get:

HTTP/1.1 200 OK
Date: Fri, 21 Sep 2012 16:52:22 GMT
Server: Apache/2
Last-Modified: Mon, 09 Jan 2012 17:43:58 GMT
ETag: "2fb3-4b61bf1721380"
Accept-Ranges: bytes
Content-Length: 12211
Keep-Alive: timeout=15, max=100
Connection: Keep-Alive
Content-Type: text/html

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
[etc]

Do you get any more headers than that?

Theo

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Clear screen under Task Window

The following bytes were arranged on 21 Sep 2012 by tenox@tenox.tc:

> So is there any way to clear the screen output of Edit window from a "text
> mode" application?

From a brief test, the answer would appear to be no. Sorry.

You could try using StrongED instead:

http://stronged.iconbar.com/

Both BASIC's CLS and VDU 12 seem to work with that.

IIRC, getting StrongED to claim TaskWindow requests is a bit tricky.
There's an obey file called !StrongED.Defaults.ClaimFiles, which you
need to edit before StrongED will act as your default text editor. It's
also a good idea to add StrongED to your 'Look at' list on bootup (Menu
over task manager, Configure, Boot, Look at).

> I found that when I double click on the binary produced by gcc it opens a
> small roughly 80x24 window where VDU codes do work.

That's a command window, which doesn't multitask, unlike a TaskWindow.
It's provided by RISC OS, so it can do a lot of things TaskWindows
can't, such as draw graphics.

--
__<^>__ Red sky in the morning: Shepherd's warning
/ _ _ \ Red sky at night: Shepherd's delight
( ( |_| ) ) Mince and potatoes: Shepherd's pie
\_> <_/ ======================= Martin Bazley ==========================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Clear screen under Task Window

> Look, if you reboot the computer and press Ctrl+F12, which
> application's icon gets added to the icon bar? That's the one you
> just ran to provide you with your TaskWindow.

I found it, it's Edit.

So is there any way to clear the screen output of Edit window from a "text
mode" application?

I found that when I double click on the binary produced by gcc it opens a
small roughly 80x24 window where VDU codes do work. But they do not seem to
work inside Edit. Is there any other way?

Thank you,
Antoni


_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Sys$ReturnCode (was: Parsing wget output)

In message <76bd60d252.martin@blueyonder.co.uk>
Martin Bazley <martin.bazley@blueyonder.co.uk> wrote:

> Ah - I've just discovered that it *will* set Sys$ReturnCode to
> zero/nonzero.

Ah I thought you were already doing that.

> So with a bit more tweaking, it should be possible to
> change MBBack to use that.
>
> However, I did discover that the value of this variable is very
> volatile. How safe is it to read? I'm planning on calling wget in a
> TaskWindow with Wimp_StartTask, and examining the return code once I get
> the TaskCloseDown Wimp message. Is it possible that some other task may
> alter its value between wget quitting and MBBack learning of it,
> assuming that I am not the only program calling Wimp_Poll in the
> desktop?
>
> An alternative solution would be to write a wrapper Obey file and
> Wimp_StartTask that, but I'd rather avoid that.

I would write a small C program launching your wget command via system()
or fork()/execve() and use the return code to set a system variable which
is somehow registered. And call that C program using Wimp_StartTask
TaskWindow instead.

I don't think Sys$ReturnCode is preserved in a TaskWindow context switch.

John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

[gccsdk] Sys$ReturnCode (was: Parsing wget output)

Ah - I've just discovered that it *will* set Sys$ReturnCode to
zero/nonzero. So with a bit more tweaking, it should be possible to
change MBBack to use that.

However, I did discover that the value of this variable is very
volatile. How safe is it to read? I'm planning on calling wget in a
TaskWindow with Wimp_StartTask, and examining the return code once I get
the TaskCloseDown Wimp message. Is it possible that some other task may
alter its value between wget quitting and MBBack learning of it,
assuming that I am not the only program calling Wimp_Poll in the
desktop?

An alternative solution would be to write a wrapper Obey file and
Wimp_StartTask that, but I'd rather avoid that.

--
__<^>__
/ _ _ \ You always find something in the last place you look.
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

In message <d17a5ad252.martin@blueyonder.co.uk>
Martin Bazley <martin.bazley@blueyonder.co.uk> wrote:

> It's easy enough when I actually want to download the file, as I can
> just check if the requested target file exists or not, but returning
> results from the program seems to be a minefield when there's no human
> supervision. Surely someone must be using wget to do this, somewhere?

Could very well be a bug in that particular version (or Debian patches
if any). Have you tried mining the wget ML/IRC and/or manual mentioned
at www.gnu.org/software/wget/ ?

John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Parsing wget output

On 09/20/2012 02:17 PM, Martin Bazley wrote:

> It's easy enough when I actually want to download the file, as I can
> just check if the requested target file exists or not, but returning
> results from the program seems to be a minefield when there's no human
> supervision. Surely someone must be using wget to do this, somewhere?
>

The AB uses wget because that's what I've been using for years and years.
However, there's really nothing stopping it being changed to curl - you
might find its error handling more satisfactory.




_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

[gccsdk] Parsing wget output

I've just been testing wget 1.13.4 from the autobuilder to make sure
it's ARMv7 safe, and while I was at it I checked it was still compatible
with MBBack. The answer to the second question turned out to be no.

wget 1.11.4, which is what I was using before, could be persuaded to
check for the existence of a remote file without downloading it. I did
this by a combination of the switches '-nv' (non-verbose output), '-o'
(write stderr to file) and '--spider' (don't download). I discovered
through experimentation that the -nv option only considered error
messages 'important' enough to write, and therefore if the output file
was blank then the operation could be presumed a success.

Unfortunately the behaviour of the -nv option has changed, and now it
reports success as well. It doesn't even report it consistently:

2012-09-20 21:52:01 URL: http://www.somethi.ng/valid.html 200 OK

http://not.val.id/at.all:
Remote file does not exist -- broken link!!!

What should I do with this? Check the last three bytes of the output
file for consecutive exclamation marks? Assume that a single line
containing the string "200 OK" is the only possible non-error output,
just as no output was previously the only possible non-error output?

Is there any way of calling wget from a computer program at all?

It's easy enough when I actually want to download the file, as I can
just check if the requested target file exists or not, but returning
results from the program seems to be a minefield when there's no human
supervision. Surely someone must be using wget to do this, somewhere?

--
__<^>__
/ _ _ \ I don't have a problem with God; it's his fan club I can't stand.
( ( |_| ) )
\_> <_/ ======================= Martin Bazley ==========================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Clear screen under Task Window

The following bytes were arranged on 20 Sep 2012 by tenox@tenox.tc:

> > > Is there any way one can clear the screen under Task Window?
> >
> > Important question: Which program are you running to provide this
> > TaskWindow? Edit, Zap, StrongED or something else?
>
> I middle click on the RiscOS icon in bottom right corner and select Task
> Window ^F12, so I would call it the built-in. Or is it the Edit?

There isn't a "built-in". All TaskWindows are provided by a text
editor - there is a protocol followed by applications to trap TaskWindow
requests and provide the service themselves.

Look, if you reboot the computer and press Ctrl+F12, which application's
icon gets added to the icon bar? That's the one you just ran to provide
you with your TaskWindow.

--
__<^>__ === RISC OS is a work of art. Some people adore it, ===
/ _ _ \ === others can't see the point of it, and it's really ===
( ( |_| ) ) === expensive. ===
\_> <_/ ======================= Martin Bazley ===================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Clear screen under Task Window

> > Is there any way one can clear the screen under Task Window?
>
> Important question: Which program are you running to provide this
> TaskWindow? Edit, Zap, StrongED or something else?

I middle click on the RiscOS icon in bottom right corner and select Task
Window ^F12, so I would call it the built-in. Or is it the Edit?

Thanks,
Antoni


_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] Clear screen under Task Window

The following bytes were arranged on 19 Sep 2012 by tenox@tenox.tc:

> Is there any way one can clear the screen under Task Window?

Important question: Which program are you running to provide this
TaskWindow? Edit, Zap, StrongED or something else?

--
__<^>__ === RISC OS is a work of art. Some people adore it, ===
/ _ _ \ === others can't see the point of it, and it's really ===
( ( |_| ) ) === expensive. ===
\_> <_/ ======================= Martin Bazley ===================

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

[gccsdk] Clear screen under Task Window

Hi,

Another question.

Is there any way one can clear the screen under Task Window?

I've tried sending form feed (VDU 12), all vt100, ANSI and other codes I
could think of but nothing works.

I also tried CLS and VDU 12 in BASIC which are supposed to work but they do
not.

I'm looking at OS_GSTrans() as possible escape codes but I don't really see
it working.

I saw some mentions of Curses port to RISC OS. Does it exist?

Thanks,
Antoni



_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] GCC file format not recognized

> -----Original Message-----
> From: Duncan Moore [mailto:duncan.moore@gmx.com]
> Sent: Tuesday, September 18, 2012 2:43 AM
> To: Antoni Sawicki
> Cc: gcc@gccsdk.riscos.info
> Subject: Re: [gccsdk] GCC file format not recognized
>
> gcc uses Unix filename semantics - these are translated behind the scenes.
> Also, for historical reasons (old limits on leafname length and number of
files
> per directory), gcc uses a non-obvious source file organisation. Name your
> file c.hello (RISC OS semantics), i.e. a file 'hello' in directory 'c'.
Then use 'gcc
> hello.c' to compile it - from the directory with 'c' in it. Similarly,
header files
> will go in a directory 'h'. When including headers though, use #include
"abc.h"
> as you would normally.
>

Thank you very much, it worked!

Antoni


_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: Multiple call for Iconv

In message <52d131dde6mail@JohnWoodhouse.plus.com>
<mail@JohnWoodhouse.plus.com> wrote:

> Having installed Reporter for another unrelated reason, I have noticed that
> on every occasion a Netsurf window is opened, Netsurf calls for the Iconv
> module.
> e.g. :
>
> 16:08:19.82 [Appl/NetSurf] RMEnsure Iconv 0.04 RMload System:Modules.Iconv
> 16:08:19.82 [Appl/NetSurf] RMEnsure Iconv 0.04 Error 16_10F iconv support
> requires the Iconv module 0.04 or newer
>
> [...] The above is repeated 32 times each time a new
> window is opened!

Right, this is done by UnixLib for each iconv_open() call and should
actually only be done once when it is successful. I'll queue a fix
for UnixLib but it could take a while before this gets into a new
NetSurf build.

John.
--
John Tytgat
joty@netsurf-browser.org

Multiple call for Iconv

Having installed Reporter for another unrelated reason, I have noticed that
on every occasion a Netsurf window is opened, Netsurf calls for the Iconv
module.
e.g. :

16:08:19.82 [Appl/NetSurf] RMEnsure Iconv 0.04 RMload System:Modules.Iconv
16:08:19.82 [Appl/NetSurf] RMEnsure Iconv 0.04 Error 16_10F iconv support
requires the Iconv module 0.04 or newer

I have a copy of Iconv :
==> Help on keyword Iconv
Module is: Iconv 0.11 (04 Jan 2011)

in System:Modules, so I don't understand why it is constantly being called
for - and a lot of times. The above is repeated 32 times each time a new
window is opened!

John

--
____
/__ __________________________
/____Mail from mail@JohnWoodhouse.plus.com

Re: [gccsdk] GCC file format not recognized

On 18/09/2012 08:21, Antoni Sawicki wrote:
> I'm new to the list and RiscOS in general. I have a quick question. I
Welcome.
> *gcc hello/c
> hello/c: file not recognized: file format not recognized
> collect2: ld returned 1 exit status
Have a look at the Examples directory that came with !GCC, for how to
organise your files, and also !GCC.!Help, particularly section 1.2.

gcc uses Unix filename semantics - these are translated behind the
scenes. Also, for historical reasons (old limits on leafname length and
number of files per directory), gcc uses a non-obvious source file
organisation. Name your file c.hello (RISC OS semantics), i.e. a file
'hello' in directory 'c'. Then use 'gcc hello.c' to compile it - from
the directory with 'c' in it. Similarly, header files will go in a
directory 'h'. When including headers though, use #include "abc.h" as
you would normally.



_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

[gccsdk] GCC file format not recognized

Hi,

I'm new to the list and RiscOS in general. I have a quick question. I
have
installed GCC version 4.1.2 (GCCSDK GCC 4.1.2 Release 1) on RiscOS 4.
When I
try to compile hello world I get following error:

*gcc hello/c
hello/c: file not recognized: file format not recognized
collect2: ld returned 1 exit status

Can anyone help?

Thanks,
Antoni

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

[gccsdk] GCC file format not recognized

Hi,

I'm new to the list and RiscOS in general. I have a quick question. I have
installed GCC version 4.1.2 (GCCSDK GCC 4.1.2 Release 1) on RiscOS 4. When I
try to compile hello world I get following error:

*gcc hello/c
hello/c: file not recognized: file format not recognized
collect2: ld returned 1 exit status

Can anyone help?

Thanks,
Antoni


_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

NetSurf developer weekend

The next NetSurf Developer weekend will be taking place between 2nd
and 4th November in Cambridge, UK.

The rough plan is to arrive in Cambridge on Friday afternoon/evening
meet at the venue possibly have a social meetup. Return to the meeting
rooms on Saturday and Sunday and work on NetSurf with food and snack
breaks as required.

Collabora are going to sponsor the event by providing access to their
conference rooms.

The address of the location is:

Kett House
Station Rd
Cambridge CB1 2JH
United Kingdom

Google maps ref http://goo.gl/maps/F1CG and the location is very close
to the railway station.

For building security purposes all physical attendees *must* register
(send me an email at least a week before the event) with Name, email
and contact details.

Already Registered are:

John-Mark Bell
Daniel Silverstone
Rob Kendrick
Vincent Sanders

Assistance with accommodation is available and I have a vehicle for
collecting and dropping people off from wherever they are
staying. There are a limited number of parking spaces available at
Kett house if necessary.

If you would like to join us for what is usually a very productive and
enjoyable weekend please let me know as soon as you can.

--
Regards Vincent

Re: [gccsdk] GCC on ROOL compatibility list

In message <603bbdcf52.martin@blueyonder.co.uk> you wrote:

> At present I'm slowly working through ROOL's ARMv6/v7 compatibility list
> and trying to turn it into something which can really be relied on, with
> cooperation from software authors wherever possible.
>
> http://www.riscosopen.org/wiki/documentation/show/ARMv6%2Fv7%20software%20compatibility%20list
>
> [...]
> There are two grades of 'compatible' - 'Seems OK', which means nobody's
> made it go wrong yet, and the 'ARMv7 OK' logo, which carries the
> explicit endorsement of its developers.

As far as I know, the latest official GCCSDK version released (4.1.2 Rel 1)
is 'ARMv7 OK', i.e. "The program is believed to be compatible with ARMv6/v7
by the person responsible for providing the newest version (whether official
or unofficial). (This doesn't necessarily imply continuing support.)"

Note this is about using the GCC compiler on ARMv7 RISC OS machine. Most
probably people want to know whether that compiler can also be used to
produce ARMv7 compatible programs. Again, as far as I know, that's the
case as well.

> The problem with collaborative projects like GCC (currently marked
> 'Seems OK') is that it's difficult to know who to ask, or indeed if
> there is anybody to ask. (The response I got from the NetSurf team was
> that "you have just as much right as anyone else to declare it
> compatible".) Hence this post.

BTW, all GCCSDK and Autobuilder communication happens best via its mailing
list mentioned at gccsdk.riscos.info. So you did the right thing. :-)

> Even more thorny than the issue of compatibility is the question of
> support, as it effectively carries a guarantee that the maintainer(s)
> of a given piece of software will be willing to fix any problems
> discovered for the forseeable future.

Speaking for myself but I'm not giving any guarantee nor warranty for
GCCSDK releases (like mentioned in !GCC/docs/index.html, see "Licenses
and warranty" topic, and what "gcc --version" says).

On the other hand, I believe our trackrecord shows we've been willing
to fix any problems and/or accept/consider all patches supplied for
fixing and enhancing GCCSDK. And this for, ahem, 10+ years.

But again, not guarantee nor warranty.

> As this is quite a commitment, I
> always make sure to ask the person responsible for providing such
> support for permission before marking their software as 'Supported'.
> Again, I have no idea who that may be - or even if the only support that
> can be guaranteed is the lesser grade of 'Source Available', meaning
> that if something's wrong, you'll probably have to fix it yourself.
>
> So, from someone who feels qualified to answer, would it be OK to change
> GCC's status to 'ARMv7 OK' and 'Supported'?

My believe this is a reasonable change.

> Also, which is the earliest version of GCC which is ARMv7 compatible?

Go for 4.1.2 Release 1. I'm sure earlier versions of the compiler itself
where ok but I don't see the value of spending time figuring that out.

> And finally, since I'm not sure where the best place to ask about this
> is either, what's the status of the riscos.info packages of ported
> software, such as wget, DosBox, etc.? Are they compatible, are they
> supported? Is there any way of finding out for a given package who's
> responsible for fixing the software?

Cfr. Theo's answer. He's doing a great job atm. :-)

John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK

Re: [gccsdk] GCC on ROOL compatibility list

On Sat, Sep 15, 2012 at 08:27:14PM +0100, Martin Bazley wrote:
> The problem with collaborative projects like GCC (currently marked
> 'Seems OK') is that it's difficult to know who to ask, or indeed if
> there is anybody to ask. (The response I got from the NetSurf team was
> that "you have just as much right as anyone else to declare it
> compatible".) Hence this post.

Generally everything GCC-related is tagged 'GCC developers', with the email
address of this list. So this is the best place to ask.

> So, from someone who feels qualified to answer, would it be OK to change
> GCC's status to 'ARMv7 OK' and 'Supported'? Also, which is the earliest
> version of GCC which is ARMv7 compatible?

The question is more complex than you might expect, because the compiler may
run on an ARMv7 system but still generate code that isn't ARMv7 compatible.
John Tytgat is the person to give you a definitive answer and he'll probably
be along in a minute, but AIUI it's GCC 4.1.2 release 1 which both runs and
generates suitable code.

> And finally, since I'm not sure where the best place to ask about this
> is either, what's the status of the riscos.info packages of ported
> software, such as wget, DosBox, etc.? Are they compatible, are they
> supported? Is there any way of finding out for a given package who's
> responsible for fixing the software?

I'm gradually working my way through rebuilding them (nontrivial for various
reasons). To a first approximation, if you look at the filesystem:
http://www.riscos.info/packages/arm/
anything with a date of 2012 is likely to work, but earlier builds are
unlikely.

As for responsibility for maintenance, it's collectively 'GCC developers'
again. Which is both everyone and, worst case, no-one. You can look into
the SVN logs[1] to see who committed the original package and they might
have more motivation to fix it, but generally the first place to complain if
something's broken is here.

Theo

[1]
http://www.riscos.info/websvn/listing.php?repname=gccsdk&path=/trunk/autobuilder/

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK