Re: [gccsdk] GCC for RISC OS 4.7.4 release 3

On Tue, Jan 31, 2017 at 01:01:50PM +0000, alan buckley wrote:
> I see there has been a check in to fix this. I'll rebuild the compilers and upload them again.

There's a build #125 here:
http://ci.riscos.info/job/gcc-4.7-native/lastSuccessfulBuild/artifact/gcc4/release-area/riscpkg/autobuilder_website/

It might be better to use that since we track what it was built with.
Are you able to test and confirm functionality?

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] GCC for RISC OS 4.7.4 release 3

 

Chris Gransden wrote on 28 January 2017 13:00

 

> In article

<LOXP123MB13184E8FA8FCF8721E824C77F0760@LOXP123MB1318.GBRP123.PROD.OUTLOOK.COM>,

   alan buckley <alan_baa@hotmail.com> wrote:

> > I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear in the PackMan package list and the riscos info packages website tomorrow. The direct download should be available now.

 

> > I'll do a proper release announcement early next week and also update the GCC Releases page at riscos info. Is there anything else that needs updating?

 

> gcc -v and g++ -v are still showing as release 2.

 

I see there has been a check in to fix this. I'll rebuild the compilers and upload them again.

 

Regards,

Alan

Re: [gccsdk] GCC for RISC OS 4.7.4 release 3

Theo Markettos, on 30 Jan, wrote:

> On Mon, Jan 30, 2017 at 12:30:57PM +0000, David Pitt wrote:
> > Some time ago I could do commits, for !Thump. There is a bit of a C
> > expertise deficit here, I only tried this commit as it was simple stuff.
[
snip]

> What does 'svn commit' do?

nano opens.

I wrote a log entry, exited nano and confirmed the commit.


--
David Pitt

_______________________________________________
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 for RISC OS 4.7.4 release 3

On Mon, Jan 30, 2017 at 12:30:57PM +0000, David Pitt wrote:
> Some time ago I could do commits, for !Thump. There is a bit of a C
> expertise deficit here, I only tried this commit as it was simple stuff.
>
> Something may have got fouled up, I now see this repeatedly.
>
> pittdj@Ubuntu1604box:/media/pittdj/GCCSDK$ svn co
> svn://svn.riscos.info/gccsdk/trunk/gcc4/ gcc4
>
> Fetching external item into 'gcc4/riscos/asasm':
>
> Fetching external item into 'gcc4/riscos/asasm/decaof':
> Checked out external at revision 7129.
>
>
> Fetching external item into 'gcc4/riscos/asasm/elftoolchain':
> Checked out external at revision 7129.
>
> Checked out external at revision 7129.
>
> Checked out revision 7129.
> pittdj@Ubuntu1604box:/media/pittdj/GCCSDK$
>
> If that is my fault then my apologies.

That's normal - GCCSDK uses svn externals, in other words there are child
repos that also get checked out when you check out the top level.

What does 'svn commit' do?

> > I've committed a variant as r7129 - it's now a Makefile variable so is
> > easier to spot at the top of the file.
>
> A much better idea.

On the subject of curl, it looks like it's breaking in Jenkins because it
depends on cacertificates producing a library tarball for the curl job,
which it doesn't. I'm not sure what the build dependency on cacertificates
is meant to mean, given the certs are not compiled into the binary (AIUI).
(is this an install-time dependency?)

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] GCC for RISC OS 4.7.4 release 3

Theo Markettos, on 30 Jan, wrote:

[snip]
>
> The list of things we can't build is a bit long:
> http://ci.riscos.info/job/packages/ click on Last Failure for a full list.
> I suspect some of the failures are simple - any help in fixing them up is
> appreciated.

curl has just built here.

https://www.riscosopen.org/forum/forums/11/topics/8983
--
David Pitt

_______________________________________________
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 for RISC OS 4.7.4 release 3

Theo Markettos, on 30 Jan, wrote:

> On Mon, Jan 30, 2017 at 10:48:19AM +0000, David Pitt wrote:
> > This info is baked into gcc4/Makefile. The following fixed it.
> >
> > GCC_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release
> > 3' \ --with-bugurl=http://gccsdk.riscos.info/ BINUTILS_CONFIG_ARGS +=
> > --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
> > --with-bugurl=http://gccsdk.riscos.info/ GDB_CONFIG_ARGS +=
> > --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
> > --with-bugurl=http://gccsdk.riscos.info/
> >
> > I tried to commit that but it does not appear to have worked.
>
> You probably don't have commit rights - if you're going to be making more
> changes I can make you an account.

Some time ago I could do commits, for !Thump. There is a bit of a C
expertise deficit here, I only tried this commit as it was simple stuff.

Something may have got fouled up, I now see this repeatedly.

pittdj@Ubuntu1604box:/media/pittdj/GCCSDK$ svn co
svn://svn.riscos.info/gccsdk/trunk/gcc4/ gcc4

Fetching external item into 'gcc4/riscos/asasm':

Fetching external item into 'gcc4/riscos/asasm/decaof':
Checked out external at revision 7129.


Fetching external item into 'gcc4/riscos/asasm/elftoolchain':
Checked out external at revision 7129.

Checked out external at revision 7129.

Checked out revision 7129.
pittdj@Ubuntu1604box:/media/pittdj/GCCSDK$

If that is my fault then my apologies.

> I've committed a variant as r7129 - it's now a Makefile variable so is
> easier to spot at the top of the file.

A much better idea.

--
David Pitt

_______________________________________________
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 for RISC OS 4.7.4 release 3

On Mon, Jan 30, 2017 at 10:48:19AM +0000, David Pitt wrote:
> This info is baked into gcc4/Makefile. The following fixed it.
>
> GCC_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
> --with-bugurl=http://gccsdk.riscos.info/
> BINUTILS_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release
> 3' \
> --with-bugurl=http://gccsdk.riscos.info/
> GDB_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
> --with-bugurl=http://gccsdk.riscos.info/
>
> I tried to commit that but it does not appear to have worked.

You probably don't have commit rights - if you're going to be making more
changes I can make you an account.

I've committed a variant as r7129 - it's now a Makefile variable so is
easier to spot at the top of the file.

That's running through Jenkins (ci.riscos.info) now.
http://ci.riscos.info/job/gcc-4.7-native/124/

Yesterday I gave various of the CI machines a cleanout - Jenkins eats disc
for breakfast and several had tied themselves in knots (eg wouldn't boot).
Things should be back to normal now.

I started a full build of all the packages in the interests of cleaning the
cobwebs out of the infrastructure. Do we know how new packages need to be
to have ARMv8 support? Should we aim to do a full release of everything we
can build?

The list of things we can't build is a bit long:
http://ci.riscos.info/job/packages/
click on Last Failure for a full list.
I suspect some of the failures are simple - any help in fixing them up is
appreciated.

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] GCC for RISC OS 4.7.4 release 3

Chris Gransden, on 28 Jan, wrote:

> In article
>
<LOXP123MB13184E8FA8FCF8721E824C77F0760@LOXP123MB1318.GBRP123.PROD.OUTLOOK.COM>,
> alan buckley <alan_baa@hotmail.com> wrote:
> > I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear
> > in the PackMan package list and the riscos info packages website
> > tomorrow. The direct download should be available now.
>
> > I'll do a proper release announcement early next week and also update
> > the GCC Releases page at riscos info. Is there anything else that needs
> > updating?
>
> > Regards, Alan
>
> gcc -v and g++ -v are still showing as release 2.

This info is baked into gcc4/Makefile. The following fixed it.

GCC_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
--with-bugurl=http://gccsdk.riscos.info/
BINUTILS_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release
3' \
--with-bugurl=http://gccsdk.riscos.info/
GDB_CONFIG_ARGS += --with-pkgversion='GCCSDK GCC $(GCC_VERSION) Release 3' \
--with-bugurl=http://gccsdk.riscos.info/

I tried to commit that but it does not appear to have worked.

--
David Pitt

_______________________________________________
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] Qt riscos qpa plugin

On 29/01/17 12:46, Michael Grunditz wrote:
> In message <555f887d-4b80-7012-c42f-fe1981f07d9f@sky.com>
> Lee Noar <leenoar@sky.com> wrote:
>
>> On 29/01/17 09:48, Michael Grunditz wrote:
>>> Hi
>>>
>>> I am thinking of creating a riscos qpa plugin. Has anyone started
>>> working on this allready?
>
>> There's already a RISC OS platform plugin that uses a RISC OS sprite
>> as a window backing store that Qt can then render into. If you're
>> talking about something that makes Qt use native RISC OS gadgets, then
>> no there's nothing like that yet.
>
>> Lee.
>
> Ah sorry, I was under the impression that it used X11 libs.

No, fortunately, these type of libraries tend to be more
decoupled from X11 these days making porting easier. GTK3 did
the same thing.

Lee.


_______________________________________________
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] Qt riscos qpa plugin

In message <555f887d-4b80-7012-c42f-fe1981f07d9f@sky.com>
Lee Noar <leenoar@sky.com> wrote:

> On 29/01/17 09:48, Michael Grunditz wrote:
>> Hi
>>
>> I am thinking of creating a riscos qpa plugin. Has anyone started
>> working on this allready?

> There's already a RISC OS platform plugin that uses a RISC OS sprite
> as a window backing store that Qt can then render into. If you're
> talking about something that makes Qt use native RISC OS gadgets, then
> no there's nothing like that yet.

> Lee.

Ah sorry, I was under the impression that it used X11 libs.

--
Michael Grunditz

_______________________________________________
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] Qt riscos qpa plugin

On 29/01/17 09:48, Michael Grunditz wrote:
> Hi
>
> I am thinking of creating a riscos qpa plugin. Has anyone started
> working on this allready?

There's already a RISC OS platform plugin that uses a RISC OS sprite
as a window backing store that Qt can then render into. If you're
talking about something that makes Qt use native RISC OS gadgets, then
no there's nothing like that yet.

Lee.


_______________________________________________
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] Qt riscos qpa plugin

Hi

I am thinking of creating a riscos qpa plugin. Has anyone started
working on this allready?

--
Michael Grunditz

_______________________________________________
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] Problem with autobuilder (newbie questions)

In article <399fb40556.Matthew@sinenomine.freeserve.co.uk>,
Matthew Phillips <spam2011m@yahoo.co.uk> wrote:

> $ cd build
> $ ../autobuilder/build -v wget

> as instructed. As with gpsbabel the first thing it tries to do is fetch and
> compile native-zip. Again, after doing the patching, make starts and we get
> the same errors as before:

> make[1]: Entering directory `/home/matthew/gccsdk/build/native-zip/zip30'
> cc -c -I. -DUNIX -DFORRISCOS -DUIDGID_NOT_16BIT -DUNICODE_SUPPORT -DHAVE_DIRENT_
> H -DHAVE_TERMIOS_H zip.c
> cc -c -I. -DUNIX -DFORRISCOS -DUIDGID_NOT_16BIT -DUNICODE_SUPPORT -DHAVE_DIRENT_
> H -DHAVE_TERMIOS_H zipfile.c
> In file included from unix/osdep.h:31:0,
> from tailor.h:93,
> from zip.h:88,
> from zip.c:16:
> /home/matthew/gccsdk/cross/lib/gcc/arm-unknown-riscos/4.7.4/../../../../arm-unkn
> own-riscos/include/sys/stat.h:117:1: error: redefinition of 'struct stat'
> In file included from acorn/osdep.h:9:0,
> from tailor.h:80,
> from zip.h:88,
> from zip.c:16:
> acorn/riscos.h:30:8: note: originally defined here

> etc.

> Any further ideas? I'm happy to try wiping it all and starting from scratch
> with the GCC stable compiler instead, or to post any log files that might
> help.

native-zip should be getting built by the host compiler but instead it
looks like the autobuilder installed compiler is being used.
Have you set any envionment variables? Does echo $CC output anything.
Are any of the autobuilder folders in your PATH? e.g.
/home/matthew/gccsdk/env.

> That's impressive! We haven't seen any MMO files yet. We came across
> gpsbabel because another user has a GPS device that produces FIT files, and
> there is a NetSurf-friendly online converter tool based on GPSBabel. (The
> next release of RiscOSM will handle FIT files natively.)

> I thought it would be useful to have GPSBabel available on RISC OS if it
> wasn't too difficult.

You can download gpsbabel from
http://www.riscosports.co.uk/gpsbabel_1.5.3-1.zip. It also depends on
Qt5Base being installed. If you use PackMan to install it the rest of the
dependencies will be installed as well.


_______________________________________________
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] Problem with autobuilder (newbie questions)

In message <560575292cchrisg@care4free.net>
on 28 Jan 2017 Chris Gransden wrote:

> In article <eb913a0556.Matthew@sinenomine.freeserve.co.uk>,
> Matthew Phillips <spam2011m@yahoo.co.uk> wrote:
>
> > I then tried
> > ./build -d cli/gpsbabel
>
> It looks like you're running the build from the autobuilder folder which is
> why you're getting errors.
> First create a build folder in the gccsdk folder then create a file in the
> build folder called build-setvars.
> Put the following in this file. Change the path to suit where you have
> installed the autobuilder,
>
> GCCSDK_INSTALL_CROSSBIN=/home/user/gccsdk/cross/bin
> GCCSDK_INSTALL_ENV=/home/user/gccsdk/env
> RO_SHAREDLIBS=yes
> AB_ELFBUILD=yes
>
> Then from within the build folder use the command,
>
> ./autobuilder/build -v gpsbabel
>
> This ensures all the autobuilder scripts and libraries are found in the
> correct place otherwise the host libraries and scripts may be used which
> can cause the errors you are getting.

Thanks for your interest, Chris. It would be good to have gpsbabel available
to download. I would like to try getting to grips with the autobuilder as
well, though, as I had a couple of other projects in mind and was trying
gpsbabel as a possibly straightforward first project to kick off the learning
process!

The rest of the message is about the latest attempt. I am still getting
similar errors after following your instructions. I'd be grateful for any
further help or ideas.

Yes, I was running the autobuilder from the wrong place. I was misled by step
5 in the "Executive summary" on
http://www.riscos.info/index.php/Autobuilder_development

I have now deleted various files and directories that were created inside
autobuilder as a consequence, such as native-zip, Sources-stable-contrib.gz
and so on.

I created ~/gccsdk/build as you instructed and the build-setvars:
GCCSDK_INSTALL_CROSSBIN=/home/matthew/gccsdk/cross/bin
GCCSDK_INSTALL_ENV=/home/matthew/gccsdk/env
RO_SHAREDLIBS=yes
AB_ELFBUILD=yes

I re-read the instructions on http://www.riscos.info/index.php/Using_GCCSDK
-- I am pretty sure I built the development gccsdk using these instructions:

$ mkdir gccsdk
$ cd gccsdk
$ svn co svn://svn.riscos.info/gccsdk/trunk/autobuilder/ autobuilder
$ svn co svn://svn.riscos.info/gccsdk/trunk/gcc4/ gcc4
$ cd gcc4
$ ./build-world

I thought this time I would just try building wget, the example given on the
http://www.riscos.info/index.php/Using_GCCSDK web page. I did

$ cd build
$ ../autobuilder/build -v wget

as instructed. As with gpsbabel the first thing it tries to do is fetch and
compile native-zip. Again, after doing the patching, make starts and we get
the same errors as before:

make[1]: Entering directory `/home/matthew/gccsdk/build/native-zip/zip30'
cc -c -I. -DUNIX -DFORRISCOS -DUIDGID_NOT_16BIT -DUNICODE_SUPPORT -DHAVE_DIRENT_
H -DHAVE_TERMIOS_H zip.c
cc -c -I. -DUNIX -DFORRISCOS -DUIDGID_NOT_16BIT -DUNICODE_SUPPORT -DHAVE_DIRENT_
H -DHAVE_TERMIOS_H zipfile.c
In file included from unix/osdep.h:31:0,
from tailor.h:93,
from zip.h:88,
from zip.c:16:
/home/matthew/gccsdk/cross/lib/gcc/arm-unknown-riscos/4.7.4/../../../../arm-unkn
own-riscos/include/sys/stat.h:117:1: error: redefinition of 'struct stat'
In file included from acorn/osdep.h:9:0,
from tailor.h:80,
from zip.h:88,
from zip.c:16:
acorn/riscos.h:30:8: note: originally defined here

etc.

Any further ideas? I'm happy to try wiping it all and starting from scratch
with the GCC stable compiler instead, or to post any log files that might
help.

> I've got a port of 1.5.3 which successfully converted mmo to gpx and loaded
> OK into RiscOSM. I can make it available for download.

That's impressive! We haven't seen any MMO files yet. We came across
gpsbabel because another user has a GPS device that produces FIT files, and
there is a NetSurf-friendly online converter tool based on GPSBabel. (The
next release of RiscOSM will handle FIT files natively.)

I thought it would be useful to have GPSBabel available on RISC OS if it
wasn't too difficult.

--
Matthew Phillips
Durham

_______________________________________________
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] Problem with autobuilder (newbie questions)

In article <eb913a0556.Matthew@sinenomine.freeserve.co.uk>,
Matthew Phillips <spam2011m@yahoo.co.uk> wrote:
> ./build -d cli/gpsbabel

gpsbabel has now been added to the 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

Re: [gccsdk] GCC for RISC OS 4.7.4 release 3

In article
<LOXP123MB13184E8FA8FCF8721E824C77F0760@LOXP123MB1318.GBRP123.PROD.OUTLOOK.COM>,
alan buckley <alan_baa@hotmail.com> wrote:
> I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear in the PackMan package list and the riscos info packages website tomorrow. The direct download should be available now.

> I'll do a proper release announcement early next week and also update the GCC Releases page at riscos info. Is there anything else that needs updating?

> Regards,
> Alan

gcc -v and g++ -v are still showing as release 2.


_______________________________________________
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] Problem with autobuilder (newbie questions)

In article <eb913a0556.Matthew@sinenomine.freeserve.co.uk>,
Matthew Phillips <spam2011m@yahoo.co.uk> wrote:

> I then tried
> ./build -d cli/gpsbabel

It looks like you're running the build from the autobuilder folder which is
why you're getting errors.
First create a build folder in the gccsdk folder then create a file in the
build folder called build-setvars.
Put the following in this file. Change the path to suit where you have
installed the autobuilder,

GCCSDK_INSTALL_CROSSBIN=/home/user/gccsdk/cross/bin
GCCSDK_INSTALL_ENV=/home/user/gccsdk/env
RO_SHAREDLIBS=yes
AB_ELFBUILD=yes

Then from within the build folder use the command,

../autobuilder/build -v gpsbabel

This ensures all the autobuilder scripts and libraries are found in the
correct place otherwise the host libraries and scripts may be used which
can cause the errors you are getting.

The command line gpgbabel needs some of the Qt libraries installed in the
autobuilder to build successfully.

I've got a port of 1.5.3 which successfully converted mmo to gpx and loaded
OK into RiscOSM. I can make it available for download.


_______________________________________________
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 for RISC OS 4.7.4 release 3

alan buckley, on 27 Jan, wrote:

> I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear
> in the PackMan package list and the riscos info packages website tomorrow.
> The direct download should be available now.

To confirm that that is good on the Raspberry Pi 3.

> I'll do a proper release announcement early next week and also update the
> GCC Releases page at riscos info.

> Is there anything else that needs updating?

DeskLib? That built but as I recall the file formats were linux style rather
that RISC OS, header/h not h.header, manually corrected here.

HTH. Many Thanks.
--
David Pitt

_______________________________________________
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] Problem with autobuilder (newbie questions)

I am trying the autobuilder for the first time. I am using Linux Mint and I
seem to have downloaded and compiled GCCSDK successfully. I built a little
program of "Hello, world" complexity,compiled it, and ran it successfully on
RISC OS.

Next I tried to set up a new package for the autobuilder to build. I am
trying gpsbabel, which converts between various geodata file formats. I'm not
sure how far I'll get as I'll need to avoid compiling the Qt-based GUI and
just do the command-line version. Anyhow, I set up a very basic setvars whose
only active lines are:

AB_GIT=https://github.com/gpsbabel/gpsbabel.git
AB_ROVERSION=1
AB_PRECONF=:
AB_INSTALL=no

(I realise this may well not be sufficient.)

I then tried
./build -d cli/gpsbabel

It started building support/native-zip for me, and succeeded with various
patches. Then it started on configure, and got through that. Then make was
invoked, entering ./autobuilder/native-zip/zip30

I was expecting it to be built inside ./autobuilder/support/native-zip rather
than ./autobuilder/native-zip -- I don't know if that is a problem.

Then the C compiler started giving errors:

In file included from unix/osdep.h:31:0,
from tailor.h:93,
from zip.h:88,
from zipfile.c:16:
/home/matthew/gccsdk/cross/lib/gcc/arm-unknown-riscos/4.7.4/../../../../arm-unkn
own-riscos/include/sys/stat.h:117:1: error: redefinition of 'struct stat'
In file included from acorn/osdep.h:9:0,
from tailor.h:80,
from zip.h:88,
from zipfile.c:16:
acorn/riscos.h:30:8: note: originally defined here

There were a couple like that and also "conflicting types for 'stat'", and a complaint about chmod:

In file included from unix/osdep.h:31:0,
from tailor.h:93,
from zip.h:88,
from zip.c:16:
/home/matthew/gccsdk/cross/lib/gcc/arm-unknown-riscos/4.7.4/../../../../arm-unkn
own-riscos/include/sys/stat.h:241:12: error: conflicting types for 'chmod'
In file included from acorn/osdep.h:9:0,
from tailor.h:80,
from zip.h:88,
from zip.c:16:
acorn/riscos.h:106:5: note: previous declaration of 'chmod' was here

and then it finished:

make[1]: *** [zipfile.o] Error 1
make[1]: *** Waiting for unfinished jobs....
make[1]: *** [zip.o] Error 1
make[1]: Leaving directory `/home/matthew/gccsdk/autobuilder/native-zip/zip30'
make: *** [generic] Error 2

I wonder whether the code has changed and needs more patching somehow? There
is something in configure to check if a 64-bit stat is supported and the
struct definitions differ if that's available. Zip contains its own RISC OS
specific header file containing stat.

I assume I need to get this fixed before I can proceed to the next stage of
failing abysmally to cross-compile gpsbabel. Can anyone suggest what to do
next?

Out of interest I've had a look at
http://ci.riscos.info/job/packages/job/native-zip/5/console This also failed,
but for different reasons. In the configure stage it had:

Check for Large File Support
off_t is 8 bytes
-- yes we have Large File Support!

whereas on my machine I got

Check for Large File Support
./conftest: 1: ./conftest: Syntax error: word unexpected (expecting ")")
-- no Large File Support - no 64-bit stat

It looks like the Large File Support decision is not being made correctly, as
I would have expected it to be "no" for RISC OS. This means on Jenkins
-DLARGE_FILE_SUPPORT is passed to the C compiler. For whatever reason that
seems to avoid the duplicated stat definition problem, but the Jenkins build
fails at the end with

Archiving artifacts
ERROR: No artifacts found that match the file pattern "autobuilder/autobuilder_packages/**, autobuilder/autobuilder_libraries/**". Configuration error?
Build step 'Archive the artifacts' changed build result to FAILURE
[WS-CLEANUP] Deleting project workspace...[WS-CLEANUP] done
Finished: FAILURE


Thanks,

Matthew

--
Matthew Phillips
Durham

_______________________________________________
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 for RISC OS 4.7.4 release 3

On 27/01/17 13:18, alan buckley wrote:
> I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear
> in the PackMan package list and the riscos info packages website
> tomorrow. The direct download should be available now.
>
>
>
> I'll do a proper release announcement early next week and also update
> the GCC Releases page at riscos info. Is there anything else that needs
> updating?

Have you updated SharedLibs-C2 containing the libunixlib shared library?

Lee.


_______________________________________________
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 for RISC OS 4.7.4 release 3

I've uploaded release 3 of GCC 4.7.4 for RISC OS today. It should appear in the PackMan package list and the riscos info packages website tomorrow. The direct download should be available now.

 

I'll do a proper release announcement early next week and also update the GCC Releases page at riscos info. Is there anything else that needs updating?

 

Regards,

Alan

Re: RESOLVED: [Was] Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <5603e20e5ctim@timil.com>,
Tim Hill <tim@timil.com> wrote:
> In article <5603a241a4brian.jordan9@btinternet.com>, Brian Jordan
> <brian.jordan9@btinternet.com> wrote:

[Snip]

> <link type="text/css" rel="StyleSheet" href="style.css?version=latest"
> />

> Appending any fake variable to the URL forces it to be loaded every
> time the page is viewed and not just when the browser feels like it.

You live and learn! Many thanks for this tip.

--
_____________________________________________________________________

Brian Jordan
Virtual RPC-AdjustSA on Windows 10 Pro 64-bit
RISC OS 6.20
_____________________________________________________________________


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: RESOLVED: [Was] Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <5603a241a4brian.jordan9@btinternet.com>, Brian Jordan
<brian.jordan9@btinternet.com> wrote:

[Snip]

> I suspect two versions of the relevant style sheet were
> being stored and that one was invoked when I loaded the online version
> of the site and the other when I loaded the local version.

[Snip]


There's that 'thing' when changes have been made to a stylesheet but a
browser doesn't load it because it already has the old one. Which is what
may have happened here. To prevent it, have something like this in your
Head:

<link type="text/css" rel="StyleSheet" href="style.css?version=latest" />

Appending any fake variable to the URL forces it to be loaded every time
the page is viewed and not just when the browser feels like it.

--

Tim Hill

timil.com : tjrh.eu : butterwick.eu : blue-bike.uk : youngtheatre.co.uk

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <56039cd7babrian.jordan9@btinternet.com>, Brian Jordan
<brian.jordan9@btinternet.com> wrote:
> FWIW the site was designed and laid out
> using RISC OS and NetSurf in particular.

I used to do that until trying to make my sites more mobile compliant
for the 25% of mobile visitors and I discovered Flex containers were a
handy cross browser solution.
Flex isn't implemented in NetSurf, of course. :-(

--

Tim Hill

timil.com : tjrh.eu : butterwick.eu : blue-bike.uk : youngtheatre.co.uk

RESOLVED: [Was] Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <56039cd7babrian.jordan9@btinternet.com>,
Brian Jordan <brian.jordan9@btinternet.com> wrote:
> In article <56033109cfbrian.jordan9@btinternet.com>,
> Brian Jordan <brian.jordan9@btinternet.com> wrote:

[Snip]

>Today I discovered that NetSurf on the VA displays the site as intended
> when it is loaded from my local mirror although it still doesn't work
> properly with files from the www. "Aha!" you say "go through the site
> until you find the discrepancy, it's probably a style sheet." "But," I
> reply,"it still renders correctly on the same version of NetSurf on the
> Pi when viewed via the www." Mystified and digging deeper.

[Snip]

After clearing out
HostFS::HardDisc4.$.!Boot.Choices.Users.Single.WWW.NetSurf
the problem has gone away. I suspect two versions of the relevant style
sheet were being stored and that one was invoked when I loaded the online
version of the site and the other when I loaded the local version. Anyway
the site now displays as I intended in NetSurf #3966 whether it is
downloaded or loaded locally both on both platforms and with both.
versions of RISC OS.

--
_____________________________________________________________________

Brian Jordan
Virtual RPC-AdjustSA on Windows 10 Pro 64-bit
RISC OS 6.20
_____________________________________________________________________


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <56033109cfbrian.jordan9@btinternet.com>,
Brian Jordan <brian.jordan9@btinternet.com> wrote:

[Snip]

>If there is an inconsistency relating to OS versions I will file a
> report to the NetSurf developers, if not I shall have to dig a lot
> deeper here. Thanks B

[Snip]

Thanks all for your responses. I have been looking more closely at this
and am beginning to think the problem is of my own making so, until I
have probed further, I shall hold fire on sending anything to the
developers. Today I discovered that NetSurf on the VA displays the site
as intended when it is loaded from my local mirror although it still
doesn't work properly with files from the www. "Aha!" you say "go through
the site until you find the discrepancy, it's probably a style sheet."
"But," I reply,"it still renders correctly on the same version of NetSurf
on the Pi when viewed via the www."
Mystified and digging deeper.

PS. Jim, pinch away the design ideas aren't originally mine - I just
pinched and modified them. FWIW the site was designed and laid out using
RISC OS and NetSurf in particular.

--
_____________________________________________________________________

Brian Jordan
Virtual RPC-AdjustSA on Windows 10 Pro 64-bit
RISC OS 6.20
_____________________________________________________________________


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

On Tue, Jan 24, 2017 at 12:42:08AM +0000, Brian Jordan wrote:
> The home page of www.clubmans.org.uk has a news area in the lower part.
> It is intended that this is a scrolling <div> and in all mainstream
> browsers I have tried this is the case [1]. it is also the case in
> NetSurf #3966 running under RISC OS 5.23 on my Raspberry Pi. However,
> using the same browser to view the same page on this emulator with RISC
> OS Select 6.20 the <div> overflows. The most likely explanation is, as
> ever, that this is a local issue and that I have missed the obvious but
> to help define the problem I wonder if some of you would take a look and
> let me know what you see and the OS version you are using? If there is an
> inconsistency relating to OS versions I will file a report to the NetSurf
> developers, if not I shall have to dig a lot deeper here.

My first and totally uninformed guess is the different FontManagers and
fonts in use. Could you perhaps create a but on the tracker, and attach
screenshots of both and the config files for both?

B.

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

Brian Jordan wrote on 24 Jan:
> The home page of www.clubmans.org.uk has a news area in the lower
> part. It is intended that this is a scrolling <div> and in all
> mainstream browsers I have tried this is the case [1]. it is also the
> case in NetSurf #3966 running under RISC OS 5.23 on my Raspberry Pi.
> ...

Here, it looks as you describe it should. ArmX6 with Ro 5.23 and
Netsurf #3955.

(Might pinch some of your design ideas!)

--
Jim Nagel www.archivemag.co.uk

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

On 24/01/17 00:42, Brian Jordan wrote:
> The home page of www.clubmans.org.uk has a news area in the lower part.
> It is intended that this is a scrolling <div> and in all mainstream
> browsers I have tried this is the case [1]. it is also the case in
> NetSurf #3966 running under RISC OS 5.23 on my Raspberry Pi. However,
> using the same browser to view the same page on this emulator with RISC
> OS Select 6.20 the <div> overflows.

It might be a difference in config on the two systems. Try visiting the
URL "about:config" on the two systems, and look for any difference in
the options marked as having "user" provenance.

Either that, or you might have a user stylesheet on the system causing
the problem.

Cheers,

--

Michael Drake http://www.codethink.co.uk/

Re: Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

In article <56033109cfbrian.jordan9@btinternet.com>,
Brian Jordan <brian.jordan9@btinternet.com> wrote:
> The home page of www.clubmans.org.uk has a news area in the lower part.
> It is intended that this is a scrolling <div> and in all mainstream
> browsers I have tried this is the case [1]. it is also the case in
> NetSurf #3966 running under RISC OS 5.23 on my Raspberry Pi. However,
> using the same browser to view the same page on this emulator with RISC
> OS Select 6.20 the <div> overflows. The most likely explanation is, as
> ever, that this is a local issue and that I have missed the obvious but
> to help define the problem I wonder if some of you would take a look and
> let me know what you see and the OS version you are using? If there is an
> inconsistency relating to OS versions I will file a report to the NetSurf
> developers, if not I shall have to dig a lot deeper here.
> Thanks
> B

> [1] HTML tidy reports the page as <!DOCTYPE html PUBLIC "-//W3C//DTD HTML
> 4.01 Transitional//EN"> and I test in the Windows browsers because these
> are what the majority of visitors us.

On VRPC RISC OS 4.02 with NetSurf #3966 I see the lower section as
scrolling information as you describe. Can't see any problem here.

Netsurf inconsistent behaviour between RISC OS 5.23 and 6.20

The home page of www.clubmans.org.uk has a news area in the lower part.
It is intended that this is a scrolling <div> and in all mainstream
browsers I have tried this is the case [1]. it is also the case in
NetSurf #3966 running under RISC OS 5.23 on my Raspberry Pi. However,
using the same browser to view the same page on this emulator with RISC
OS Select 6.20 the <div> overflows. The most likely explanation is, as
ever, that this is a local issue and that I have missed the obvious but
to help define the problem I wonder if some of you would take a look and
let me know what you see and the OS version you are using? If there is an
inconsistency relating to OS versions I will file a report to the NetSurf
developers, if not I shall have to dig a lot deeper here.
Thanks
B

[1] HTML tidy reports the page as <!DOCTYPE html PUBLIC "-//W3C//DTD HTML
4.01 Transitional//EN"> and I test in the Windows browsers because these
are what the majority of visitors us.

--
_____________________________________________________________________

Brian Jordan
Virtual RPC-AdjustSA on Windows 10 Pro 64-bit
RISC OS 6.20
_____________________________________________________________________


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: [gccsdk] GCCSDK Raspberry Pi 3 status

[Reposted as it didn't make it to the list last time]

On 09/01/17 13:10, alan buckley wrote:
> I've just seen something in the ROOL forums implying that the SharedULib
> needs updating for the Raspberry Pi 3. Is this correct? I thought that
> programs needed recompiling, but the SharedULib module was OK.

There seems to be some confusion about what SharedULib is. There are 2
separate entities; the module, SharedUnixLibrary and the library
libunixlib which exists as an ELF shared library and as a static
archive.
The problem with SWP on the RPi3 existed in the ELF shared library and
in the static archive and has been fixed. There was no problem in the
module.
This means that programs linked to the static archive will have to be
re-linked to the fixed version. Any programs that are dynamically
linked to the ELF shared library should work correctly once libunixlib
in !SharedLibs is replaced with a fixed version.

> What is the current status of GCCSDK for the Raspberry Pi 3?

Providing you are using a recent version then GCC and any code generated
by it should work on the Raspberry Pi 3.

> Is it worth creating a new RISC OS version of !GCC so it can be used by
> Raspberry Pi 3 users? Or is there more to do?

The version packaged on riscos.info does not include the fixes, so
could do with being updated.

Lee.


_______________________________________________
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] GCCSDK Raspberry Pi 3 status

alan buckley, on 16 Jan, wrote:

> David Pitt wrote on 09 January 2017 15:57
>
> > alan buckley, on 9 Jan, wrote:

[snip]

> > > What is the current status of GCCSDK for the Raspberry Pi 3?
>
> > I have a locally built !GCC from June 2006 that is RPi3 compatible, it
> > being built after the SWP fix was checked in. The build also generated
> > an updated !SharedLibs which is definitely required and a SharedUlib
> > which might not be.
>
> > > Is it worth creating a new RISC OS version of !GCC so it can be used
> > > by Raspberry Pi 3 users? Or is there more to do?
>
> > I have not used !GCC extensively but as far as I am aware the build that
> > I have is good on the RPi3.
>
> I'm really trying to determine if I (or someone) should produce a new
> version of !GCC for the riscos info website or if it needs something else.

> > The problem on the RPi3 is more extensive than just !GCC, !PDF is
> > another casualty. Again I have a local RPi3 good build.
>
> Did you use the GCCSDK autobuilder to build !PDF? If so did you have to
> make any changes to the build scripts?

The autobuilder was used.

!PDF builds as is, and is good on the RPi3. Revision 7084 was required.

> If any of the riscos-info stuff is failing on the RPi3 and will build OK
> from the autobuilder, I can build a new version and upload it. Just let me
> know what the packages are, and please test them after they are uploaded
> as I don't have a RPi3. Unless there are quick fixes to the build I won't
> have time to sort them out though.

Wget and !GCC also build and are OK on my RPi3.

> > All of this was discussed at the time on the ROOL forum.
>
> I've seen the ROOL forum discussions, I'm trying to get input from The
> GCCSDK developers as they will have a better idea of the current state of
> play.

--
David Pitt

_______________________________________________
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] GCCSDK Raspberry Pi 3 status

David Pitt wrote on  09 January 2017 15:57

 

> alan buckley, on 9 Jan, wrote:

 

> > I've just seen something in the ROOL forums implying that the SharedULib

> > needs updating for the Raspberry Pi 3. Is this correct? I thought that

> > programs needed recompiling, but the SharedULib module was OK.

 

> Hmm!! I thought that about the SharedULib module too, but others would seem

> to be saying otherwise.

 

> > What is the current status of GCCSDK for the Raspberry Pi 3?

 

> I have a locally built !GCC from June 2006 that is RPi3 compatible, it being

> built after the SWP fix was checked in. The build also generated an updated

>!SharedLibs which is definitely required and a SharedUlib which might not

> be.

 

> > Is it worth creating a new RISC OS version of !GCC so it can be used by

> > Raspberry Pi 3 users? Or is there more to do?

 

> I have not used !GCC extensively but as far as I am aware the build that I

> have is good on the RPi3.

 

I'm really trying to determine if I (or someone) should produce a new

version of !GCC for the riscos info website or if it needs something

else.

 

> The problem on the RPi3 is more extensive than just !GCC, !PDF is another

> casualty. Again I have a local RPi3 good build.

 

Did you use the GCCSDK autobuilder to build !PDF? If so did you have to

make any changes to the build scripts?

 

If any of the riscos-info stuff is failing on the RPi3 and will build OK

from the autobuilder, I can build a new version and upload it. Just

let me know what the packages are, and please test them after

they are uploaded as I don't have a RPi3.

Unless there are quick fixes to the build I won't have time to sort them

out though.

 

> All of this was discussed at the time on the ROOL forum.

 

I've seen the ROOL forum discussions, I'm trying to get input from

The GCCSDK developers as they will have a better idea of the

current state of play.

 

Thanks,

Alan

Bug 2506: can't move hotlist, history, cookies windows off screen

diff --git a/frontends/riscos/templates/de b/frontends/riscos/templates/de
index 0fb4a9d..a4ec1f4 100644
--- a/frontends/riscos/templates/de
+++ b/frontends/riscos/templates/de
@@ -2542,7 +2542,7 @@ wimp_window {
xscroll:0
yscroll:0
next:wimp_TOP
- window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT
+ window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT | wimp_WINDOW_NO_BOUNDS
title_fg:wimp_COLOUR_BLACK
title_bg:wimp_COLOUR_LIGHT_GREY
work_fg:wimp_COLOUR_BLACK
diff --git a/frontends/riscos/templates/en b/frontends/riscos/templates/en
index 7746b86..25be55f 100644
--- a/frontends/riscos/templates/en
+++ b/frontends/riscos/templates/en
@@ -2750,7 +2750,7 @@ wimp_window {
xscroll:0
yscroll:0
next:wimp_TOP
- window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT
+ window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT | wimp_WINDOW_NO_BOUNDS
title_fg:wimp_COLOUR_BLACK
title_bg:wimp_COLOUR_LIGHT_GREY
work_fg:wimp_COLOUR_BLACK
diff --git a/frontends/riscos/templates/fr b/frontends/riscos/templates/fr
index 6779264..fafe0ce 100644
--- a/frontends/riscos/templates/fr
+++ b/frontends/riscos/templates/fr
@@ -2554,7 +2554,7 @@ wimp_window {
xscroll:0
yscroll:0
next:wimp_TOP
- window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT
+ window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT | wimp_WINDOW_NO_BOUNDS
title_fg:wimp_COLOUR_BLACK
title_bg:wimp_COLOUR_LIGHT_GREY
work_fg:wimp_COLOUR_BLACK
diff --git a/frontends/riscos/templates/nl b/frontends/riscos/templates/nl
index 704206f..5ed9cac 100644
--- a/frontends/riscos/templates/nl
+++ b/frontends/riscos/templates/nl
@@ -2796,7 +2796,7 @@ wimp_window {
xscroll:0
yscroll:0
next:wimp_TOP
- window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT
+ window_flags:wimp_WINDOW_MOVEABLE | wimp_WINDOW_SCROLL_REPEAT | wimp_WINDOW_IGNORE_XEXTENT | wimp_WINDOW_IGNORE_YEXTENT | wimp_WINDOW_BOUNDED_ONCE | wimp_WINDOW_BACK_ICON | wimp_WINDOW_CLOSE_ICON | wimp_WINDOW_TITLE_ICON | wimp_WINDOW_TOGGLE_ICON | wimp_WINDOW_VSCROLL | wimp_WINDOW_SIZE_ICON | wimp_WINDOW_HSCROLL | wimp_WINDOW_NEW_FORMAT | wimp_WINDOW_NO_BOUNDS
title_fg:wimp_COLOUR_BLACK
title_bg:wimp_COLOUR_LIGHT_GREY
work_fg:wimp_COLOUR_BLACK
These windows are based on the "tree" template, which does not have the
wimp_WINDOW_NO_BOUNDS flag bit set (and there is no code to set it after
creating the window). The attached patches OR the flag bit into each of
the four template definitions that we have (de, en, fr, nl).

The bug cropped up when the tree view code was replaced.

It's my belief that this flag bit /should/ be in the templates - it's
part of the fundamental window behaviour definitions, which is what
RISC OS has templates for. Code should only be altering flag bits if
they need to be altered back and forth dynamically, e.g. to grey icons
out. I cannot see any reason why the ability to drag windows off
screen would ever need to be altered dynamically. Please speak up
if you believe differently!

OTOH if the team agrees with me, then perhaps there should be another
stage of moving this flag bit into all the window definitions. Again,
please speak up!

When applying these patches, please bear in mind that some of the
files contain top bit set characters for the non-English languages,
which may upset your text editor. The files in question are all
textual.

Dave

Re: [gccsdk] GCCSDK Raspberry Pi 3 status

On 09/01/17 13:10, alan buckley wrote:
> I've just seen something in the ROOL forums implying that the SharedULib
> needs updating for the Raspberry Pi 3. Is this correct? I thought that
> programs needed recompiling, but the SharedULib module was OK.

There seems to be some confusion about what SharedULib is. There are 2
separate entities; the module, SharedUnixLibrary and the library
libunixlib which exists as an ELF shared library and as a static
archive.
The problem with SWP on the RPi3 existed in the ELF shared library and
in the static archive and has been fixed. There was no problem in the
module.
This means that programs linked to the static archive will have to be
re-linked to the fixed version. Any programs that are dynamically
linked to the ELF shared library should work correctly once libunixlib
in !SharedLibs is replaced with a fixed version.

> What is the current status of GCCSDK for the Raspberry Pi 3?

Providing you are using a recent version then GCC and any code generated
by it should work on the Raspberry Pi 3.

> Is it worth creating a new RISC OS version of !GCC so it can be used by
> Raspberry Pi 3 users? Or is there more to do?

The version packaged on riscos.info does not include the fixes, so
could do with being updated.

Lee.


_______________________________________________
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] GCCSDK Raspberry Pi 3 status

alan buckley, on 9 Jan, wrote:

> I've just seen something in the ROOL forums implying that the SharedULib
> needs updating for the Raspberry Pi 3. Is this correct? I thought that
> programs needed recompiling, but the SharedULib module was OK.

Hmm!! I thought that about the SharedULib module too, but others would seem
to be saying otherwise.

> What is the current status of GCCSDK for the Raspberry Pi 3?

I have a locally built !GCC from June 2006 that is RPi3 compatible, it being
built after the SWP fix was checked in. The build also generated an updated
!SharedLibs which is definitely required and a SharedUlib which might not
be.

> Is it worth creating a new RISC OS version of !GCC so it can be used by
> Raspberry Pi 3 users? Or is there more to do?

I have not used !GCC extensively but as far as I am aware the build that I
have is good on the RPi3.

The problem on the RPi3 is more extensive than just !GCC, !PDF is another
casualty. Again I have a local RPi3 good build.

All of this was discussed at the time on the ROOL forum.

--
David Pitt

_______________________________________________
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] GCCSDK Raspberry Pi 3 status

I've just seen something in the ROOL forums implying that the SharedULib needs updating for the Raspberry Pi 3. Is this correct? I thought that programs needed recompiling, but the SharedULib module was OK.

 

What is the current status of GCCSDK for the Raspberry Pi 3?

 

Is it worth creating a new RISC OS version of !GCC so it can be used by Raspberry Pi 3 users? Or is there more to do?

 

Thanks,

Alan

 

Re: Non-displaying page

In article <d06acb50-44e8-3d6a-8e5b-af68cf28fc1c@codethink.co.uk>,
Michael Drake <michael.drake@codethink.co.uk> wrote:
> On 05/01/17 13:27, Richard Torrens (lists) wrote:
> > Is this a bug or a style fault on the page?

> The "body { visibility: hidden !important; }" style means "hide the
> whole page from view". There is some JavaScript to unset it,
> revealing the page.

> It's daft, because the page won't display at all in any browser with
> JavaScript disabled. In NetSurf, it doesn't work with JavaScript
> enabled because our JavaScript support is far from complete.

Thanks for the explanation. I don't know how can you hope to keep up with
such daft www programmers!

--
Richard Torrens.
http://www.Torrens.org.uk for genealogy, natural history, wild food, walks, cats
and more!

Re: Web page magnification

On 2 Jan, brian.jordan9@btinternet.com typed:

> <F11> when in Netsurf will bring up the "Scale view" dialogue; I
> imagine you know this already but to make this the default you need to
> <Menu> over the NetSurf window and in the "Display" sub menu click on
> "Save as default". Good luck!

Yes, and the value is stored in text file:

!Boot.Choices.WWW.NetSurf.Choices

by the defualt % scale in the line:

scale:100

--
Bernard

Re: Non-displaying page

On 05/01/17 13:27, Richard Torrens (lists) wrote:
> Is this a bug or a style fault on the page?

The "body { visibility: hidden !important; }" style means "hide the
whole page from view". There is some JavaScript to unset it,
revealing the page.

It's daft, because the page won't display at all in any browser with
JavaScript disabled. In NetSurf, it doesn't work with JavaScript
enabled because our JavaScript support is far from complete.

--

Michael Drake http://www.codethink.co.uk/

Non-displaying page

Is this a bug or a style fault on the page?

http://www.seattletimes.com/seattle-news/health/popular-heartburn-drugs-may-increase-esophageal-cancer-risk/

The first <style that occurs is

<style id='chartbeat-flicker-control-style' type='text/css'>
/* Will be removed by ChartBeat's async flicker control below */
body { visibility: hidden !important; }
</style>

Remove this and the page displays. So I suspect it may not be a bug?

--
Richard Torrens.
http://www.Torrens.org.uk for genealogy, natural history, wild food, walks, cats
and more!

Web page magnification

Thank you all for your helpful replies. They are very useful.

--
David Wild using RISC OS on broadband
www.davidhwild.me.uk

Re: Web page magnification

In article <55f8413d86dhwild@talktalk.net>, David H Wild
<dhwild@talktalk.net> wrote:
> I am having problems reading because of a cataract which needs removing,

> i find that I can read the screen fairly well if I increases the
> magnification to 150%. Is there any way of telling NetSurf that this is
> how I want all pages?

Not a solution to the default, but my ButtonBar for NetSurf has grow and
shrink buttons (because I couldn't ever remember the key short-cuts).

http://timil.com/riscos

--

Tim Hill

timil.com : tjrh.eu : butterwick.eu : blue-bike.uk : youngtheatre.co.uk

Re: Web page magnification

On 2 Jan 2017 Brian Jordan wrote:

> In article <55f8413d86dhwild@talktalk.net>,
> David H Wild <dhwild@talktalk.net> wrote:
>> I am having problems reading because of a cataract which needs removing,

>> i find that I can read the screen fairly well if I increases the
>> magnification to 150%. Is there any way of telling NetSurf that this is
>> how I want all pages?

> <F11> when in Netsurf will bring up the "Scale view" dialogue; I imagine
> you know this already but to make this the default you need to <Menu>
> over the NetSurf window and in the "Display" sub menu click on "Save as
> default". Good luck!

Yes it does. There is an entry in choices: Scale:150 . Ctrl-W is a
quicker route to 150%. It's useful for enlarging photos on the screen
(goes to 150 - 200 - 300% and more). Ctrl-Q reduces the magnification
correspondingly.

--
Richard Porter http://www.minijem.plus.com/
Skype: minijem2 mailto:ricp@minijem.plus.com
I don't want a "user experience" - I just want stuff that works.

Re: Web page magnification

On 2 Jan 2017 David H Wild wrote:

> I am having problems reading because of a cataract which needs removing,

> i find that I can read the screen fairly well if I increases the
> magnification to 150%. Is there any way of telling NetSurf that this
> is how I want all pages?

Enter Ctrl-W on the keyboard. This sets the magnification to 150%.
The magnification persists if you follow links including in a new
window, but not if you open a new window from the icon bar.

I can't see anything in Choices that sets the default magnification
(perhaps a feature request would be in order) but increasing the
default and minimum font sizes might be help.

--
Richard Porter http://www.minijem.plus.com/
Skype: minijem2 mailto:ricp@minijem.plus.com
I don't want a "user experience" - I just want stuff that works.

Re: Web page magnification

In article <55f8413d86dhwild@talktalk.net>,
David H Wild <dhwild@talktalk.net> wrote:
> I am having problems reading because of a cataract which needs removing,

> i find that I can read the screen fairly well if I increases the
> magnification to 150%. Is there any way of telling NetSurf that this is
> how I want all pages?

<F11> when in Netsurf will bring up the "Scale view" dialogue; I imagine
you know this already but to make this the default you need to <Menu>
over the NetSurf window and in the "Display" sub menu click on "Save as
default". Good luck!

--
_____________________________________________________________________

Brian Jordan
Virtual RPC-AdjustSA on Windows 10 Pro 64-bit
RISC OS 6.20
_____________________________________________________________________


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Web page magnification

I am having problems reading because of a cataract which needs removing,

i find that I can read the screen fairly well if I increases the
magnification to 150%. Is there any way of telling NetSurf that this is how
I want all pages?

Thanks.

--
David Wild using RISC OS on broadband
www.davidhwild.me.uk