Re: [Rpcemu] BASIC mouse-pointer issue

Thanks for the workaround!

On 22 January 2018 at 17:08, Peter Howkins <rpcemu.howkins@marutan.net> wrote:

On Sat, Jan 20, 2018 at 03:34:29PM +0000, J Percival wrote:
>    RPCEmu Version: 0.8.15 and later
>    Config: RPC-StrongARM/16MB/No VRAM/Sound Enabled
>    I'm not sure if this is an emulator bug or not but I've noticed that the
>    BASIC "MOUSE TO" command under RO3.71 changes the size of the pointer
>    instead of its location - specifically the 'x' argument has no effect and
>    the 'y' argument modifies the pointer's height.
>    To reproduce:
>    MOUSE ON:MOUSE TO 0,20

Thanks for this report, I've been able to reproduce it and am
investigating. Oddly the reduction in the size of the pointer is related
to the amount that is offscreen when moving TO the Y coord, e.g. MOUSE TO
0, 2 means you have an even smaller pointer.

This only affects 'Follow host mouse' mode, so if you need a temporary
workaround, use 'Capture Mouse' mode by deselecting 'Follow host mouse'
(0.8.100) or chosing Capture Mouse (0.8.15).

Peter


--
Peter Howkins
peter.howkins@marutan.net

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

Re: [Rpcemu] RPCEmu pre release test version 0.8.100

On Tue, Jan 02, 2018 at 03:01:37AM +0000, J Percival wrote:
> Hi, I did a little more experimentation to see exactly at what point
> RiscOS locks up when running the universal-boot-2 sequence under the
> recompiler.
> It seems to be on execution of !Boot.Utils.BootVars so probably a bug in
> the recompiler code.

To let you know we have reproduced this and are investigating. Initial
looks says it's not a quick fix (nor a regression in 0.8.100), as such
it's unlikely to be fixed before the next release version and the
workaround of choosing 'StrongARM' is probably a good idea here.

Peter

--
Peter Howkins
peter.howkins@marutan.net

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

Re: [Rpcemu] BASIC mouse-pointer issue

On Sat, Jan 20, 2018 at 03:34:29PM +0000, J Percival wrote:
> RPCEmu Version: 0.8.15 and later
> Config: RPC-StrongARM/16MB/No VRAM/Sound Enabled
> I'm not sure if this is an emulator bug or not but I've noticed that the
> BASIC "MOUSE TO" command under RO3.71 changes the size of the pointer
> instead of its location - specifically the 'x' argument has no effect and
> the 'y' argument modifies the pointer's height.
> To reproduce:
> MOUSE ON:MOUSE TO 0,20

Thanks for this report, I've been able to reproduce it and am
investigating. Oddly the reduction in the size of the pointer is related
to the amount that is offscreen when moving TO the Y coord, e.g. MOUSE TO
0, 2 means you have an even smaller pointer.

This only affects 'Follow host mouse' mode, so if you need a temporary
workaround, use 'Capture Mouse' mode by deselecting 'Follow host mouse'
(0.8.100) or chosing Capture Mouse (0.8.15).

Peter


--
Peter Howkins
peter.howkins@marutan.net

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

Re: [NetSurf 0002266]: URLs about:welcome, about:credits and about:licence don't load localized version when the corresponding files are available

On 21 Jan, NetSurf Bug Tracker <help@netsurf-browser.org> wrote:
> The following issue has been RESOLVED.
> ======================================================================
> http://bugs.netsurf-browser.org/mantis/view.php?id=2266
> ----------------------------------------------------------------------
> (0001714) John-Mark Bell (administrator) - 2018-01-21 14:29
> http://bugs.netsurf-browser.org/mantis/view.php?id=2266#c1714
> ----------------------------------------------------------------------
> Fixed in 4302. sprintf approach is fine, but I disliked the assumption that
> ro_gui_default_language() returned a string that was 2 characters long, so
> the preceding length calculation has been updated to take account of the
> actual length of the language identifier.

I'd, perhaps wrongly, assumed that ro_gui_default_language() always returned
an ISO3166-1 country code, and hence the 2 character length was OK to
hardcode.

Regardless your mods look fine to me - thanks for integrating this,
Sprow.

PS: Can't seem to add comments to tickets once they're resolved.

[Rpcemu] BASIC mouse-pointer issue

RPCEmu Version: 0.8.15 and later

Config: RPC-StrongARM/16MB/No VRAM/Sound Enabled

I'm not sure if this is an emulator bug or not but I've noticed that the BASIC "MOUSE TO" command under RO3.71 changes the size of the pointer instead of its location - specifically the 'x' argument has no effect and the 'y' argument modifies the pointer's height.

To reproduce:

MOUSE ON:MOUSE TO 0,20

Kind Regards,

James

[gccsdk] Special case made for zlib -requires DIR set

I posted this problem yesterday, but though the post in the archive for
January, I haven't received it back so I cant follow the same thread.
Is anyone receiving posts?

After searching the configure file for zlib.h, I found that they have made
a special case and it required --with-zlib=$GCCSDK_INSTALL_ENV otherwise
it wont look.
I have my doubts wether telegram-cli will be suitable for the autobuilder,
As I've run into a lot of things needing manual intervention in the
makefile.
I may need to add in libconfig to obtain the internal library makefiles,
if the CYGWIN readme is anything to go by.

Autobuilder has output the root makefile, but as is, it is not enough for
the build, missing rules and the LD variables are including native linux
path which I dont think will ever fly, so some editting is needed.

Thankfully the makefile is only one page in size.

Ron M.



_______________________________________________
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] zlib.h not found suitable

I'm having a go at getting telegram.cli working and hopefully in the
autobuilder.
A dependancy libevent from testing seems to have built OK, but would be
good if someone knowledgeable about pthreads etc could give it a look and
include it in the autobuilder.
There is an existing Event.h from OSLib which doesn't appear to conflict.

A problem with telegram-cli is it is only available from git as a zip and
on my local (ftp) network anyway, the autobuilder wont load it unless I
convert it to a tar/xx first.

Having met the dependancies I then get an error to do with zlib.h
It is in the same env/include directory as the new event.h and must be to
do with version, though the current 1.2.8 was installed recently.

The tail of the failure log:

checking for stdint.h... yes
checking for unistd.h... yes
checking event2/event.h usability... yes
checking event2/event.h presence... yes
checking for event2/event.h... yes
checking for openssl/ssl.h in /home/ron/gccsdk/env... yes
checking whether compiling and linking against OpenSSL works... yes
checking if zlib is wanted... yes
checking for inflateEnd in -lz... yes
checking zlib.h usability... no
checking zlib.h presence... no
checking for zlib.h... no
configure: error: No zlib found
Autobuilder: Running make command: /home/ron/gccsdk/env/ro-make
Autobuilder: No known build method

Ron M.



_______________________________________________
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