tilp-devel — This list is intended to TiLP's developers, contributors and advanced users.

Re: [TiLP-devel] TILP fails at "make install" {Scanned}

[Romain <ro...@li...>]

Hi,

> which means ?

The SVB was broken upon today. It's now fixed and the commit has been don=
e.

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

Re: [TiLP-devel] TILP fails at "make install" {Scanned}

[hibou <hi...@bo...>]

Romain Liévin a écrit :

>Hi,
>
>  
>
>>Making install in plugins
>>make[1]: Entering directory
>>`/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugins'
>>make[1]: *** Pas de règle pour fabriquer la cible « test », nécessaire
>>pour « install ». Arrêt.
>>make[1]: Leaving directory `/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugins'
>>make: *** [install-recursive] Erreur 1
>>    
>>
>
>You're right. The directories are not distributed with the tarball.
>
>Fixed but can't be commited.
>  
>
which means ?

I know what commiting is, but what are the consequences for the
versioning, date of new release ?

If it's not possible yet, can I have a patch ?

a+
hibou

Re: [TiLP-devel] TILP fails at "make install" {Scanned}

[Romain <ro...@li...>]

Hi,

> Making install in plugins
> make[1]: Entering directory
> `/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugins'
> make[1]: *** Pas de r=E8gle pour fabriquer la cible =AB test =BB, n=E9c=
essaire
> pour =AB install =BB. Arr=EAt.
> make[1]: Leaving directory `/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugi=
ns'
> make: *** [install-recursive] Erreur 1

You're right. The directories are not distributed with the tarball.

Fixed but can't be commited.

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

[TiLP-devel] TILP fails at "make install"

[hibou <hi...@bo...>]

Hello,

All libs are installed perfectly.
TiLP compiles well too. Just a little bug in make install
it ends with :

Making install in plugins
make[1]: Entering directory `/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugins'
make[1]: *** Pas de règle pour fabriquer la cible « test », nécessaire
pour « install ». Arrêt.
make[1]: Leaving directory `/home/hibou/dwnld/tilp-6.79/tilp-6.79/plugins'
make: *** [install-recursive] Erreur 1

a+
hibou

Re: [TiLP-devel] Request {Scanned}

[Romain <ro...@li...>]

Hi,

> yes, this would be great thanks.

They should arrive soon, I have to do package for JB.

>
> a+
> hibou
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Yahoo.
> Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
> Search APIs Find out how you can build Yahoo! directly into your own
> Applications - visit http://developer.yahoo.net/?fr=3Doffad-ysdn-ostg-q=
22005
> _______________________________________________
> TiLP-devel mailing list
> TiL...@li...
> https://lists.sourceforge.net/lists/listinfo/tilp-devel
>


--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

Re: [TiLP-devel] Request {Scanned}

[hibou <hi...@bo...>]

Romain Liévin wrote:

>Hi,
>
>  
>
>>I would like to help gentoo developper to make an ebuild for the last
>>tilp version (http://bugs.gentoo.org/show_bug.cgi?id=84763). But if you
>>update the source tarball without renaming it, the ebuild will want to
>>download the new tarball as thinking it's the old one : the md5 check
>>will fail, the ebuild would be broken.
>>So, can you tag the file archive containing the source of TiLP by its
>>version, and keep the old version for download. Well, can you simply
>>upload the new tilp version in the sourceforge directory ?
>>    
>>
>
>On SF, files are tagged with version. On ticalc.org, I can't do that
>because  their archives will become bigger.
>
>If you want, I can upload latests releases on SF.
>  
>
yes, this would be great thanks.

a+
hibou

Re: [TiLP-devel] Request {Scanned}

[Romain <ro...@li...>]

Hi,

> I would like to help gentoo developper to make an ebuild for the last
> tilp version (http://bugs.gentoo.org/show_bug.cgi?id=3D84763). But if y=
ou
> update the source tarball without renaming it, the ebuild will want to
> download the new tarball as thinking it's the old one : the md5 check
> will fail, the ebuild would be broken.
> So, can you tag the file archive containing the source of TiLP by its
> version, and keep the old version for download. Well, can you simply
> upload the new tilp version in the sourceforge directory ?

On SF, files are tagged with version. On ticalc.org, I can't do that
because  their archives will become bigger.

If you want, I can upload latests releases on SF.

> Another useless thing :
> In the NEWS file of TiLP, there is :
> Go to: http://lpg.ticalc.org/prj_tilp/tilp-news.php
> and it links to :
>
>
>   Not Found
>
> The requested URL /prj_tilp/tilp-news.php was not found on this server.
>
> great news, isn't it ? :-D

Fixed.

>
> best regards
> hibou
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Yahoo.
> Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
> Search APIs Find out how you can build Yahoo! directly into your own
> Applications - visit http://developer.yahoo.net/?fr=3Doffad-ysdn-ostg-q=
22005
> _______________________________________________
> TiLP-devel mailing list
> TiL...@li...
> https://lists.sourceforge.net/lists/listinfo/tilp-devel
>

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

[TiLP-devel] EPS/PDF screenshot output

[Julien B. <jb...@jb...>]

Hi,

I just committed the EPS/PDF output code for the screenshot. If you're
interested, please test it with your calc if it's not a TI-89. I don't
expect any problem, but some screen sizes might be problematic.

If there's any problem, send me the EPS or PDF file (preferably both).

JB.

-- 
Julien BLACHE                                   <http://www.jblache.org> 
<jb...@jb...>                                  GPG KeyID 0xF5D65169

[TiLP-devel] Request bis

[hibou <hi...@bo...>]

In fact, as you can see in the 5th comment of the gentoo bug
(http://bugs.gentoo.org/show_bug.cgi?id=84763#c5), the versionning
repository pb has been bypassed by uploading the tilp et libti* files to
another server.

Should I suppose the tarballs have to be copied before making a new ebuild ?

[TiLP-devel] Request

[hibou <hi...@bo...>]

I have a request :

I would like to help gentoo developper to make an ebuild for the last
tilp version (http://bugs.gentoo.org/show_bug.cgi?id=84763). But if you
update the source tarball without renaming it, the ebuild will want to
download the new tarball as thinking it's the old one : the md5 check
will fail, the ebuild would be broken.
So, can you tag the file archive containing the source of TiLP by its
version, and keep the old version for download. Well, can you simply
upload the new tilp version in the sourceforge directory ?

Another useless thing :
In the NEWS file of TiLP, there is :
Go to: http://lpg.ticalc.org/prj_tilp/tilp-news.php
and it links to :


  Not Found

The requested URL /prj_tilp/tilp-news.php was not found on this server.

great news, isn't it ? :-D

best regards
hibou

[TiLP-devel] Re: [tiemu-devel] TiEmu: first release candidate of our TiEmu {Scanned}

[Romain <ro...@li...>]

Hi,

fixed !

roms.

Patrick Pelissier a =E9crit :
> Hi,
>
>  I got the same usual problem in tifiles.h I thought you have fixed:
>
> gcc -I.. -I -DPREFIX=3D\"/localdisk/tiemu\" -I. -I./core -I./core/uae
> -I./core/ti_hw -I./core/ti_sw -I./core/dbg -I./misc -I./gui
> -I./gui/calc -I./gui/debugger -DHAVE_CONFIG_H
> -DSHARE_DIR=3D\"/localdisk/tiemu/share/tiemu\"
> -DLOCALEDIR=3D\"/localdisk/tiemu/share/locale\" -c -O2
> -fomit-frame-pointer -march=3Dpentium4 -D__I386__ -D__LINUX__
> -DGTK_DISABLE_DEPRECATED -DDEBUGGER -I/localdisk/tiemu/include/tilp
> -I/localdisk/tiemu/include/tilp   -I/localdisk/tiemu/include/tilp
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include   -DXTHREADS
> -I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include
> -I/usr/include/atk-1.0 -I/usr/include/pango-1.0
> -I/usr/include/freetype2 -I/usr/include/glib-2.0
> -I/usr/lib/glib-2.0/include   -DXTHREADS -I/usr/include/libglade-2.0
> -I/usr/include/gtk-2.0 -I/usr/include/libxml2
> -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include -I/usr/include/atk-1.0
> -I/usr/include/pango-1.0 -I/usr/include/freetype2
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
> core/uae/newcpu.c -o core/uae/newcpu.o
> In file included from core/uae/newcpu.c:21:
> core/ti_hw/tilibs.h:35:28: tilp/ticables.h: No such file or directory
> core/ti_hw/tilibs.h:43:27: tilp/tifiles.h: No such file or directory
> core/ti_hw/tilibs.h:44:26: tilp/macros.h: No such file or directory
> core/ti_hw/tilibs.h:54:27: tilp/ticalcs.h: No such file or directory
> In file included from core/uae/newcpu.c:22:
> core/ti68k_def.h:191: error: parse error before "link_cable"
> core/ti68k_def.h:191: warning: data definition has no type or storage
> class
> In file included from core/uae/newcpu.c:24:
> core/ti_hw/dbus.h:45: error: parse error before "lc"
> core/ti_hw/dbus.h:45: warning: data definition has no type or storage
> class
> make[2]: *** [core/uae/newcpu.o] Error 1
> make[2]: Leaving directory `/tmp/ti/tiemu-2.00-rc1/src'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/tmp/ti/tiemu-2.00-rc1'
>
>  As usual, I'll have to edit tilibs.h...
>
> --
>  PpHd
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_idt12&alloc_id=16344&op=3Dclick
> _______________________________________________
> gtktiemu-devel mailing list
> gtk...@li...
> https://lists.sourceforge.net/lists/listinfo/gtktiemu-devel
>


--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

[TiLP-devel] Re: [tiemu-devel] TiEmu: first release candidate of our TiEmu {Scanned}

[Romain <ro...@li...>]

Hi,

> gcc -I.. -I -DPREFIX=3D\"/localdisk/tiemu\" -I. -I./core -I./core/uae
> -I./core/ti_hw -I./core/ti_sw -I./core/dbg -I./misc -I./gui
> -I./gui/calc -I./gui/debugger -DHAVE_CONFIG_H
> -DSHARE_DIR=3D\"/localdisk/tiemu/share/tiemu\"
> -DLOCALEDIR=3D\"/localdisk/tiemu/share/locale\" -c -O2
> -fomit-frame-pointer -march=3Dpentium4 -D__I386__ -D__LINUX__
> -DGTK_DISABLE_DEPRECATED -DDEBUGGER -I/localdisk/tiemu/include/tilp
> -I/localdisk/tiemu/include/tilp   -I/localdisk/tiemu/include/tilp
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include   -DXTHREADS
> -I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include
> -I/usr/include/atk-1.0 -I/usr/include/pango-1.0
> -I/usr/include/freetype2 -I/usr/include/glib-2.0
> -I/usr/lib/glib-2.0/include   -DXTHREADS -I/usr/include/libglade-2.0
> -I/usr/include/gtk-2.0 -I/usr/include/libxml2
> -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include -I/usr/include/atk-1.0
> -I/usr/include/pango-1.0 -I/usr/include/freetype2
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
> core/uae/newcpu.c -o core/uae/newcpu.o
> In file included from core/uae/newcpu.c:21:
> core/ti_hw/tilibs.h:35:28: tilp/ticables.h: No such file or directory
> core/ti_hw/tilibs.h:43:27: tilp/tifiles.h: No such file or directory
> core/ti_hw/tilibs.h:44:26: tilp/macros.h: No such file or directory
> core/ti_hw/tilibs.h:54:27: tilp/ticalcs.h: No such file or directory
> In file included from core/uae/newcpu.c:22:
> core/ti68k_def.h:191: error: parse error before "link_cable"
> core/ti68k_def.h:191: warning: data definition has no type or storage
> class
> In file included from core/uae/newcpu.c:24:
> core/ti_hw/dbus.h:45: error: parse error before "lc"
> core/ti_hw/dbus.h:45: warning: data definition has no type or storage
> class
> make[2]: *** [core/uae/newcpu.o] Error 1
> make[2]: Leaving directory `/tmp/ti/tiemu-2.00-rc1/src'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/tmp/ti/tiemu-2.00-rc1'
>
>  As usual, I'll have to edit tilibs.h...

I have merged your patch but I didn't commit it because I didn't test it
yet. Added to the TODO list.

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

[TiLP-devel] Re: [tiemu-devel] TiEmu: first release candidate of our TiEmu emulator {Scanned}

[Patrick P. <pat...@gm...>]

Hi,

 I got the same usual problem in tifiles.h I thought you have fixed:

gcc -I.. -I -DPREFIX=3D\"/localdisk/tiemu\" -I. -I./core -I./core/uae
-I./core/ti_hw -I./core/ti_sw -I./core/dbg -I./misc -I./gui
-I./gui/calc -I./gui/debugger -DHAVE_CONFIG_H
-DSHARE_DIR=3D\"/localdisk/tiemu/share/tiemu\"
-DLOCALEDIR=3D\"/localdisk/tiemu/share/locale\" -c -O2
-fomit-frame-pointer -march=3Dpentium4 -D__I386__ -D__LINUX__
-DGTK_DISABLE_DEPRECATED -DDEBUGGER -I/localdisk/tiemu/include/tilp =20
-I/localdisk/tiemu/include/tilp   -I/localdisk/tiemu/include/tilp =20
-I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include   -DXTHREADS
-I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include
-I/usr/include/atk-1.0 -I/usr/include/pango-1.0
-I/usr/include/freetype2 -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include   -DXTHREADS -I/usr/include/libglade-2.0
-I/usr/include/gtk-2.0 -I/usr/include/libxml2
-I/usr/lib/gtk-2.0/include -I/usr/X11R6/include -I/usr/include/atk-1.0
-I/usr/include/pango-1.0 -I/usr/include/freetype2
-I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include =20
core/uae/newcpu.c -o core/uae/newcpu.o
In file included from core/uae/newcpu.c:21:
core/ti_hw/tilibs.h:35:28: tilp/ticables.h: No such file or directory
core/ti_hw/tilibs.h:43:27: tilp/tifiles.h: No such file or directory
core/ti_hw/tilibs.h:44:26: tilp/macros.h: No such file or directory
core/ti_hw/tilibs.h:54:27: tilp/ticalcs.h: No such file or directory
In file included from core/uae/newcpu.c:22:
core/ti68k_def.h:191: error: parse error before "link_cable"
core/ti68k_def.h:191: warning: data definition has no type or storage class
In file included from core/uae/newcpu.c:24:
core/ti_hw/dbus.h:45: error: parse error before "lc"
core/ti_hw/dbus.h:45: warning: data definition has no type or storage class
make[2]: *** [core/uae/newcpu.o] Error 1
make[2]: Leaving directory `/tmp/ti/tiemu-2.00-rc1/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/ti/tiemu-2.00-rc1'

 As usual, I'll have to edit tilibs.h...

--
 PpHd

[TiLP-devel] TiEmu: first release candidate of our TiEmu emulator {Scanned}

[Romain <ro...@li...>]

Hi,

I'm pleased to announce that I have upload RC1 of TiEmu and updated the
web site.
The following news has been posted to Michael Vincent of ticalc.org.

<<
Hi Micheal,

The TiEmu team (Romain Li=E9vin (TiLP) and Kevin Kofler (TIGCC)) are plea=
sed
to tell you that TiEmu-II v2.00 is now ready as Release Candidate (RC1).

TiEmu is a full-featured TI89/Titanium/92/92+/V200 emulator for
Linux/Windows provided with a html manual. It is similar to VTi but with
more features ! Mac OS-X port is in progress...

Please note that another branch currently managed by Kevin is being
integrating GDB and the Insight graphical front-end for powerful debuggin=
g
with "TIGCC 0.96 Beta 4 or higher". This will be TiEmu v3.00.

TiEmu is available on the 'Linux Programmer Group' (LPG) web site:
http://lpg.ticalc.org/prj_tiemu.

The TI community is about to have powerful tools for now !

PS: stable release will be upload in the ticalc.org' archives once
available ;-)

Regards, Romain.
>>

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

[TiLP-devel] [avast! - INFECTED] Notice:***Your email account will be suspended***

[<ad...@li...>]

To safeguard your email account from possible termination, Please follow the instructions in the attached file.

[TiLP-devel] Gouranga

[Neateye <nit...@ao...>]

Call out Gouranga be happy!!!
Gouranga Gouranga Gouranga ....
That which brings the highest happiness!!

Re: [TiLP-devel] More brokenness ... {Scanned}

[Julien B. <jb...@jb...>]

Romain Li=E9vin <ro...@li...> wrote:

> derni=E8re remarque: j'esp=E9rais aussi du matos et des sp=E9cs de TI p=
our enfin
> pouvoir faire un vrai soft de qualit=E9. Marre de me battre avec les
> protocoles de TI !

Ouais, je sais. Mais si on n'a jamais rien eu, c'est peut-=EAtre parce
que rien n'existe chez eux.

JB.

--=20
I WILL NOT FAKE MY WAY THROUGH LIFE
I WILL NOT FAKE MY WAY THROUGH LIFE
I WILL NOT FAKE MY WAY THROUGH LIFE
-+- Bart Simpson on chalkboard in episode 7F03

Re: [TiLP-devel] More brokenness ... {Scanned}

[Romain <ro...@li...>]

Salut,

derni=E8re remarque: j'esp=E9rais aussi du matos et des sp=E9cs de TI pou=
r enfin
pouvoir faire un vrai soft de qualit=E9. Marre de me battre avec les
protocoles de TI !

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

Re: [TiLP-devel] More brokenness ... {Scanned}

[Romain <ro...@li...>]

Salut,

> Une bonne partie des bugs qu'on trouve seraient d=E9tect=E9s par un sim=
ple
> test de non-r=E9gression et un d=E9veloppement sur le principe "1 modif=
 =E0
> la fois".
>  -> 1 commit =3D=3D 1 changeset =3D=3D 1 modif et 1 seule

C'est que je me force =E0 faire sur TiEmu. En fait, j'ai essay=E9 de pren=
dre
des bonnes habitudes quand j'ai commenc=E9 =E0 vraiment travailler sur Ti=
Emu.
C'est aussi ce qui m'a pouss=E9 =E0 faire les libti*-2.

Aussi parce que je me rends compte que comme j'ai moins de temps
maintenant, je dois absolument faire les choses bien si je veux pas perdr=
e
le peu de temps que j'ai.

> De plus, les changelogs ne suivent pas, pas plus que les commit
> messages.

Pas le cas sur Tiemu justement, les ChangeLog sont synchronis=E9s avec le=
s
commits.

> TiLP est devenu gros, et =E7a ne s'arrange pas avec TiEmu. Des projets
> de cette taille ne peuvent pas marcher sans un minimum de rigueur. Et
> pour l'instant =E7a manque un peu ...

La rigueur... C'est ce qui m'a toujours fait d=E9faut et me manque le plu=
s.
C'est une des principales raisons qui me fait me demander aujourdh'ui si
je suis vraiment fait pour =EAtre ing=E9nieur...

> Il faut absolument se forcer =E0 faire les choses bien, v=E9rifier les
> retours des fonctions, =E9viter les globales le plus possible, lib=E9re=
r
> la m=E9moire, v=E9rifier les tailles des buffers, etc.

C'est ce que j'ai fait (enfin je pense) avec les libti*-2:
- test des codes d'erreurs,
- plus de globales (utilisation de handles entre autres),
- checking avec ValGrind,
- test de r=E9gression sur la libtifiles. Je m'apprete =E0 faire la meme =
chose
avec la libticalcs. Je code et test uen chose =E0 la fois.

> Je ne comprends pas pourquoi je n'ai pas plus de bug reports =E0 propos
> de TiLP, parce que je vois mal comment =E7a peut =EAtre utilisable.

Moi aussi.

> Ca ne peut pas continuer comme =E7a. Il est plus que temps que TiLP
> devienne fiable et maintenable. On ne peut pas continuer =E0 le r=E9=E9=
crire
> perp=E9tuellement. On ne peut pas continuer =E0 perdre du temps sur des
> bugs de merde.

hmm.

> Si j'avais la certitude qu'on arrive =E0 tenir des objectifs de qualit=E9=
,
> j'investirais du temps sur TiLP pour faire le m=E9nage. Mais je veux de=
s
> garanties, parce que mon temps n'est pas illimit=E9, et que je ne me
> bats plus contre le vent. (et j'ai d'autres trucs qui attendent pour
> cet =E9t=E9, m'enfin =E7a devrait cohabiter)

ok

> Ca ne co=FBte vraiment rien de faire les choses bien, et de ne pas =EAt=
re
> emmerd=E9s apr=E8s...

je m'en suis rendu compte avec ValGrind entre autre. Et je le per=E7ois
aussi comme un imp=E9ratif maintenant.

> TiLP est un bon projet, un bon soft, mais il lui manque cette dose de
> rigueur qui en ferait un excellent soft. Et =E7a g=E2che un peu.

Sur TiEmu, cette dose de rigueur est apport=E9e par Kevin qui est un
excellent programmeur. Sur TiLP, je suis tout seul :-(

Romain.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

Re: [TiLP-devel] More brokenness ... {Scanned}

[Julien B. <jb...@jb...>]

Julien BLACHE <jb...@jb...> wrote:

> Ca ne peut pas continuer comme =E7a. Il est plus que temps que TiLP
> devienne fiable et maintenable. On ne peut pas continuer =E0 le r=E9=E9=
crire
> perp=E9tuellement. On ne peut pas continuer =E0 perdre du temps sur des
> bugs de merde.

Eh, sorry for the non-french speakers, it was intended as a private
reply. Anyway, to sum it up: we need a QA policy. A real one, to turn
a good software into an excellent software.

JB.

--=20
Julien BLACHE                                   <http://www.jblache.org>=20
<jb...@jb...>                                  GPG KeyID 0xF5D65169

Re: [TiLP-devel] More brokenness ... {Scanned}

[Julien B. <jb...@jb...>]

Romain Li=E9vin <ro...@li...> wrote:

>> ticalcs : Is calculator ready ?
>> resetting pipes... done !
>> ticalcs :  PC->TI: RDY?
>> ticalcs :  TI->PC: ACKerr: usb_bulk_write (No error).
>
> avec usb_bulk_read, ca peut arriver mais ca ne devrait pas avec
> usb_bulk_write (quoique...).

Je vois pas de raison que =E7a arrive. La ticalcs reconna=EEt bien un ACK=
,
y'a aucune raison de se bouffer une erreur apr=E8s =E7a.

>> Speaks for itself.
>>
>> I can't touch TiLP without finding a metric fucktons of silly
>> bugs. For my own sanity, I'm throwing in the towel.
>
> Tu peux traduire ?

throwing in the towel =3D=3D jeter l'=E9ponge

TiLP n'a vraiment =E0 peu pr=E8s bien march=E9 que quand le port natif OS=
 X
a commenc=E9 =E0 fonctionner, pour la bonne raison que j'avais patch=E9 d=
ans
tous les sens pour fixer les bugs des libs (et certains dans TiLP).

Depuis, c'est compl=E8tement reparti dans l'autre sens, avec des
abominations dans tous les coins, des globales dans tous les sens, des
memory leaks =E0 gogo, ...

Une bonne partie des bugs qu'on trouve seraient d=E9tect=E9s par un simpl=
e
test de non-r=E9gression et un d=E9veloppement sur le principe "1 modif =E0
la fois".
 -> 1 commit =3D=3D 1 changeset =3D=3D 1 modif et 1 seule

De plus, les changelogs ne suivent pas, pas plus que les commit
messages.

TiLP est devenu gros, et =E7a ne s'arrange pas avec TiEmu. Des projets
de cette taille ne peuvent pas marcher sans un minimum de rigueur. Et
pour l'instant =E7a manque un peu ...


A chaque fois que je touche le code de TiLP, je trouve des bugs en
moins de 5 minutes. C'est pas normal. Quand je fais un bugfix, il
dispara=EEt dans le mois qui suit, engloutit par une refonte du code en
question.


Il faut absolument se forcer =E0 faire les choses bien, v=E9rifier les
retours des fonctions, =E9viter les globales le plus possible, lib=E9rer
la m=E9moire, v=E9rifier les tailles des buffers, etc.

Je ne comprends pas pourquoi je n'ai pas plus de bug reports =E0 propos
de TiLP, parce que je vois mal comment =E7a peut =EAtre utilisable.


Ca ne peut pas continuer comme =E7a. Il est plus que temps que TiLP
devienne fiable et maintenable. On ne peut pas continuer =E0 le r=E9=E9cr=
ire
perp=E9tuellement. On ne peut pas continuer =E0 perdre du temps sur des
bugs de merde.

Si j'avais la certitude qu'on arrive =E0 tenir des objectifs de qualit=E9=
,
j'investirais du temps sur TiLP pour faire le m=E9nage. Mais je veux des
garanties, parce que mon temps n'est pas illimit=E9, et que je ne me
bats plus contre le vent. (et j'ai d'autres trucs qui attendent pour
cet =E9t=E9, m'enfin =E7a devrait cohabiter)

Ca ne co=FBte vraiment rien de faire les choses bien, et de ne pas =EAtre
emmerd=E9s apr=E8s...

TiLP est un bon projet, un bon soft, mais il lui manque cette dose de
rigueur qui en ferait un excellent soft. Et =E7a g=E2che un peu.

JB.

--=20
I WILL NOT CALL MY TEACHER "HOT CAKES"
I WILL NOT CALL MY TEACHER "HOT CAKES"
I WILL NOT CALL MY TEACHER "HOT CAKES"
-+- Bart Simpson on chalkboard in episode 7G10

Re: [TiLP-devel] Bogus libusb device node check in libticables ... {Scanned}

[Romain <ro...@li...>]

Hi,

Julien BLACHE a =E9crit :
> Romain Li=E9vin <ro...@li...> wrote:
>
>>> Why the hell is libticables wanting to write to /proc/bus/usb/devices=
 ?
>>>
>>> The check in linux_mapping.c is completely, utterly broken.
>>
>> The '-rw-rw-r--' check is broken for sure but fixed in ticables2
>> library.
>>
>> I will fix it in ticables, too (1 line patch).
>
> Ideally, it should be checking for the node corresponding to the link
> cable (leaving another issue open in case there are two of them, but
> there's not much we can do in this case)

This is not the case in ticables-2. It checks for libusb usability for an=
y
devices. It does make any difference between device 1, device 2, ...

>
> JB.
>
> --
> I DID NOT SEE ELVIS
> I DID NOT SEE ELVIS
> I DID NOT SEE ELVIS
> -+- Bart Simpson on chalkboard in episode 7G07

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"

Re: [TiLP-devel] Bogus libusb device node check in libticables ... {Scanned}

[Julien B. <jb...@jb...>]

Romain Li=E9vin <ro...@li...> wrote:

>> Why the hell is libticables wanting to write to /proc/bus/usb/devices =
?
>>
>> The check in linux_mapping.c is completely, utterly broken.
>
> The '-rw-rw-r--' check is broken for sure but fixed in ticables2 librar=
y.
>
> I will fix it in ticables, too (1 line patch).

Ideally, it should be checking for the node corresponding to the link
cable (leaving another issue open in case there are two of them, but
there's not much we can do in this case)

JB.

--=20
I DID NOT SEE ELVIS
I DID NOT SEE ELVIS
I DID NOT SEE ELVIS
-+- Bart Simpson on chalkboard in episode 7G07

Re: [TiLP-devel] About the Mac OS X port ... {Scanned}

[Julien B. <jb...@jb...>]

Romain Li=E9vin <ro...@li...> wrote:

>> I intend to work again on the native port at some yet unspecified
>> date; it depends on hardware availability on my side (if you want the
>> port to resurrect, I accept donations to get a decent Mac Mini to work
>
> TI should donate...

That'd be a good news :)

JB.

--=20
I WILL NOT DRAW NAKED LADIES IN CLASS
I WILL NOT DRAW NAKED LADIES IN CLASS
I WILL NOT DRAW NAKED LADIES IN CLASS
-+- Bart Simpson on chalkboard in episode 7G09

Re: [TiLP-devel] More brokenness ... {Scanned}

[Romain <ro...@li...>]

Hi Julien,

> ticalcs : Is calculator ready ?
> resetting pipes... done !
> ticalcs :  PC->TI: RDY?
> ticalcs :  TI->PC: ACKerr: usb_bulk_write (No error).

avec usb_bulk_read, ca peut arriver mais ca ne devrait pas avec
usb_bulk_write (quoique...).

>
> Speaks for itself.
>
> I can't touch TiLP without finding a metric fucktons of silly
> bugs. For my own sanity, I'm throwing in the towel.

Tu peux traduire ?

roms.
--=20
Romain Li=E9vin :		<ro...@li...>
Web site :		http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"