Released version 3.0.3
 

Hello,

I have released and packaged namcap 3.0.3. It includes Dan's patches
for the PKGBUILD analysis and a fix for the dangling symlink
detection.

I hope this will be the last bugfix release before the move from
[testing] to [extra].

Regards,
Rémy.

Released version 3.0.3
 

Hello,

I have released and packaged namcap 3.0.3. It includes Dan's patches
for the PKGBUILD analysis and a fix for the dangling symlink
detection.

I hope this will be the last bugfix release before the move from
[testing] to [extra].

Regards,
Rémy.

Released version 3.0.3
 

On Sat, Mar 26, 2011 at 6:26 AM, Rémy Oudompheng
<remyoudompheng@gmail.com> wrote:
> Hello,
>
> I have released and packaged namcap 3.0.3. It includes Dan's patches
> for the PKGBUILD analysis and a fix for the dangling symlink
> detection.
>
> I hope this will be the last bugfix release before the move from
> [testing] to [extra].
>
> Regards,
> Rémy.
>
>

Hi,

I'm still getting this false positive:
PKGBUILD (wxsvg) W: Non standard variable 'fvwm_icon_width' doesn't
start with an underscore

Wasn't Dan's patch supposed to fix that?

Eric

Released version 3.0.3
 

On 2011/3/27 Eric Bélanger <snowmaniscool@gmail.com> wrote:
> Hi,
>
> I'm still getting this false positive:
> PKGBUILD (wxsvg) W: Non standard variable 'fvwm_icon_width' doesn't
> start with an underscore
>
> Wasn't Dan's patch supposed to fix that?

No, Dan's patch is for variables declared in the PKGBUILD, it does not
ignore environment variables. I suggest two possible fixes:
* add a "/usr/bin/env -i" in /usr/bin/parsepkgbuild so that it ignores
your environment
* modify also the Python code to check that the name of the culprit
variable actually exists in the PKGBUILD

Maybe both.

Rémy.

Released version 3.0.3
 

On Sun, Mar 27, 2011 at 4:54 AM, Rémy Oudompheng
<remyoudompheng@gmail.com> wrote:
> On 2011/3/27 Eric Bélanger <snowmaniscool@gmail.com> wrote:
>> Hi,
>>
>> I'm still getting this false positive:
>> PKGBUILD (wxsvg) W: Non standard variable 'fvwm_icon_width' doesn't
>> start with an underscore
>>
>> Wasn't Dan's patch supposed to fix that?
>
> No, Dan's patch is for variables declared in the PKGBUILD, it does not
> ignore environment variables. I suggest two possible fixes:
> * add a "/usr/bin/env -i" in /usr/bin/parsepkgbuild so that it ignores
> your environment
> * modify also the Python code to check that the name of the culprit
> variable actually exists in the PKGBUILD
>
> Maybe both.

The first seems like the more correct solution here.

-Dan