Re: [GsoC] your latest commit of tasks_udd.py
Hi,
On Wed, Jul 29, 2015 at 03:37:39AM +0530, Akshita Jha wrote:
> > I can not see a diff here - perhaps just whitespace? Such things should
> > somehow vanish from the output.
>
> This is solved now.
Good. This enabled me spotting one thing. For official Debian packages
we have no information about the license in UDD. However, we can assume
that it is "DFSG free". This is what I had inserted for official
packages in main.
In contrest to packages in contrib / non-free the string "non-free" is
injected (which is not fully correct for packages in contrib - feel free
to add the first enhancement here and use "DFSG free" also for packages
in contrib).
> > While I follow your arguing that english definitely needs to be OK I
> > think we should compare at least one additional language (preferably one
> > featuring non-ASCII characters) as we have seen that these frequently
> > cause problems. For me German sounds like a good candidate since it has
> > such special characters and is quite good translated. May be having
> > some command line argument specifying an additional language in addition
> > to English would suffice.
> >
>
> As you suggested, I have made a slight change to allow an optional second
> command line argument to be given to the script. This argument has to be a
> lang in the form of 'de', 'da', 'fi' etc. This generates the diff of the
> tasks files in the specified language along with the diff in english
> (default).
>
> Usage: ./test_output.py <Blend> <optional lang> eg: ./test_output.py
> debian-med de
$ ./test_output.py debian-med de
Tasks files created are not the same
$ ./test_output.py debian-med
$
Seems there is some glitch here for the language addition
>
> <start_tag> : th <attr> : ('class', 'version') <data> : Version <end_tag>
> : th
> <start_tag> : th <attr> : ('class', 'archs') <data> : Architectures
> <end_tag> : th
> <end_tag> : tr
> +<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : stretch <end_tag> : td
> +<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-2 <end_tag>
> : td
> +<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> +<end_tag> : tr
> +<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : wheezy <end_tag> : td
> +<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-1 <end_tag>
> : td
> +<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> +<end_tag> : tr
> +<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : sid <end_tag> : td
> +<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-2 <end_tag>
> : td
> +<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> +<end_tag> : tr
> <data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : jessie <end_tag> : td
> <start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-2 <end_tag>
> : td
> <start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> <end_tag> : tr
> -<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : sid <end_tag> : td
> -<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-2 <end_tag>
> : td
> -<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> -<end_tag> : tr
> -<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : stretch <end_tag> : td
> -<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-2 <end_tag>
> : td
> -<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> -<end_tag> : tr
> -<data> : <start_tag> : tr <start_tag> : td <attr> : ('class', 'release')
> <data> : wheezy <end_tag> : td
> -<start_tag> : td <attr> : ('class', 'version') <data> : 1.3.1-1 <end_tag>
> : td
> -<start_tag> : td <attr> : ('class', 'archs') <data> : all <end_tag> : td
> -<end_tag> : tr
>
> The diff is due to the difference in order: tasks.html : (sid, stretch,
> wheezy) and tasks_udd.html (stretch, wheezy, sid). Should this also be in
> alphabetical order ?
>
> > So far for my first notes.
It should be in release order (as per the sort field in the release table)
and the list of architectures in alphabethical order.
> Please have a look at the above changes and kindly suggest what else can be
> improved/changed ?
I'm happy about the test script since it uncovered needed changes in the
code:
1. Licenses of official packages
2. Fixing order od versions + archs
Kind regards
Andreas.
--
http://fam-tille.de
Reply to: