[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#974909: explicitly differentiate between files and fileds in Check:



Hi Private,

On Mon, Nov 16, 2020 at 3:45 AM Pirate Praveen <praveen@onenetbeyond.org> wrote:
>
> As there is no standard field in debian/watch or there is no
> debian/watch/standard file. It would be obvious to someone who is
> experienced, but making it clearly spelt is helpful to newbies.

I sympathize with your desire for clarity and have put a lot of time
toward that goal. In the case of the watch file, I know what you mean
but I do not arrive at the same conclusion.

Perhaps you can agree that clarity is improved when different things
have different names. (My favorite example for that is driving in
Atlanta, where a significant number of streets are named Peachtree.)
In Lintian, we try to reserve the term"version" for package versions.
They are probably the single most confusing, most controversial and
least thought-out area in Debian. Part of it is, of course, that
upstreams are free to choose whichever system or format they like.

As a mental exercise, you can try to sort version numbers (for the
same package). We attempt to do that, however poorly, on lintian.d.o.
I do not see an advantage to mixing the concept of standard compliance
anywhere, such as for watch files, with package versions. As a mental
bridge, which I need sometimes as well, I find it helpful to think of
Debian's very own Standards-Version in d/control.

> Check: fields/mail-address could be,
>
> Check: Uploaders/Maintainer

I am not sure I understand your proposal here. As Chris wrote, the
check field was traditionally not published because our tags reside in
a global namespace.

There was a proposal for private name spaces [1] which was
implemented, but the concept was unpopular [2]. I really thought it
could help shorten tag names. It is currently only used for two
classification tags in the check continuous-integration/salsa.

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=943525
[2] https://lists.debian.org/debian-devel/2019/11/msg00288.html

Kind regards
Felix Lechner


Reply to: