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

Re: Bug#664794: lintian: should we compress some collections (file-info and index)?



On 2012-03-25 21:17, Russ Allbery wrote:
> Niels Thykier <niels@thykier.net> writes:
> 
> [...]
>> I had a look at some other candidates and I am thinking that java-info,
>> copyright-file and md5sums.  However, as it is we sometimes just leave
>> an empty file for these collections (if there is no information etc.).
>>   For copyright-file and java-info this is probably going to be common
>> case (symlinked u/s/d/$pkg and no jar files respectively).
> 
>> My personal view is that we could do without the empty files and then
>> only leave a file if there is any information.  It will probably require
>> some changes to checks (or collections) that access these directly, but
>> I think we should take that as an oppertunity of improving (the usage
>> of) L::Collect. :)
> 
> Is it maybe time to start installing our Perl modules in the Perl search
> path?

Personally, this has been one of the things I have been wanting for a
while[0], so generally I approve of that suggestion.  :)

>  We'd probably need to add a BEGIN block to the lintian frontend to
> search the command-line options for --root and insert a "use lib"
> statement if one was found to be sure we get the new modules, but I think
> that should be sufficient.

Actually, I do not see why the frontends would need change (a lot).
Also, I assume we are not going to install checks in the perl search
path, so we would still need to update @INC (in lintian).

>  And then any out-of-tree stuff that wants to
> parse the lab can start using a documented API for doing so, with the
> understanding that it's still in flux and could change further.
> 

I would probably want to see index() return objects first and solve
"index vs $colls"[1] first.  I especially suspect that the latter will
cause some people issues.

Revisiting the issue (after todays commits), I am inclined to just
remove the extra layer of quoting when reading index files in
L::Collect.  It would save us some duplicated "strip quoting".
  If the stripped quoting is (or had been) an issue we already have an
issue as chekcs/binaries (and others) use the filenames from
(sorted_)file_info rather than the "quoted" (sorted_)index.

> I should separately redo license-count in the Policy package to use
> Lintian::Collect.  That would be fairly easy to do.
> 

Except L::Collect does not cover the copyright-file yet.  :)

But once it does, you can throw in a Lintian::Lab::visit_packages and
license-count will be lab (layout) agnostic as well! :P

~Niels

[0] You may remember
http://lists.debian.org/debian-lint-maint/2011/07/msg00044.html

[1] http://lists.debian.org/debian-lint-maint/2011/08/msg00291.html


Reply to: