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

Bug#759403: lintian: Please publish machine-readable report for all packages



Control: tag -1 - moreinfo

Hi Chris!

Chris Lamb:
>> on the pkg-perl team, we would like to track the hardening status of
>> our packages (same goes in Tails, to track the hardening status of the
>> packages we ship).
>> 
>> The easiest way for us would possibly be to retrieve data about all
>> packages from lintian.d.o, filter on the maintainer field, and build
>> statistics and graphs from that.

> Can we close this in favour of the "lintian" table in UDD?

tl;dr: yes.


Thanks for the pointer, I had no idea that existed.

I've tested https://udd.debian.org/lintian/ for the two usecases I had
in mind:

1. Debian Perl team
===================

  $ curl 'https://udd.debian.org/lintian/?email1=pkg-perl-maintainers%40lists.alioth.debian.org&email2=&email3=&packages=&ignpackages=&format=yaml#all'

… gives me exactly what I was asking for: 
for example, I can easily extract the list of pkg-perl -maintained
packages with the hardening-no-bindnow tag. Woohoo! :))


2. Packages used to build Tails
===============================

(Assuming the current working directory contains a Tails build
manifest called latest.build-manifest, e.g. downloaded from
https://nightly.tails.boum.org/build_Tails_ISO_feature-buster/lastSuccessful/archive/latest.build-manifest)

  $ curl 'https://udd.debian.org/lintian/?email1=&email2=&email3=&packages='$(perl -mYAML::Tiny -E 'print join("+", map { $_->{package} } @{YAML::Tiny->read("latest.build-manifest")->[0]->{packages}->{binary}});')'&bin2src=on&ignpackages=&format=yaml#all'
  <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
  <html><head>
  <title>414 Request-URI Too Long</title>
  </head><body>
  <h1>Request-URI Too Long</h1>
  <p>The requested URL's length exceeds the capacity
  limit for this server.<br />
  </p>
  <hr>
  <address>Apache Server at udd.debian.org Port 443</address>
  </body></html>

… so I'll have to loop over this large list of packages, query UDD
with smaller subsets at a time, and somehow merge the results (that
can have duplicates since the input in a list of binary packages).
Or learn to query UDD directly without going through the web
interface, I guess.

This is not ideal but it's good enough so feel free to close this bug
report :)

Cheers,
-- 
intrigeri


Reply to: