public inbox for gentoo-dev-announce@lists.gentoo.org
 help / color / mirror / Atom feed
From: "Michał Górny" <mgorny@gentoo.org>
To: gentoo-dev-announce@lists.gentoo.org
Cc: gentoo-dev@lists.gentoo.org
Subject: [gentoo-dev-announce] PROPERTIES=test_network now available in git portage for packages whose only tests need Internet
Date: Wed, 19 May 2021 10:21:14 +0200	[thread overview]
Message-ID: <f369625200658c922b7551a7ddfde7fb20241b78.camel@gentoo.org> (raw)

Hello, everyone.

I've just pushed support PROPERTIES=test_network to git portage.  It can
now be used in ::gentoo to mark packages that are test-restricted
because their tests need to access the Internet, i.e.:

  # ebuild snippet
  PROPERTIES="test_network"
  RESTRICT="test"

Normally, this doesn't change anything.  But if you use a new enough
Portage (3.0.19+ when it's released, or 9999) and set in your make.conf:

  # make.conf snippet
  ALLOW_TEST="network"

Then Portage is going to unrestrict tests in the packages using the new
property, and run the test phase with network-sandbox disabled.  This
way developers and arch testers can improve test coverage without
affecting users that do not wish random Internet access.


This was originally proposed to EAPI 8 as RESTRICT="network-test". 
However, we've decided that by using a combination of PROPERTIES +
regular RESTRICT="test" we can add this to all EAPIs and benefit from it
immediately.

It is quite possible that future versions of Portage will allow
additional test_* properties and ALLOW_TEST values, e.g. "test_docker".

I'd say that this is an '80%' solution.  Generally, if we classified
packages on a non-linear scale of 1..5 that:

1 - all tests run locally
2 - there are some insignificant tests using Internet
3 - there are some local tests and some Internet tests and both are
needed to meaningfully test the package
4 - the vast majority of tests require Internet, the few tests that run
locally are insignificant
5 - all tests require Internet

This solution addresses 4 & 5.  1 is good, for 2 it should be good
enough to deselect the few tests.  3 is unsolved right now.

-- 
Best regards,
Michał Górny




                 reply	other threads:[~2021-05-19  8:23 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=f369625200658c922b7551a7ddfde7fb20241b78.camel@gentoo.org \
    --to=mgorny@gentoo.org \
    --cc=gentoo-dev-announce@lists.gentoo.org \
    --cc=gentoo-dev@lists.gentoo.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox