lilypond-devel
[Top][All Lists]
Advanced

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

Stop issue verification?


From: Jean Abou Samra
Subject: Stop issue verification?
Date: Sun, 25 Oct 2020 20:43:31 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0

Hi all,

Before 2.21.80 is released, I'd like to restart this debate
(which might be contentious; we'll see).

Basically, I think our current processes obviate the effort
to go through all fixed issues and verify them. The primary job
was to ensure that patches had landed in master, due to a largely
hand-powered workflow that could be error-prone. In my experience,
many of the issues cannot be verified easily (especially when
involving the build system, Guile 2, etc.), and many others are
not actually bugs but documentation wishes for example. This all
takes a good amount of time, with little actual benefits as far
as I could see.The bugs for which verification is workable are
exactly those for which we are adding regression tests.

In my opinion, the limited total working time of the development
team is better spent on fixing bugs (there are plenty of easy ones
for newbies) than undertaking this task.

I propose that we:

- Remove this procedure.

- Also remove the "Status" scoped label, adding non-scoped labels
  when necessary.

  - Status::new and Status::accepted: no longer applicable since
    there is no longer a formal bug squad.

  - Status::fixed and Status::verified: not needed anymore if there
    is no verification (closed issues are fixed).

  - Status::started: assign the issue to yourself if you are working
    on it. This appears in the UI and you also see if an issue has
    an associated merge request.

  - Status::invalid, Status::duplicate and Status::shelved converted
    to plain labels.

- Explicitely state in [CG 9.1] that whenever applicable, bug fixes
  should contain a regression test − currently it says this needs a policy.

What do you think?

Regards,
Jean

[CG 9.1] http://lilypond.org/doc/v2.21/Documentation/contributor/introduction-to-regression-tests




reply via email to

[Prev in Thread] Current Thread [Next in Thread]