[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Ten patches for xgps
From: |
James Browning |
Subject: |
Re: Ten patches for xgps |
Date: |
Thu, 18 Jul 2024 13:59:11 -0700 |
On Thursday, July 18, 2024 1:16:12 PM PDT Fred Wright wrote:
>
> On Thu, 18 Jul 2024, Gary E. Miller wrote:
>
> > Please make these individual Merge Requests.
I can totally see opening up a bunch of merge requests as a
justification to permanently scrap everything I am working on. Probably
on a premise like "Who can keep track of all of this?"
> I'm not sure whether per-commit MRs, or a single MR for all of them,
or
> something in between, is the best choice, but patches via email
certainly
> isn't. Patches via email are more tedious and more error-prone for
all
> parties concerned.
>
> The main criterion for including multiple commits in one MR is the
> expectation that they share a common criterion for acceptance. If
they
> do, then a single MR is more convenient. But if they have differing
> expectations for acceptance, then putting them in different MRs avoids
the
> need for partial merging. Though splitting them up gets more
complicated
> if they're interdependent.
>
> As far as pylint goes, note that different versions of pylint may
behave
> differently, and making something work better for one version
sometimes
> makes it work worse for another. An especially obnoxious property is
> where a given version of pylint complains about trying to disable a
> warning that it doesn't know about, which is *really* stupid.
>
> Essentially, it goes:
>
> User: "Don't warn me about X."
> Pylint: "I don't know about X, so I'll warn you about telling me not
to
> warn you about it."
The seventh patch caused the score to drop by 0.23 from the previous
slightly positive climb. Therefore the obvious thing to do is to drop
support for the affected Python 2; this is only four and a half years
after Python 2.7 went end-of-life. Whose cold undead hands are still
grasping it?
-30-
signature.asc
Description: This is a digitally signed message part.