Re: master 1aa83da46d1 2/2: ; * etc/NEWS: Rename read-process-output-fast here also.

[Robert Pluim <rpluim@gmail.com>]

>>>>> On Tue, 11 Jun 2024 15:28:44 +0300, Dmitry Gutov <dmitry@gutov.dev> said:

    Dmitry> On 11/06/2024 11:32, Robert Pluim wrote:
    >> Po Lu> -To undo the change, set 'read-process-output-fast' to nil.
    >> Po Lu> +To undo the change, set 'fast-read-process-output' to nil.
    >> I realise weʼve renamed this one already, but would it not make
    >> more
    >> sense for the variable to be called something starting with
    >> 'process-', such as 'process-fast-read-output'? Thatʼs in line with
    >> other variables affecting the processes process data, such as
    >> `process-adaptive-read-buffering'.

    Dmitry> The idea behind this name was to match one specific C function (which
    Dmitry> might be familiar to many who read Emacs's code), but I don't have a
    Dmitry> strong opinion either way. process-read-output-fast, maybe?

If I was being pedantic [1] Iʼd say it should be
'process-read-output-simple', the 'fast' bit is just the side effect
;-)

More seriously: the variable name should describe the semantics. What
underlying function is used as a result is not relevant to users of
the variable.

    Dmitry> Also note that both previous names have been already mentioned in
    Dmitry> bug#71452 comments, so I suppose the new one should be mentioned too,
    Dmitry> in order to avoid keeping an outdated source on that subject.

Investigation for me normally goes

'search bug# in commit log' -> git show <revision> 

so it doesnʼt really matter if the bug comments match the code.

Robert

Footnotes:
[1]  Moi, pedantic? Never.

--