On 11/27/20 2:10 PM, Joe Perches wrote:
> On Fri, 2020-11-27 at 13:43 -0800, t...@redhat.com wrote:
>> From: Tom Rix <t...@redhat.com>
>>
>> From
>> RFC MAINTAINERS tag for cleanup robot
>> https://lkml.org/lkml/2020/11/21/190
> I think this should be RFC.
> It looks as as if it's only for subsystems through A
Yes only A, i wanted to show the direction in a subset.
>
>> The prefix used by subsystems in their commit log is arbitrary.
>> To elimitate the time and effort to manually find a reasonable
>> prefix, store the preferred prefix in the MAINTAINERS file.
>>
>> Populate with reasonable prefixes using this algorithm.
>> What did the maintainers use in their commits?
>> If there were no maintainers commits then what did everyone
>> else use in their commits.
> The algorithm used to produce these prefixes should also be published.
Ok.
>> The results were manually reviewed and about 25% were rejected.
> Because this isn't necessarily the best option.
>
> I think an exception mechanism would be better than a specific
> mechanism added to various entries.
Can you give an example of what you mean ?
>> # check MAINTAINERS entries for the right ordering too
>> - my $preferred_order = 'MRLSWQBCPTFXNK';
>> + my $preferred_order = 'MRLSWQBCPTFXNKD';
>> if ($rawline =~ /^\+[A-Z]:/ &&
>> $prevrawline =~ /^[\+ ][A-Z]:/) {
>> $rawline =~ /^\+([A-Z]):\s*(.*)/;
> I'd prefer to keep the file and keyword list last.
>
So change to
my $preferred_order = 'MRLSWQBCPTDFXNK';
?
>
>
