On Sat, 2010-09-25 at 22:10 -0400, Ted Ts'o wrote: > On Sat, Sep 25, 2010 at 07:03:11PM -0700, Joe Perches wrote: > > When you define "it" that way, not as any simple > > file pattern match, but as a control for what > > "git log -- path" > > to inspect, it's quite feasible to use the pattern > > match rather than the file name. > > So, thanks, that's a good suggestion. > Don't make it an option, though. If you must use hueristics, then at > least *try* to make the hueristics smarter. If the file name falls > into certain patterns, such as: > fs/*/*.[ch] > drivers/scsi/*/*.[ch] > drivers/net/*/*.[ch] > etc., then you really should be doing the analysis by subdirectory, > and not by file. I think if there's an exact pattern "depth" match, then git history should be searched by that subdirectory. So for example: F: drivers/scsi/megaraid/ A lookup for drivers/scsi/megaraid/mega_common.h (same depth, search all commits in drivers/scsi/megaraid) now shows: $ ./scripts/get_maintainer.pl -f drivers/scsi/megaraid/mega_common.h --rolestats --git Neela Syam Kolli <megaraidlinux@xxxxxxx> (maintainer:MEGARAID SCSI DRI...) "James E.J. Bottomley" <James.Bottomley@xxxxxxx> (maintainer:SCSI SUBSYSTEM,commit_signer:20/27=74%) linux-scsi@xxxxxxxxxxxxxxx (open list:MEGARAID SCSI DRI...) linux-kernel@xxxxxxxxxxxxxxx (open list) > But what was shipped as part of 2.6.35 was really, really, *REALLY* > bad. It makes me wonder how much testing someone did with the > hueristics before turning it loose for the newbies to use. A fair bit with all public discussions. It's been shipped more or less that way since 2.6.30 beginning in April, 2009. I get annoyed when people suggest that the maintainer of that silly script isn't responsive. -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
