Re: GSoC - Designing a faster index format

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

 



On Sat, Mar 24, 2012 at 2:50 PM, elton sky <eltonsky9404@xxxxxxxxx> wrote:
> Thanks again Nguyen,
>
>> Which reminds me, we cannot abandon current index format. Users should
>> be allowed to choose which format to use. It may be hard to keep the
>> code support two formats while still taking advantage of the new one.
>> Maybe you could internally convert old format to new one in memory so
>> that git code only has to deal with one format, but that adds more
>> cost on using old format. I don't know..
>
> I understand we should allow user to switch between old & new format.
> But I guess that should only happens when user init a working dir,
> isn't it? Otherwise I have to transform them back n forth. If a user
> chooses to use old format, I assume their repository is not large, so
> there should not be big delay for new format.

Users may choose to stick with old format because other git tools rely
on that (or they want to use older git versions at the same time).
Note that current index works "fine" with ~50k files in working
directory. Not huge, but not small either. Overhead on old format
should be reasonable.
-- 
Duy
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]