[PATCH 0 of 2] Issue 1371: Inotify Patch with a test

Matt Mackall mpm at selenic.com
Sun Nov 2 13:56:55 CST 2008


On Sun, 2008-11-02 at 17:53 +0200, Gerard Korsten wrote:
> 
> On Sun, Nov 2, 2008 at 1:54 PM, Benoit Boissinot <bboissin at gmail.com>
> wrote:
>         On Sun, Nov 2, 2008 at 5:44 AM, Gerard Korsten
>         <soonkia77 at gmail.com> wrote:
>         
>         
>         I folded the patches in one patch. And slightly modified them.
>         They are now pushed to crew:
>         http://hg.intevation.org/mercurial/crew/rev/972737252d05
>         
>         It seems the test might race (deletion of 'b' isn't detected)
>         we'll
>         see if it causes
>         a problem.
>         
> 
> Hi there,
> 
> Thanks for adding it and making it better.
> 
> Not sure about the race condition as b is specifically deleted with
> only as normal system delete, and not with "hg rm".  
> 
> So, b is detected correctly by the test as with a status of "! b"
> which is what I wanted in the test.
> 
> I found that if you're not careful with hg's INotify, you can screw up
> the "hg status" output - so I decided to try and test for all file
> operations.

Two things to note here:

First, inotify now automatically starts when enabled instead of
complaining obscurely about no daemon running.

Second, there's a test mode that actually compares what inotify reports
with what normal status reports:

[inotify]
debug=1

-- 
Mathematics is the supreme nostalgia of our time.



More information about the Mercurial-devel mailing list