.hgignore in subfolders

[Gregory Szorc]

On Tue, Aug 11, 2015 at 10:50 AM, Matt Mackall <mpm at selenic.com> wrote:

> On Tue, 2015-08-11 at 11:06 +0000, Harry . wrote:
> > Why is .hgignore in subfolders ignored (unless specifically included by
> the top-level .hgignore)?
>
> Two reasons. First, there is a substantial cost to trying to open a file
> in every directory on a large project. Second, we can build an optimized
> regular expression matcher if we have all the data at the beginning, but
> not if we load it incrementally.


Could we not maintain a cache of known .hgignore files? It could be updated
during transactions when changegroups are imported, like fncache. Managing
deleted .hgignore files could be problematic, so we'd need one set per head
or to incur extra I/O for deleted .hgignore files. I think it is doable.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20150811/c8059821/attachment-0002.html>