[PATCH] dirstate: use repository root as base for relative paths read from ui.ignore
mpm at selenic.com
Sun Apr 21 17:32:20 CDT 2013
On Fri, 2013-04-19 at 10:53 +0200, Isaac Jurado wrote:
> On Fri, Apr 19, 2013 at 1:16 AM, Matt Mackall <mpm at selenic.com> wrote:
> > On Thu, 2013-04-18 at 17:47 -0500, Matt Mackall wrote:
> >> On Thu, 2013-04-18 at 21:56 +0200, diptongo at gmail.com wrote:
> >> > # HG changeset patch
> >> > # User Isaac Jurado <diptongo at gmail.com>
> >> > # Date 1366314160 -7200
> >> > # Thu Apr 18 21:42:40 2013 +0200
> >> > # Node ID 380de75c43f4a186a1b8255393cf52c3af43820a
> >> > # Parent 7d31f2e42a8afb54c8fae87e8e3e29a63578aea4
> >> > dirstate: use repository root as base for relative paths read from ui.ignore
> >> >
> >> > Convert relative paths from the ui.ignore* configurations to absolute by
> >> > prepending the path to the repository root. This breaks the old, and probably
> >> > unused, behaviour where relative paths were interpreted from the working
> >> > directory.
> >> >
> >> > Absolute paths are kept untouched.
> >> Seems safe enough. Queued for default with a (BC) marker.
> > Or maybe not. Something just destroyed my
> > rsync-to-box-in-cloud-and-test script, and this is the most likely
> > culprit.
> Ok, I'll resend on May. Meanwhile, it could be useful to have more
> details abut such failure, if possible.
Turns out this was probably the fault of the recent color extension bug
and nothing wrong with your patch.
Mathematics is the supreme nostalgia of our time.
More information about the Mercurial-devel