| commit | a9a6152725262b9061e5ee55d69efc579817bbf4 | |
| author | Kyle J. McKay <mackyle@gmail.com> | |
| Thu, 17 Nov 2016 00:03:30 +0000 (16 16:03 -0800) | ||
| committer | Kyle J. McKay <mackyle@gmail.com> | |
| Thu, 17 Nov 2016 00:03:30 +0000 (16 16:03 -0800) | ||
| tree | 7f3b94c444caa0bea962cbcbbb3b58a8fb677efa | treesnapshot (tar.gz zip) |
| parent | 16d371440df14473a38522534b3dc7532168f999 | commitdiff |
update.sh: fix bzr mirror update marks handling
Ever since bzr support was first added in 94186c32b5 (Add support for
mirroring bzr repos, 2010-01-28), the clone part of the process has been
writing marks to a bfe-marks file but the update side of the process has
been expecting a file named dfe-marks instead -- apparently a result of
a copy-and-paste error when bzr support was first added.
Correct this by making the update side of the process use the correct,
bfe-marks, file name.
Apparently the bzr support has not been used very much.
It would still probably work it's just that the first update would end
up redoing all the work of the initial clone and possibly timing out if
the clone is large.
Signed-off-by: Kyle J. McKay <mackyle@gmail.com>
Ever since bzr support was first added in 94186c32b5 (Add support for
mirroring bzr repos, 2010-01-28), the clone part of the process has been
writing marks to a bfe-marks file but the update side of the process has
been expecting a file named dfe-marks instead -- apparently a result of
a copy-and-paste error when bzr support was first added.
Correct this by making the update side of the process use the correct,
bfe-marks, file name.
Apparently the bzr support has not been used very much.
It would still probably work it's just that the first update would end
up redoing all the work of the initial clone and possibly timing out if
the clone is large.
Signed-off-by: Kyle J. McKay <mackyle@gmail.com>
| jobd/update.sh | diffblobblamehistory |