branchmap: server should not advertise secret changeset in branchmap (Issue3303)
Discovery now use an overlay above branchmap to prune invisible "secret"
changeset from branchmap.
To minimise impact on the code during the code freeze, this is achieve by
recomputing non-secret heads on the fly when any secret changeset exists. This
is a computation heavy approach similar to the one used for visible heads. But
few sever should contains secret changeset anyway. See comment in code for more
robust approach.
On local repo the wrapper is applied explicitly while the wire-protocol take
care of wrapping branchmap call in a transparent way. This could be unified by
the Peter Arrenbrecht and Sune Foldager proposal of a `peer` object.
An inappropriate `(+i heads)` may still appear when pushing new changes on a
repository with secret changeset. (see Issue3394 for details)
Failed qimport of patches from files should cleanup by recording successfully
imported patches in series file.
$ echo "[extensions]" >> $HGRCPATH
$ echo "mq=" >> $HGRCPATH
$ hg init repo
$ cd repo
$ echo a > a
$ hg ci -Am'add a'
adding a
$ cat >b.patch<<EOF
> diff --git a/a b/a
> --- a/a
> +++ b/a
> @@ -1,1 +1,2 @@
> a
> +b
> EOF
empty series
$ hg qseries
qimport valid patch followed by invalid patch
$ hg qimport b.patch fakepatch
adding b.patch to series file
abort: unable to read file fakepatch
[255]
valid patches before fail added to series
$ hg qseries
b.patch
$ hg pull -q -r 0 . # update phase
$ hg qimport -r 0
abort: revision 0 is not mutable
(see "hg help phases" for details)
[255]