view tests/test-hooklib-reject_new_heads.t @ 50366:a445194f0a4d stable

backup: fix issue when the backup end up in a different directory

Because of store encoding, we might end up with the backup in a different
directory than the initial copy (for example if the backup path make it cross
the 120 char limit).

This can create crash, especially since 6.4 where backup are used during revlog
split.

Making sure the directory exists fixes these crash We added a test covering this
case.

Strictly speaking, this has always been broken, however the new code in 6.4
triggers it more easily.

  $ cat <<EOF >> $HGRCPATH
  > [extensions]
  > hooklib =
  > 
  > [phases]
  > publish = False
  > EOF
  $ hg init a
  $ hg --cwd a debugbuilddag '.:parent.*parent'
  $ hg --cwd a log -G
  o  changeset:   2:fa942426a6fd
  |  tag:         tip
  |  parent:      0:1ea73414a91b
  |  user:        debugbuilddag
  |  date:        Thu Jan 01 00:00:02 1970 +0000
  |  summary:     r2
  |
  | o  changeset:   1:66f7d451a68b
  |/   user:        debugbuilddag
  |    date:        Thu Jan 01 00:00:01 1970 +0000
  |    summary:     r1
  |
  o  changeset:   0:1ea73414a91b
     tag:         parent
     user:        debugbuilddag
     date:        Thu Jan 01 00:00:00 1970 +0000
     summary:     r0
  
  $ hg init b
  $ cat <<EOF >> b/.hg/hgrc
  > [hooks]
  > pretxnclose.reject_new_heads = \
  >   python:hgext.hooklib.reject_new_heads.hook
  > EOF
  $ hg --cwd b pull ../a
  pulling from ../a
  requesting all changes
  adding changesets
  adding manifests
  adding file changes
  error: pretxnclose.reject_new_heads hook failed: Changes on branch 'default' resulted in multiple heads
  transaction abort!
  rollback completed
  abort: Changes on branch 'default' resulted in multiple heads
  [255]
  $ hg --cwd b pull ../a -r 1ea73414a91b
  pulling from ../a
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 0 changes to 0 files
  new changesets 1ea73414a91b (1 drafts)
  (run 'hg update' to get a working copy)