view tests/test-merge-remove.t @ 42377:0546ead39a7e stable

manifest: avoid corruption by dropping removed files with pure (issue5801)

Previously, removed files would simply be marked by overwriting the first byte
with NUL and dropping their entry in `self.position`.  But no effort was made to
ignore them when compacting the dictionary into text form.  This allowed them to
slip into the manifest revision, since the code seems to be trying to minimize
the string operations by copying as large a chunk as possible.  As part of this,
compact() walks the existing text based on entries in the `positions` list, and
consumed everything up to the next position entry.   This typically resulted in
a ValueError complaining about unsorted manifest entries.

Sometimes it seems that files do get dropped in large repos- it seems to
correspond to there being a new entry that would take the same slot.  A much
more trivial problem is that if the only changes were removals, `_compact()`
didn't even run because `__delitem__` doesn't add anything to `self.extradata`.
Now there's an explicit variable to flag this, both to allow `_compact()` to
run, and to avoid searching the manifest in cases where there are no removals.

In practice, this behavior was mostly obscured by the check in fastdelta() which
takes a different path that explicitly drops removed files if there are fewer
than 1000 changes.  However, timeless has a repo where after rebasing tens of
commits, a totally different path[1] is taken that bypasses the change count
check and hits this problem.

[1] https://www.mercurial-scm.org/repo/hg/file/2338bdea4474/mercurial/manifest.py#l1511

  $ hg init

  $ echo foo > foo
  $ echo bar > bar
  $ hg ci -qAm 'add foo bar'

  $ echo foo2 >> foo
  $ echo bleh > bar
  $ hg ci -m 'change foo bar'

  $ hg up -qC 0
  $ hg mv foo foo1
  $ echo foo1 > foo1
  $ hg cat foo >> foo1
  $ hg ci -m 'mv foo foo1'
  created new head

  $ hg merge
  merging foo1 and foo to foo1
  1 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)

  $ hg debugstate --no-dates
  m   0         -2 unset               bar
  m   0         -2 unset               foo1
  copy: foo -> foo1

  $ hg st -q
  M bar
  M foo1


Removing foo1 and bar:

  $ cp foo1 F
  $ cp bar B
  $ hg rm -f foo1 bar

  $ hg debugstate --no-dates
  r   0         -1 set                 bar
  r   0         -1 set                 foo1
  copy: foo -> foo1

  $ hg st -qC
  R bar
  R foo1


Re-adding foo1 and bar:

  $ cp F foo1
  $ cp B bar
  $ hg add -v foo1 bar
  adding bar
  adding foo1

  $ hg debugstate --no-dates
  n   0         -2 unset               bar
  n   0         -2 unset               foo1
  copy: foo -> foo1

  $ hg st -qC
  M bar
  M foo1
    foo


Reverting foo1 and bar:

  $ hg revert -vr . foo1 bar
  saving current version of bar as bar.orig
  saving current version of foo1 as foo1.orig
  reverting bar
  reverting foo1

  $ hg debugstate --no-dates
  n   0         -2 unset               bar
  n   0         -2 unset               foo1
  copy: foo -> foo1

  $ hg st -qC
  M bar
  M foo1
    foo

  $ hg diff

Merge should not overwrite local file that is untracked after remove

  $ rm *
  $ hg up -qC
  $ hg rm bar
  $ hg ci -m 'remove bar'
  $ echo 'memories of buried pirate treasure' > bar
  $ hg merge
  bar: untracked file differs
  abort: untracked files in working directory differ from files in requested revision
  [255]
  $ cat bar
  memories of buried pirate treasure

Those who use force will lose

  $ hg merge -f
  file 'bar' was deleted in local [working copy] but was modified in other [merge rev].
  What do you want to do?
  use (c)hanged version, leave (d)eleted, or leave (u)nresolved? u
  merging foo1 and foo to foo1
  0 files updated, 1 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat bar
  bleh
  $ hg st
  M bar
  M foo1