tests/test-convert-bzr-ghosts.t

lfs: avoid quadratic performance in processing server responses

This is also adapted from the Facebook repo[1].  Unlike there, we were already
reading the download stream in chunks and immediately writing it to disk, so we
basically avoided the problem on download.  There shouldn't be a lot of data to
read on upload, but it's better to get rid of this pattern.

[1] https://github.com/facebookexperimental/eden/commit/82df66ffe97e21f3ee73dfec093c87500fc1f6a7

Differential Revision: https://phab.mercurial-scm.org/D7882

#require bzr

  $ . "$TESTDIR/bzr-definitions"
  $ cat > ghostcreator.py <<EOF
  > import sys
  > from bzrlib import workingtree
  > wt = workingtree.WorkingTree.open('.')
  > 
  > message, ghostrev = sys.argv[1:]
  > wt.set_parent_ids(wt.get_parent_ids() + [ghostrev])
  > wt.commit(message)
  > EOF

ghost revisions

  $ mkdir test-ghost-revisions
  $ cd test-ghost-revisions
  $ bzr init -q source
  $ cd source
  $ echo content > somefile
  $ bzr add -q somefile
  $ bzr commit -q -m 'Initial layout setup'
  $ echo morecontent >> somefile
  $ "$PYTHON" ../../ghostcreator.py 'Commit with ghost revision' ghostrev
  $ cd ..
  $ hg convert source source-hg
  initializing destination source-hg repository
  scanning source...
  sorting...
  converting...
  1 Initial layout setup
  0 Commit with ghost revision
  $ glog -R source-hg
  o  1@source "Commit with ghost revision" files+: [], files-: [], files: [somefile]
  |
  o  0@source "Initial layout setup" files+: [somefile], files-: [], files: []
  

  $ cd ..