tests/test-narrow-share.t

py3: make stdout line-buffered if connected to a TTY

Status messages that are to be shown on the terminal should be written to the
file descriptor before anything further is done, to keep the user updated.

One common way to achieve this is to make stdout line-buffered if it is
connected to a TTY. This is done on Python 2 (except on Windows, where libc,
which the CPython 2 streams depend on, does not properly support this).

Python 3 rolls it own I/O streams. On Python 3, buffered binary streams can't be
set line-buffered. The previous code (added in 227ba1afcb65) incorrectly
assumed that on Python 3, pycompat.stdout (sys.stdout.buffer) is already
line-buffered. However the interpreter initializes it with a block-buffered
stream or an unbuffered stream (when the -u option or the PYTHONUNBUFFERED
environment variable is set), never with a line-buffered stream.

One example where the current behavior is unacceptable is when running
`hg pull https://www.mercurial-scm.org/repo/hg` on Python 3, where the line
"pulling from https://www.mercurial-scm.org/repo/hg" does not appear on the
terminal before the hg process blocks while waiting for the server.

Various approaches to fix this problem are possible, including:

1. Weaken the contract of procutil.stdout to not give any guarantees about
   buffering behavior. In this case, users of procutil.stdout need to be
   changed to do enough flushes. In particular,
  1. either ui must insert enough flushes for ui.write() and friends, or
  2. ui.write() and friends get split into flushing and fully buffered
     methods, or
  3. users of ui.write() and friends must flush explicitly.
2. Make stdout unbuffered.
3. Make stdout line-buffered. Since Python 3 does not natively support that for
   binary streams, we must implement it ourselves.

(2.) is problematic because using unbuffered I/O changes the performance
characteristics significantly compared to line-buffered (which is used on
Python 2) and this would be a regression.

(1.2.) and (1.3) are a substantial amount of work. It’s unclear whether the
added complexity would be justified, given that raw performance doesn’t matter
that much when writing to a terminal much faster than the user could read it.

(1.1.) pushes complexity into the ui class instead of separating the concern of
how stdout is buffered. Other users of procutil.stdout would still need to take
care of the flushes.

This patch implements (3.). The general performance considerations are very
similar to (1.1.). The extra method invocation and method forwarding add a
little more overhead if the class is used. In exchange, it doesn’t add overhead
if not used.


For the benchmarks, I compared the previous implementation (incorrect on Python
3), (1.1.), (3.) and (2.). The command was chosen so that the streams were
configured as if they were writing to a TTY, but actually write to a pager,
which is also the default:
HGRCPATH=/dev/null python3 ./hg --cwd ~/vcs/mozilla-central --time --pager yes --config pager.pager='cat > /dev/null' status --all

previous:
time: real 7.880 secs (user 7.290+0.050 sys 0.580+0.170)
time: real 7.830 secs (user 7.220+0.070 sys 0.590+0.140)
time: real 7.800 secs (user 7.210+0.050 sys 0.570+0.170)

(1.1.) using Yuya Nishihara’s patch:
time: real 9.860 secs (user 8.670+0.350 sys 1.160+0.830)
time: real 9.540 secs (user 8.430+0.370 sys 1.100+0.770)
time: real 9.830 secs (user 8.630+0.370 sys 1.180+0.840)

(3.) using this patch:
time: real 9.580 secs (user 8.480+0.350 sys 1.090+0.770)
time: real 9.670 secs (user 8.480+0.330 sys 1.170+0.860)
time: real 9.640 secs (user 8.500+0.350 sys 1.130+0.810)

(2.) using a previous patch by me:
time: real 10.480 secs (user 8.850+0.720 sys 1.590+1.500)
time: real 10.490 secs (user 8.750+0.750 sys 1.710+1.470)
time: real 10.240 secs (user 8.600+0.700 sys 1.590+1.510)


As expected, there’s no difference on Python 2, as exactly the same code paths
are used:

previous:
time: real 6.950 secs (user 5.870+0.330 sys 1.070+0.770)
time: real 7.040 secs (user 6.040+0.360 sys 0.980+0.750)
time: real 7.070 secs (user 5.950+0.360 sys 1.100+0.760)

this patch:
time: real 7.010 secs (user 5.900+0.390 sys 1.070+0.730)
time: real 7.000 secs (user 5.850+0.350 sys 1.120+0.760)
time: real 7.000 secs (user 5.790+0.380 sys 1.170+0.710)

#testcases flat tree

  $ . "$TESTDIR/narrow-library.sh"

#if tree
  $ cat << EOF >> $HGRCPATH
  > [experimental]
  > treemanifest = 1
  > EOF
#endif

  $ cat << EOF >> $HGRCPATH
  > [extensions]
  > share =
  > EOF

  $ hg init remote
  $ cd remote
  $ for x in `$TESTDIR/seq.py 0 10`
  > do
  >   mkdir d$x
  >   echo $x > d$x/f
  >   hg add d$x/f
  >   hg commit -m "add d$x/f"
  > done
  $ cd ..

  $ hg clone --narrow ssh://user@dummy/remote main -q \
  > --include d1 --include d3 --include d5 --include d7

Ignore file called "ignored"
  $ echo ignored > main/.hgignore

  $ hg share main share
  updating working directory
  4 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg -R share tracked
  I path:d1
  I path:d3
  I path:d5
  I path:d7
  $ hg -R share files
  share/d1/f
  share/d3/f
  share/d5/f
  share/d7/f

Narrow the share and check that the main repo's working copy gets updated

# Make sure the files that are supposed to be known-clean get their timestamps set in the dirstate
  $ sleep 2
  $ hg -R main st
  $ hg -R main debugdirstate --no-dates
  n 644          2 set                 d1/f
  n 644          2 set                 d3/f
  n 644          2 set                 d5/f
  n 644          2 set                 d7/f
# Make d3/f dirty
  $ echo x >> main/d3/f
  $ echo y >> main/d3/g
  $ touch main/d3/ignored
  $ touch main/d3/untracked
  $ hg add main/d3/g
  $ hg -R main st
  M d3/f
  A d3/g
  ? d3/untracked
# Make d5/f not match the dirstate timestamp even though it's clean
  $ sleep 2
  $ hg -R main st
  M d3/f
  A d3/g
  ? d3/untracked
  $ hg -R main debugdirstate --no-dates
  n 644          2 set                 d1/f
  n 644          2 set                 d3/f
  a   0         -1 unset               d3/g
  n 644          2 set                 d5/f
  n 644          2 set                 d7/f
  $ touch main/d5/f
  $ hg -R share tracked --removeinclude d1 --removeinclude d3 --removeinclude d5
  comparing with ssh://user@dummy/remote
  searching for changes
  looking for local changes to affected paths
  deleting data/d1/f.i
  deleting data/d3/f.i
  deleting data/d5/f.i
  deleting meta/d1/00manifest.i (tree !)
  deleting meta/d3/00manifest.i (tree !)
  deleting meta/d5/00manifest.i (tree !)
  $ hg -R main tracked
  I path:d7
  $ hg -R main files
  abort: working copy's narrowspec is stale
  (run 'hg tracked --update-working-copy')
  [255]
  $ hg -R main tracked --update-working-copy
  not deleting possibly dirty file d3/f
  not deleting possibly dirty file d3/g
  not deleting possibly dirty file d5/f
  not deleting unknown file d3/untracked
  not deleting ignored file d3/ignored
# d1/f, d3/f, d3/g and d5/f should no longer be reported
  $ hg -R main files
  main/d7/f
# d1/f should no longer be there, d3/f should be since it was dirty, d3/g should be there since
# it was added, and d5/f should be since we couldn't be sure it was clean
  $ find main/d* -type f | sort
  main/d3/f
  main/d3/g
  main/d3/ignored
  main/d3/untracked
  main/d5/f
  main/d7/f

Widen the share and check that the main repo's working copy gets updated

  $ hg -R share tracked --addinclude d1 --addinclude d3 -q
  $ hg -R share tracked
  I path:d1
  I path:d3
  I path:d7
  $ hg -R share files
  share/d1/f
  share/d3/f
  share/d7/f
  $ hg -R main tracked
  I path:d1
  I path:d3
  I path:d7
  $ hg -R main files
  abort: working copy's narrowspec is stale
  (run 'hg tracked --update-working-copy')
  [255]
  $ hg -R main tracked --update-working-copy
# d1/f, d3/f should be back
  $ hg -R main files
  main/d1/f
  main/d3/f
  main/d7/f
# d3/f should be modified (not clobbered by the widening), and d3/g should be untracked
  $ hg -R main st --all
  M d3/f
  ? d3/g
  ? d3/untracked
  I d3/ignored
  C d1/f
  C d7/f

We should also be able to unshare without breaking everything:

  $ hg share main share-unshare
  updating working directory
  3 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ cd share-unshare
  $ hg unshare
  $ hg verify
  checking changesets
  checking manifests
  checking directory manifests (tree !)
  crosschecking files in changesets and manifests
  checking files
  checked 11 changesets with 3 changes to 3 files
  $ cd ..

Dirstate should be left alone when upgrading from version of hg that didn't support narrow+share

  $ hg share main share-upgrade
  updating working directory
  3 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ cd share-upgrade
  $ echo x >> d1/f
  $ echo y >> d3/g
  $ hg add d3/g
  $ hg rm d7/f
  $ hg st
  M d1/f
  A d3/g
  R d7/f
Make it look like a repo from before narrow+share was supported
  $ rm .hg/narrowspec.dirstate
  $ hg ci -Am test
  abort: working copy's narrowspec is stale
  (run 'hg tracked --update-working-copy')
  [255]
  $ hg tracked --update-working-copy
  $ hg st
  M d1/f
  A d3/g
  R d7/f
  $ cd ..