view tests/test-export.t @ 44950:f9734b2d59cc

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)

  $ hg init repo
  $ cd repo
  $ touch foo
  $ hg add foo
  $ for i in 0 1 2 3 4 5 6 7 8 9 10 11; do
  >    echo "foo-$i" >> foo
  >    hg ci -m "foo-$i"
  > done

  $ for out in "%nof%N" "%%%H" "%b-%R" "%h" "%r" "%m"; do
  >    echo
  >    echo "# foo-$out.patch"
  >    hg export -v -o "foo-$out.patch" 2:tip
  > done
  
  # foo-%nof%N.patch
  exporting patches:
  foo-01of10.patch
  foo-02of10.patch
  foo-03of10.patch
  foo-04of10.patch
  foo-05of10.patch
  foo-06of10.patch
  foo-07of10.patch
  foo-08of10.patch
  foo-09of10.patch
  foo-10of10.patch
  
  # foo-%%%H.patch
  exporting patches:
  foo-%617188a1c80f869a7b66c85134da88a6fb145f67.patch
  foo-%dd41a5ff707a5225204105611ba49cc5c229d55f.patch
  foo-%f95a5410f8664b6e1490a4af654e4b7d41a7b321.patch
  foo-%4346bcfde53b4d9042489078bcfa9c3e28201db2.patch
  foo-%afda8c3a009cc99449a05ad8aa4655648c4ecd34.patch
  foo-%35284ce2b6b99c9d2ac66268fe99e68e1974e1aa.patch
  foo-%9688c41894e6931305fa7165a37f6568050b4e9b.patch
  foo-%747d3c68f8ec44bb35816bfcd59aeb50b9654c2f.patch
  foo-%5f17a83f5fbd9414006a5e563eab4c8a00729efd.patch
  foo-%f3acbafac161ec68f1598af38f794f28847ca5d3.patch
  
  # foo-%b-%R.patch
  exporting patches:
  foo-repo-2.patch
  foo-repo-3.patch
  foo-repo-4.patch
  foo-repo-5.patch
  foo-repo-6.patch
  foo-repo-7.patch
  foo-repo-8.patch
  foo-repo-9.patch
  foo-repo-10.patch
  foo-repo-11.patch
  
  # foo-%h.patch
  exporting patches:
  foo-617188a1c80f.patch
  foo-dd41a5ff707a.patch
  foo-f95a5410f866.patch
  foo-4346bcfde53b.patch
  foo-afda8c3a009c.patch
  foo-35284ce2b6b9.patch
  foo-9688c41894e6.patch
  foo-747d3c68f8ec.patch
  foo-5f17a83f5fbd.patch
  foo-f3acbafac161.patch
  
  # foo-%r.patch
  exporting patches:
  foo-02.patch
  foo-03.patch
  foo-04.patch
  foo-05.patch
  foo-06.patch
  foo-07.patch
  foo-08.patch
  foo-09.patch
  foo-10.patch
  foo-11.patch
  
  # foo-%m.patch
  exporting patches:
  foo-foo_2.patch
  foo-foo_3.patch
  foo-foo_4.patch
  foo-foo_5.patch
  foo-foo_6.patch
  foo-foo_7.patch
  foo-foo_8.patch
  foo-foo_9.patch
  foo-foo_10.patch
  foo-foo_11.patch

Doing it again clobbers the files rather than appending:
  $ hg export -v -o "foo-%m.patch" 2:3
  exporting patches:
  foo-foo_2.patch
  foo-foo_3.patch
  $ grep HG foo-foo_2.patch | wc -l
  \s*1 (re)
  $ grep HG foo-foo_3.patch | wc -l
  \s*1 (re)

Using bookmarks:

  $ hg book -f -r 9 @
  $ hg book -f -r 11 test
  $ hg export -B test
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID 5f17a83f5fbd9414006a5e563eab4c8a00729efd
  # Parent  747d3c68f8ec44bb35816bfcd59aeb50b9654c2f
  foo-10
  
  diff -r 747d3c68f8ec -r 5f17a83f5fbd foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -8,3 +8,4 @@
   foo-7
   foo-8
   foo-9
  +foo-10
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID f3acbafac161ec68f1598af38f794f28847ca5d3
  # Parent  5f17a83f5fbd9414006a5e563eab4c8a00729efd
  foo-11
  
  diff -r 5f17a83f5fbd -r f3acbafac161 foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -9,3 +9,4 @@
   foo-8
   foo-9
   foo-10
  +foo-11

Exporting 4 changesets to a file:

  $ hg export -o export_internal 1 2 3 4
  $ grep HG export_internal | wc -l
  \s*4 (re)

Doing it again clobbers the file rather than appending:
  $ hg export -o export_internal 1 2 3 4
  $ grep HG export_internal | wc -l
  \s*4 (re)

Exporting 4 changesets to stdout:

  $ hg export 1 2 3 4 | grep HG | wc -l
  \s*4 (re)

Exporting revision -2 to a file:

  $ hg export -- -2
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID 5f17a83f5fbd9414006a5e563eab4c8a00729efd
  # Parent  747d3c68f8ec44bb35816bfcd59aeb50b9654c2f
  foo-10
  
  diff -r 747d3c68f8ec -r 5f17a83f5fbd foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -8,3 +8,4 @@
   foo-7
   foo-8
   foo-9
  +foo-10

Exporting wdir revision:

  $ echo "foo-wdir" >> foo
  $ hg export 'wdir()'
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID ffffffffffffffffffffffffffffffffffffffff
  # Parent  f3acbafac161ec68f1598af38f794f28847ca5d3
  
  
  diff -r f3acbafac161 foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -10,3 +10,4 @@
   foo-9
   foo-10
   foo-11
  +foo-wdir
  $ hg revert -q foo

Templated output to stdout:

  $ hg export -Tjson 0
  [
   {
    "branch": "default",
    "date": [0, 0],
    "desc": "foo-0",
    "diff": "diff -r 000000000000 -r 871558de6af2 foo\n--- /dev/null\tThu Jan 01 00:00:00 1970 +0000\n+++ b/foo\tThu Jan 01 00:00:00 1970 +0000\n@@ -0,0 +1,1 @@\n+foo-0\n",
    "node": "871558de6af2e8c244222f8eea69b782c94ce3df",
    "parents": [],
    "user": "test"
   }
  ]

Templated output to single file:

  $ hg export -Tjson 0:1 -o out.json
  $ cat out.json
  [
   {
    "branch": "default",
    "date": [0, 0],
    "desc": "foo-0",
    "diff": "diff -r 000000000000 -r 871558de6af2 foo\n--- /dev/null\tThu Jan 01 00:00:00 1970 +0000\n+++ b/foo\tThu Jan 01 00:00:00 1970 +0000\n@@ -0,0 +1,1 @@\n+foo-0\n",
    "node": "871558de6af2e8c244222f8eea69b782c94ce3df",
    "parents": [],
    "user": "test"
   },
   {
    "branch": "default",
    "date": [0, 0],
    "desc": "foo-1",
    "diff": "diff -r 871558de6af2 -r d1c9656e973c foo\n--- a/foo\tThu Jan 01 00:00:00 1970 +0000\n+++ b/foo\tThu Jan 01 00:00:00 1970 +0000\n@@ -1,1 +1,2 @@\n foo-0\n+foo-1\n",
    "node": "d1c9656e973cfb5aebd5499bbd2cb350e3b12266",
    "parents": ["871558de6af2e8c244222f8eea69b782c94ce3df"],
    "user": "test"
   }
  ]

Templated output to multiple files:

  $ hg export -Tjson 0:1 -o 'out-{rev}.json'
  $ cat out-0.json
  [
   {
    "branch": "default",
    "date": [0, 0],
    "desc": "foo-0",
    "diff": "diff -r 000000000000 -r 871558de6af2 foo\n--- /dev/null\tThu Jan 01 00:00:00 1970 +0000\n+++ b/foo\tThu Jan 01 00:00:00 1970 +0000\n@@ -0,0 +1,1 @@\n+foo-0\n",
    "node": "871558de6af2e8c244222f8eea69b782c94ce3df",
    "parents": [],
    "user": "test"
   }
  ]
  $ cat out-1.json
  [
   {
    "branch": "default",
    "date": [0, 0],
    "desc": "foo-1",
    "diff": "diff -r 871558de6af2 -r d1c9656e973c foo\n--- a/foo\tThu Jan 01 00:00:00 1970 +0000\n+++ b/foo\tThu Jan 01 00:00:00 1970 +0000\n@@ -1,1 +1,2 @@\n foo-0\n+foo-1\n",
    "node": "d1c9656e973cfb5aebd5499bbd2cb350e3b12266",
    "parents": ["871558de6af2e8c244222f8eea69b782c94ce3df"],
    "user": "test"
   }
  ]

Template keywrods:

  $ hg export 0 -T '# {node|shortest}\n\n{diff}'
  # 8715
  
  diff -r 000000000000 -r 871558de6af2 foo
  --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -0,0 +1,1 @@
  +foo-0

No filename should be printed if stdout is specified explicitly:

  $ hg export -v 1 -o -
  exporting patch:
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID d1c9656e973cfb5aebd5499bbd2cb350e3b12266
  # Parent  871558de6af2e8c244222f8eea69b782c94ce3df
  foo-1
  
  diff -r 871558de6af2 -r d1c9656e973c foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -1,1 +1,2 @@
   foo-0
  +foo-1

Checking if only alphanumeric characters are used in the file name (%m option):

  $ echo "line" >> foo
  $ hg commit -m " !\"#$%&(,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]"'^'"_\`abcdefghijklmnopqrstuvwxyz{|}~"
  $ hg export -v -o %m.patch tip
  exporting patch:
  ___________0123456789_______ABCDEFGHIJKLMNOPQRSTUVWXYZ______abcdefghijklmnopqrstuvwxyz____.patch

Template fragments in file name:

  $ hg export -v -o '{node|shortest}.patch' tip
  exporting patch:
  197e.patch

Backslash should be preserved because it is a directory separator on Windows:

  $ mkdir out
  $ hg export -v -o 'out\{node|shortest}.patch' tip
  exporting patch:
  out\197e.patch

Still backslash is taken as an escape character in inner template strings:

  $ hg export -v -o '{"out\{foo}.patch"}' tip
  exporting patch:
  out{foo}.patch

Invalid pattern in file name:

  $ hg export -o '%x.patch' tip
  abort: invalid format spec '%x' in output filename
  [255]
  $ hg export -o '%' tip
  abort: incomplete format spec in output filename
  [255]
  $ hg export -o '%{"foo"}' tip
  abort: incomplete format spec in output filename
  [255]
  $ hg export -o '%m{' tip
  hg: parse error at 3: unterminated template expansion
  (%m{
      ^ here)
  [255]
  $ hg export -o '%\' tip
  abort: invalid format spec '%\' in output filename
  [255]
  $ hg export -o '\%' tip
  abort: incomplete format spec in output filename
  [255]

Catch exporting unknown revisions (especially empty revsets, see issue3353)

  $ hg export
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID 197ecd81a57f760b54f34a58817ad5b04991fa47
  # Parent  f3acbafac161ec68f1598af38f794f28847ca5d3
   !"#$%&(,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
  
  diff -r f3acbafac161 -r 197ecd81a57f foo
  --- a/foo	Thu Jan 01 00:00:00 1970 +0000
  +++ b/foo	Thu Jan 01 00:00:00 1970 +0000
  @@ -10,3 +10,4 @@
   foo-9
   foo-10
   foo-11
  +line

  $ hg export ""
  hg: parse error: empty query
  [255]
  $ hg export 999
  abort: unknown revision '999'!
  [255]
  $ hg export "not all()"
  abort: export requires at least one changeset
  [255]

Check for color output
  $ cat <<EOF >> $HGRCPATH
  > [color]
  > mode = ansi
  > [extensions]
  > color =
  > EOF

  $ hg export --color always --nodates tip
  # HG changeset patch
  # User test
  # Date 0 0
  #      Thu Jan 01 00:00:00 1970 +0000
  # Node ID * (glob)
  # Parent * (glob)
   !"#$%&(,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
  
  \x1b[0;1mdiff -r f3acbafac161 -r 197ecd81a57f foo\x1b[0m (esc)
  \x1b[0;31;1m--- a/foo\x1b[0m (esc)
  \x1b[0;32;1m+++ b/foo\x1b[0m (esc)
  \x1b[0;35m@@ -10,3 +10,4 @@\x1b[0m (esc)
   foo-9
   foo-10
   foo-11
  \x1b[0;32m+line\x1b[0m (esc)


  $ cd ..