view tests/test-churn.t @ 26379:39d643252b9f

revlog: use existing file handle when reading during _addrevision

_addrevision() may need to read from revlogs as part of computing
deltas. Previously, we would flush existing file handles and open
a new, short-lived file handle to perform the reading.

If we have an existing file handle, it seems logical to reuse it
for reading instead of opening a new file handle. This patch
makes that the new behavior.

After this patch, revlog files are only reopened when adding
revisions if the revlog is switched from inline to non-inline.

On Linux when unbundling a bundle of the mozilla-central repo, this
patch has the following impact on system call counts:

Call     Before     After       Delta
write    827,639    673,390   -154,249
open     700,103    684,089    -16,014
read      74,489     74,489          0
fstat    493,924    461,896    -32,028
close    249,131    233,117    -16,014
stat     242,001    242,001          0
lstat     18,676     18,676          0
lseek     20,268     20,268          0
ioctl     14,652     13,173     -1,479
TOTAL  3,180,758  2,930,679   -250,079

It's worth noting that many of the open() calls fail due to missing
files. That's why there are many more open() calls than close().

Despite the significant system call reduction, this change does not
seem to have a significant performance impact on Linux.

On Windows 10 (not a VM, on a SSD), this patch appears to reduce
unbundle time for mozilla-central from ~960s to ~920s. This isn't
as significant as I was hoping. But a decrease it is nonetheless.
Still, Windows unbundle performance is still >2x slower than Linux.

Despite the lack of significant gains, fewer system calls is fewer
system calls. If nothing else, this will narrow the focus of potential
areas to optimize in the future.

  $ echo "[extensions]" >> $HGRCPATH
  $ echo "churn=" >> $HGRCPATH

create test repository

  $ hg init repo
  $ cd repo
  $ echo a > a
  $ hg ci -Am adda -u user1 -d 6:00
  adding a
  $ echo b >> a
  $ echo b > b
  $ hg ci -m changeba -u user2 -d 9:00 a
  $ hg ci -Am addb -u user2 -d 9:30
  adding b
  $ echo c >> a
  $ echo c >> b
  $ echo c > c
  $ hg ci -m changeca -u user3 -d 12:00 a
  $ hg ci -m changecb -u user3 -d 12:15 b
  $ hg ci -Am addc -u user3 -d 12:30
  adding c
  $ mkdir -p d/e
  $ echo abc > d/e/f1.txt
  $ hg ci -Am "add d/e/f1.txt" -u user1 -d 12:45 d/e/f1.txt
  $ mkdir -p d/g
  $ echo def > d/g/f2.txt
  $ hg ci -Am "add d/g/f2.txt" -u user1 -d 13:00 d/g/f2.txt


churn separate directories

  $ cd d
  $ hg churn e
  user1      1 ***************************************************************

churn all

  $ hg churn
  user1      3 ***************************************************************
  user3      3 ***************************************************************
  user2      2 ******************************************

churn excluding one dir

  $ hg churn -X e
  user3      3 ***************************************************************
  user1      2 ******************************************
  user2      2 ******************************************

churn up to rev 2

  $ hg churn -r :2
  user2      2 ***************************************************************
  user1      1 ********************************
  $ cd ..

churn with aliases

  $ cat > ../aliases <<EOF
  > user1 alias1
  > user3 alias3
  > not-an-alias
  > EOF

churn with .hgchurn

  $ mv ../aliases .hgchurn
  $ hg churn
  skipping malformed alias: not-an-alias
  alias1      3 **************************************************************
  alias3      3 **************************************************************
  user2       2 *****************************************
  $ rm .hgchurn

churn with column specifier

  $ COLUMNS=40 hg churn
  user1      3 ***********************
  user3      3 ***********************
  user2      2 ***************

churn by hour

  $ hg churn -f '%H' -s
  06      1 *****************
  09      2 *********************************
  12      4 ******************************************************************
  13      1 *****************


churn with separated added/removed lines

  $ hg rm d/g/f2.txt
  $ hg ci -Am "removed d/g/f2.txt" -u user1 -d 14:00 d/g/f2.txt
  $ hg churn --diffstat
  user1           +3/-1 +++++++++++++++++++++++++++++++++++++++++--------------
  user3           +3/-0 +++++++++++++++++++++++++++++++++++++++++
  user2           +2/-0 +++++++++++++++++++++++++++

churn --diffstat with color

  $ hg --config extensions.color= churn --config color.mode=ansi \
  >     --diffstat --color=always
  user1           +3/-1 \x1b[0;32m+++++++++++++++++++++++++++++++++++++++++\x1b[0m\x1b[0;31m--------------\x1b[0m (esc)
  user3           +3/-0 \x1b[0;32m+++++++++++++++++++++++++++++++++++++++++\x1b[0m (esc)
  user2           +2/-0 \x1b[0;32m+++++++++++++++++++++++++++\x1b[0m (esc)


changeset number churn

  $ hg churn -c
  user1      4 ***************************************************************
  user3      3 ***********************************************
  user2      2 ********************************

  $ echo 'with space = no-space' >> ../aliases
  $ echo a >> a
  $ hg commit -m a -u 'with space' -d 15:00

churn with space in alias

  $ hg churn --aliases ../aliases -r tip
  no-space      1 ************************************************************

  $ cd ..


Issue833: ZeroDivisionError

  $ hg init issue-833
  $ cd issue-833
  $ touch foo
  $ hg ci -Am foo
  adding foo

this was failing with a ZeroDivisionError

  $ hg churn
  test      0 
  $ cd ..

Ignore trailing or leading spaces in emails

  $ cd repo
  $ touch bar
  $ hg ci -Am'bar' -u 'user4 <[e-mail]>'
  adding bar
  $ touch foo
  $ hg ci -Am'foo' -u 'user4 < [e-mail] >'
  adding foo
  $ hg log -l2 --template '[{author|email}]\n'
  [ [e-mail] ]
  [[e-mail]]
  $ hg churn -c
  user1            4 *********************************************************
  user3            3 *******************************************
  user2            2 *****************************
  [e-mail]      2 *****************************
  with space       1 **************

Test multibyte sequences in names

  $ echo bar >> bar
  $ hg --encoding utf-8 ci -m'changed bar' -u 'El Niño <[e-mail]>'
  $ hg --encoding utf-8 churn -ct '{author|person}'
  user1           4 **********************************************************
  user3           3 ********************************************
  user2           2 *****************************
  user4           2 *****************************
  El Ni\xc3\xb1o         1 *************** (esc)
  with space      1 ***************

Test --template argument, with backwards compatiblity

  $ hg churn -t '{author|user}'
  user1      4 ***************************************************************
  user3      3 ***********************************************
  user2      2 ********************************
  nino       1 ****************
  with       1 ****************
             0 
  user4      0 
  $ hg churn -T '{author|user}'
  user1      4 ***************************************************************
  user3      3 ***********************************************
  user2      2 ********************************
  nino       1 ****************
  with       1 ****************
             0 
  user4      0 
  $ hg churn -t 'alltogether'
  alltogether     11 *********************************************************
  $ hg churn -T 'alltogether'
  alltogether     11 *********************************************************

  $ cd ..