view tests/test-diff-unified.t @ 42222:57203e0210f8

copies: calculate mergecopies() based on pathcopies() When copies are stored in changesets, we need a changeset-centric version of mergecopies() just like we have a changeset-centric version of pathcopies(). I think the natural way of thinking about mergecopies() is in terms of pathcopies() from the base to each of the commits. So if we can rewrite mergecopies() based on two such pathcopies() calls, we'll get the changeset-centric version for free. That's what this patch does. A nice bonus is that it ends up being a lot simpler. mergecopies() has accumulated a lot of technical debt over time. One good example is the code for dealing with grafts (the "partial/incomplete/dirty" stuff). Since pathcopies() already deals with backwards renames and ping-pong renames, we get that for free. I've run tests with hard-coded debug logging for "fullcopy" and while I haven't looked at every difference it produces, all the ones I have looked at seemed reasonable to me. I'm a little surprised that no more tests fail when run with '--extra-config-opt experimental.copies.read-from=compatibility' compared to before this patch. This patch also fixes the broken cases in test-annotate.t and test-fastannotate.t. It also enables the part of test-copies.t that was previously disabled exactly because mergecopies() needed to get a changeset-centric version. One drawback of the rewritten code is that we may now make remotefilelog prefetch more files. We used to prefetch files that were unique to either side of the merge compared to the other. We now prefetch files that are unique to either side of the merge compared to the base. This means that if you added the same file to each side, we would not prefetch it before, but we would now. Such cases are probably quite rare, but one likely scenario where they happen is when moving from a commit to its successor (or the other way around). The user will probably already have the files in the cache in such cases, so it's probably not a big deal. Some timings for calculating mergecopies between two revisions (revisions shown on each line, all using the common ancestor as base): In the hg repo: 4.8 4.9: 0.21s -> 0.21s 4.0 4.8: 0.35s -> 0.63s In and old copy of the mozilla-unified repo: FIREFOX_BETA_60_BASE^ FIREFOX_BETA_60_BASE: 0.82s -> 0.82s FIREFOX_NIGHTLY_59_END FIREFOX_BETA_60_BASE: 2.5s -> 2.6s FIREFOX_BETA_59_END FIREFOX_BETA_60_BASE: 3.9s -> 4.1s FIREFOX_AURORA_50_BASE FIREFOX_BETA_60_BASE: 31s -> 33s So it's measurably slower in most cases. The most significant difference is in the hg repo between revisions 4.0 and 4.8. In that case it seems to come from the fact that pathcopies() uses fctx.isintroducedafter() (in _tracefile), while the old mergecopies() used fctx.linkrev() (in _checkcopies()). That results in a single call to filectx._adjustlinkrev(), which is responsible for the entire difference in time (in my repo). So we pay a performance penalty but we get more correct code (see change in test-mv-cp-st-diff.t). Deleting the "== f.filenode()" in _tracefile() recovers the lost performance in the hg repo. There were are few other optimizations in _checkcopies() that I could not measure any impact from. One was from the "seen" set. Another was from a "continue" when the file was not in the destination manifest (corresponding to "am" in _tracefile). Also note that merge copies are not calculated when updating with a clean working copy, which is probably the most common case. I therefore think the much simpler code is worth the slowdown. Differential Revision: https://phab.mercurial-scm.org/D6255
author Martin von Zweigbergk <martinvonz@google.com>
date Thu, 11 Apr 2019 23:22:54 -0700
parents 22f19da5736e
children df44bac9413d
line wrap: on
line source

  $ hg init repo
  $ cd repo
  $ cat > a <<EOF
  > c
  > c
  > a
  > a
  > b
  > a
  > a
  > c
  > c
  > EOF
  $ hg ci -Am adda
  adding a

  $ cat > a <<EOF
  > c
  > c
  > a
  > a
  > dd
  > a
  > a
  > c
  > c
  > EOF

default context

  $ hg diff --nodates
  diff -r cf9f4ba66af2 a
  --- a/a
  +++ b/a
  @@ -2,7 +2,7 @@
   c
   a
   a
  -b
  +dd
   a
   a
   c

invalid --unified

  $ hg diff --nodates -U foo
  abort: diff context lines count must be an integer, not 'foo'
  [255]


  $ hg diff --nodates -U 2
  diff -r cf9f4ba66af2 a
  --- a/a
  +++ b/a
  @@ -3,5 +3,5 @@
   a
   a
  -b
  +dd
   a
   a

  $ hg --config diff.unified=2 diff --nodates
  diff -r cf9f4ba66af2 a
  --- a/a
  +++ b/a
  @@ -3,5 +3,5 @@
   a
   a
  -b
  +dd
   a
   a

  $ hg diff --nodates -U 1
  diff -r cf9f4ba66af2 a
  --- a/a
  +++ b/a
  @@ -4,3 +4,3 @@
   a
  -b
  +dd
   a

invalid diff.unified

  $ hg --config diff.unified=foo diff --nodates
  abort: diff context lines count must be an integer, not 'foo'
  [255]

noprefix config and option

  $ hg --config diff.noprefix=True diff --nodates
  diff -r cf9f4ba66af2 a
  --- a
  +++ a
  @@ -2,7 +2,7 @@
   c
   a
   a
  -b
  +dd
   a
   a
   c
  $ hg diff --noprefix --nodates
  diff -r cf9f4ba66af2 a
  --- a
  +++ a
  @@ -2,7 +2,7 @@
   c
   a
   a
  -b
  +dd
   a
   a
   c

noprefix config disabled in plain mode, but option still enabled

  $ HGPLAIN=1 hg --config diff.noprefix=True diff --nodates
  diff -r cf9f4ba66af2 a
  --- a/a
  +++ b/a
  @@ -2,7 +2,7 @@
   c
   a
   a
  -b
  +dd
   a
   a
   c
  $ HGPLAIN=1 hg diff --noprefix --nodates
  diff -r cf9f4ba66af2 a
  --- a
  +++ a
  @@ -2,7 +2,7 @@
   c
   a
   a
  -b
  +dd
   a
   a
   c

  $ cd ..


0 lines of context hunk header matches gnu diff hunk header

  $ hg init diffzero
  $ cd diffzero
  $ cat > f1 << EOF
  > c2
  > c4
  > c5
  > EOF
  $ hg commit -Am0
  adding f1

  $ cat > f2 << EOF
  > c1
  > c2
  > c3
  > c4
  > EOF
  $ mv f2 f1
  $ hg diff -U0 --nodates
  diff -r 55d8ff78db23 f1
  --- a/f1
  +++ b/f1
  @@ -0,0 +1,1 @@
  +c1
  @@ -1,0 +3,1 @@
  +c3
  @@ -3,1 +4,0 @@
  -c5

  $ hg diff -U0 --nodates --git
  diff --git a/f1 b/f1
  --- a/f1
  +++ b/f1
  @@ -0,0 +1,1 @@
  +c1
  @@ -1,0 +3,1 @@
  +c3
  @@ -3,1 +4,0 @@
  -c5

  $ hg diff -U0 --nodates -p
  diff -r 55d8ff78db23 f1
  --- a/f1
  +++ b/f1
  @@ -0,0 +1,1 @@
  +c1
  @@ -1,0 +3,1 @@ c2
  +c3
  @@ -3,1 +4,0 @@ c4
  -c5

  $ echo a > f1
  $ hg ci -m movef2

Test diff headers terminating with TAB when necessary (issue3357)
Regular diff --nodates, file creation

  $ hg mv f1 'f 1'
  $ echo b > 'f 1'
  $ hg diff --nodates 'f 1'
  diff -r 7574207d0d15 f 1
  --- /dev/null
  +++ b/f 1	
  @@ -0,0 +1,1 @@
  +b

Git diff, adding space

  $ hg diff --git
  diff --git a/f1 b/f 1
  rename from f1
  rename to f 1
  --- a/f1
  +++ b/f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

Git diff, adding extended headers

  $ hg diff --git --config experimental.extendedheader.index=7 --config experimental.extendedheader.similarity=True
  diff --git a/f1 b/f 1
  similarity index 0%
  rename from f1
  rename to f 1
  index 7898192..6178079 100644
  --- a/f1
  +++ b/f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

  $ hg diff --git --config experimental.extendedheader.index=-1
  invalid length for extendedheader.index: '-1'
  diff --git a/f1 b/f 1
  rename from f1
  rename to f 1
  --- a/f1
  +++ b/f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

  $ hg diff --git --config experimental.extendedheader.index=whatever
  invalid value for extendedheader.index: 'whatever'
  diff --git a/f1 b/f 1
  rename from f1
  rename to f 1
  --- a/f1
  +++ b/f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

Git diff with noprefix

  $ hg --config diff.noprefix=True diff --git --nodates
  diff --git f1 f 1
  rename from f1
  rename to f 1
  --- f1
  +++ f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

noprefix config disabled in plain mode, but option still enabled

  $ HGPLAIN=1 hg --config diff.noprefix=True diff --git --nodates
  diff --git a/f1 b/f 1
  rename from f1
  rename to f 1
  --- a/f1
  +++ b/f 1	
  @@ -1,1 +1,1 @@
  -a
  +b
  $ HGPLAIN=1 hg diff --git --noprefix --nodates
  diff --git f1 f 1
  rename from f1
  rename to f 1
  --- f1
  +++ f 1	
  @@ -1,1 +1,1 @@
  -a
  +b

Regular diff --nodates, file deletion

  $ hg ci -m addspace
  $ hg mv 'f 1' f1
  $ echo a > f1
  $ hg diff --nodates 'f 1'
  diff -r ca50fe67c9c7 f 1
  --- a/f 1	
  +++ /dev/null
  @@ -1,1 +0,0 @@
  -b

Git diff, removing space

  $ hg diff --git
  diff --git a/f 1 b/f1
  rename from f 1
  rename to f1
  --- a/f 1	
  +++ b/f1
  @@ -1,1 +1,1 @@
  -b
  +a

showfunc diff
  $ cat > f1 << EOF
  > int main() {
  >     int a = 0;
  >     int b = 1;
  >     int c = 2;
  >     int d = 3;
  >     return a + b + c + d;
  > }
  > EOF
  $ hg commit -m addfunction
  $ cat > f1 << EOF
  > int main() {
  >     int a = 0;
  >     int b = 1;
  >     int c = 2;
  >     int e = 3;
  >     return a + b + c + e;
  > }
  > EOF
  $ hg diff --git
  diff --git a/f1 b/f1
  --- a/f1
  +++ b/f1
  @@ -2,6 +2,6 @@
       int a = 0;
       int b = 1;
       int c = 2;
  -    int d = 3;
  -    return a + b + c + d;
  +    int e = 3;
  +    return a + b + c + e;
   }
  $ hg diff --config diff.showfunc=True --git
  diff --git a/f1 b/f1
  --- a/f1
  +++ b/f1
  @@ -2,6 +2,6 @@ int main() {
       int a = 0;
       int b = 1;
       int c = 2;
  -    int d = 3;
  -    return a + b + c + d;
  +    int e = 3;
  +    return a + b + c + e;
   }

If [diff] git is set to true, but the user says --no-git, we should
*not* get git diffs
  $ hg diff --nodates --config diff.git=1 --no-git
  diff -r f2c7c817fa55 f1
  --- a/f1
  +++ b/f1
  @@ -2,6 +2,6 @@
       int a = 0;
       int b = 1;
       int c = 2;
  -    int d = 3;
  -    return a + b + c + d;
  +    int e = 3;
  +    return a + b + c + e;
   }

  $ cd ..

Long function names should be abbreviated, but multi-byte character shouldn't
be broken up

  $ hg init longfunc
  $ cd longfunc

  >>> with open('a', 'wb') as f:
  ...     f.write(b'a' * 39 + b'bb' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 0 b\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b'a' * 39 + b'\xc3\xa0' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 0 a with grave (single code point)\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b'a' * 39 + b'a\xcc\x80' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 0 a with grave (composition)\n') and None
  ...     f.write(b' .\n' * 3) and None
  $ hg ci -qAm0

  >>> with open('a', 'wb') as f:
  ...     f.write(b'a' * 39 + b'bb' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 1 b\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b'a' * 39 + b'\xc3\xa0' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 1 a with grave (single code point)\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b'a' * 39 + b'a\xcc\x80' + b'\n') and None
  ...     f.write(b' .\n' * 3) and None
  ...     f.write(b' 1 a with grave (composition)\n') and None
  ...     f.write(b' .\n' * 3) and None
  $ hg ci -m1

  $ hg diff -c1 --nodates --show-function
  diff -r 3e92dd6fa812 -r a256341606cb a
  --- a/a
  +++ b/a
  @@ -2,7 +2,7 @@ aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaab
    .
    .
    .
  - 0 b
  + 1 b
    .
    .
    .
  @@ -10,7 +10,7 @@ aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\xc3\xa0 (esc)
    .
    .
    .
  - 0 a with grave (single code point)
  + 1 a with grave (single code point)
    .
    .
    .
  @@ -18,7 +18,7 @@ aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\xcc\x80 (esc)
    .
    .
    .
  - 0 a with grave (composition)
  + 1 a with grave (composition)
    .
    .
    .

  $ cd ..