merge(third_party/git): Merge squashed git subtree at v2.23.0

Merge commit '1b593e1ea4' as 'third_party/git'
This commit is contained in:
Vincent Ambo 2020-01-11 23:36:56 +00:00
commit 7ef0d62730
3629 changed files with 1139935 additions and 0 deletions

View file

@ -0,0 +1,2 @@
shebang.perl
diff-highlight

View file

@ -0,0 +1,285 @@
package DiffHighlight;
use 5.008;
use warnings FATAL => 'all';
use strict;
# Use the correct value for both UNIX and Windows (/dev/null vs nul)
use File::Spec;
my $NULL = File::Spec->devnull();
# Highlight by reversing foreground and background. You could do
# other things like bold or underline if you prefer.
my @OLD_HIGHLIGHT = (
color_config('color.diff-highlight.oldnormal'),
color_config('color.diff-highlight.oldhighlight', "\x1b[7m"),
color_config('color.diff-highlight.oldreset', "\x1b[27m")
);
my @NEW_HIGHLIGHT = (
color_config('color.diff-highlight.newnormal', $OLD_HIGHLIGHT[0]),
color_config('color.diff-highlight.newhighlight', $OLD_HIGHLIGHT[1]),
color_config('color.diff-highlight.newreset', $OLD_HIGHLIGHT[2])
);
my $RESET = "\x1b[m";
my $COLOR = qr/\x1b\[[0-9;]*m/;
my $BORING = qr/$COLOR|\s/;
my @removed;
my @added;
my $in_hunk;
my $graph_indent = 0;
our $line_cb = sub { print @_ };
our $flush_cb = sub { local $| = 1 };
# Count the visible width of a string, excluding any terminal color sequences.
sub visible_width {
local $_ = shift;
my $ret = 0;
while (length) {
if (s/^$COLOR//) {
# skip colors
} elsif (s/^.//) {
$ret++;
}
}
return $ret;
}
# Return a substring of $str, omitting $len visible characters from the
# beginning, where terminal color sequences do not count as visible.
sub visible_substr {
my ($str, $len) = @_;
while ($len > 0) {
if ($str =~ s/^$COLOR//) {
next
}
$str =~ s/^.//;
$len--;
}
return $str;
}
sub handle_line {
my $orig = shift;
local $_ = $orig;
# match a graph line that begins a commit
if (/^(?:$COLOR?\|$COLOR?[ ])* # zero or more leading "|" with space
$COLOR?\*$COLOR?[ ] # a "*" with its trailing space
(?:$COLOR?\|$COLOR?[ ])* # zero or more trailing "|"
[ ]* # trailing whitespace for merges
/x) {
my $graph_prefix = $&;
# We must flush before setting graph indent, since the
# new commit may be indented differently from what we
# queued.
flush();
$graph_indent = visible_width($graph_prefix);
} elsif ($graph_indent) {
if (length($_) < $graph_indent) {
$graph_indent = 0;
} else {
$_ = visible_substr($_, $graph_indent);
}
}
if (!$in_hunk) {
$line_cb->($orig);
$in_hunk = /^$COLOR*\@\@ /;
}
elsif (/^$COLOR*-/) {
push @removed, $orig;
}
elsif (/^$COLOR*\+/) {
push @added, $orig;
}
else {
flush();
$line_cb->($orig);
$in_hunk = /^$COLOR*[\@ ]/;
}
# Most of the time there is enough output to keep things streaming,
# but for something like "git log -Sfoo", you can get one early
# commit and then many seconds of nothing. We want to show
# that one commit as soon as possible.
#
# Since we can receive arbitrary input, there's no optimal
# place to flush. Flushing on a blank line is a heuristic that
# happens to match git-log output.
if (!length) {
$flush_cb->();
}
}
sub flush {
# Flush any queued hunk (this can happen when there is no trailing
# context in the final diff of the input).
show_hunk(\@removed, \@added);
@removed = ();
@added = ();
}
sub highlight_stdin {
while (<STDIN>) {
handle_line($_);
}
flush();
}
# Ideally we would feed the default as a human-readable color to
# git-config as the fallback value. But diff-highlight does
# not otherwise depend on git at all, and there are reports
# of it being used in other settings. Let's handle our own
# fallback, which means we will work even if git can't be run.
sub color_config {
my ($key, $default) = @_;
my $s = `git config --get-color $key 2>$NULL`;
return length($s) ? $s : $default;
}
sub show_hunk {
my ($a, $b) = @_;
# If one side is empty, then there is nothing to compare or highlight.
if (!@$a || !@$b) {
$line_cb->(@$a, @$b);
return;
}
# If we have mismatched numbers of lines on each side, we could try to
# be clever and match up similar lines. But for now we are simple and
# stupid, and only handle multi-line hunks that remove and add the same
# number of lines.
if (@$a != @$b) {
$line_cb->(@$a, @$b);
return;
}
my @queue;
for (my $i = 0; $i < @$a; $i++) {
my ($rm, $add) = highlight_pair($a->[$i], $b->[$i]);
$line_cb->($rm);
push @queue, $add;
}
$line_cb->(@queue);
}
sub highlight_pair {
my @a = split_line(shift);
my @b = split_line(shift);
# Find common prefix, taking care to skip any ansi
# color codes.
my $seen_plusminus;
my ($pa, $pb) = (0, 0);
while ($pa < @a && $pb < @b) {
if ($a[$pa] =~ /$COLOR/) {
$pa++;
}
elsif ($b[$pb] =~ /$COLOR/) {
$pb++;
}
elsif ($a[$pa] eq $b[$pb]) {
$pa++;
$pb++;
}
elsif (!$seen_plusminus && $a[$pa] eq '-' && $b[$pb] eq '+') {
$seen_plusminus = 1;
$pa++;
$pb++;
}
else {
last;
}
}
# Find common suffix, ignoring colors.
my ($sa, $sb) = ($#a, $#b);
while ($sa >= $pa && $sb >= $pb) {
if ($a[$sa] =~ /$COLOR/) {
$sa--;
}
elsif ($b[$sb] =~ /$COLOR/) {
$sb--;
}
elsif ($a[$sa] eq $b[$sb]) {
$sa--;
$sb--;
}
else {
last;
}
}
if (is_pair_interesting(\@a, $pa, $sa, \@b, $pb, $sb)) {
return highlight_line(\@a, $pa, $sa, \@OLD_HIGHLIGHT),
highlight_line(\@b, $pb, $sb, \@NEW_HIGHLIGHT);
}
else {
return join('', @a),
join('', @b);
}
}
# we split either by $COLOR or by character. This has the side effect of
# leaving in graph cruft. It works because the graph cruft does not contain "-"
# or "+"
sub split_line {
local $_ = shift;
return utf8::decode($_) ?
map { utf8::encode($_); $_ }
map { /$COLOR/ ? $_ : (split //) }
split /($COLOR+)/ :
map { /$COLOR/ ? $_ : (split //) }
split /($COLOR+)/;
}
sub highlight_line {
my ($line, $prefix, $suffix, $theme) = @_;
my $start = join('', @{$line}[0..($prefix-1)]);
my $mid = join('', @{$line}[$prefix..$suffix]);
my $end = join('', @{$line}[($suffix+1)..$#$line]);
# If we have a "normal" color specified, then take over the whole line.
# Otherwise, we try to just manipulate the highlighted bits.
if (defined $theme->[0]) {
s/$COLOR//g for ($start, $mid, $end);
chomp $end;
return join('',
$theme->[0], $start, $RESET,
$theme->[1], $mid, $RESET,
$theme->[0], $end, $RESET,
"\n"
);
} else {
return join('',
$start,
$theme->[1], $mid, $theme->[2],
$end
);
}
}
# Pairs are interesting to highlight only if we are going to end up
# highlighting a subset (i.e., not the whole line). Otherwise, the highlighting
# is just useless noise. We can detect this by finding either a matching prefix
# or suffix (disregarding boring bits like whitespace and colorization).
sub is_pair_interesting {
my ($a, $pa, $sa, $b, $pb, $sb) = @_;
my $prefix_a = join('', @$a[0..($pa-1)]);
my $prefix_b = join('', @$b[0..($pb-1)]);
my $suffix_a = join('', @$a[($sa+1)..$#$a]);
my $suffix_b = join('', @$b[($sb+1)..$#$b]);
return visible_substr($prefix_a, $graph_indent) !~ /^$COLOR*-$BORING*$/ ||
visible_substr($prefix_b, $graph_indent) !~ /^$COLOR*\+$BORING*$/ ||
$suffix_a !~ /^$BORING*$/ ||
$suffix_b !~ /^$BORING*$/;
}

View file

@ -0,0 +1,23 @@
all: diff-highlight
PERL_PATH = /usr/bin/perl
-include ../../config.mak
PERL_PATH_SQ = $(subst ','\'',$(PERL_PATH))
diff-highlight: shebang.perl DiffHighlight.pm diff-highlight.perl
cat $^ >$@+
chmod +x $@+
mv $@+ $@
shebang.perl: FORCE
@echo '#!$(PERL_PATH_SQ)' >$@+
@cmp $@+ $@ >/dev/null 2>/dev/null || mv $@+ $@
test: all
$(MAKE) -C t
clean:
$(RM) diff-highlight
.PHONY: FORCE

View file

@ -0,0 +1,223 @@
diff-highlight
==============
Line oriented diffs are great for reviewing code, because for most
hunks, you want to see the old and the new segments of code next to each
other. Sometimes, though, when an old line and a new line are very
similar, it's hard to immediately see the difference.
You can use "--color-words" to highlight only the changed portions of
lines. However, this can often be hard to read for code, as it loses
the line structure, and you end up with oddly formatted bits.
Instead, this script post-processes the line-oriented diff, finds pairs
of lines, and highlights the differing segments. It's currently very
simple and stupid about doing these tasks. In particular:
1. It will only highlight hunks in which the number of removed and
added lines is the same, and it will pair lines within the hunk by
position (so the first removed line is compared to the first added
line, and so forth). This is simple and tends to work well in
practice. More complex changes don't highlight well, so we tend to
exclude them due to the "same number of removed and added lines"
restriction. Or even if we do try to highlight them, they end up
not highlighting because of our "don't highlight if the whole line
would be highlighted" rule.
2. It will find the common prefix and suffix of two lines, and
consider everything in the middle to be "different". It could
instead do a real diff of the characters between the two lines and
find common subsequences. However, the point of the highlight is to
call attention to a certain area. Even if some small subset of the
highlighted area actually didn't change, that's OK. In practice it
ends up being more readable to just have a single blob on the line
showing the interesting bit.
The goal of the script is therefore not to be exact about highlighting
changes, but to call attention to areas of interest without being
visually distracting. Non-diff lines and existing diff coloration is
preserved; the intent is that the output should look exactly the same as
the input, except for the occasional highlight.
Use
---
You can try out the diff-highlight program with:
---------------------------------------------
git log -p --color | /path/to/diff-highlight
---------------------------------------------
If you want to use it all the time, drop it in your $PATH and put the
following in your git configuration:
---------------------------------------------
[pager]
log = diff-highlight | less
show = diff-highlight | less
diff = diff-highlight | less
---------------------------------------------
Color Config
------------
You can configure the highlight colors and attributes using git's
config. The colors for "old" and "new" lines can be specified
independently. There are two "modes" of configuration:
1. You can specify a "highlight" color and a matching "reset" color.
This will retain any existing colors in the diff, and apply the
"highlight" and "reset" colors before and after the highlighted
portion.
2. You can specify a "normal" color and a "highlight" color. In this
case, existing colors are dropped from that line. The non-highlighted
bits of the line get the "normal" color, and the highlights get the
"highlight" color.
If no "new" colors are specified, they default to the "old" colors. If
no "old" colors are specified, the default is to reverse the foreground
and background for highlighted portions.
Examples:
---------------------------------------------
# Underline highlighted portions
[color "diff-highlight"]
oldHighlight = ul
oldReset = noul
---------------------------------------------
---------------------------------------------
# Varying background intensities
[color "diff-highlight"]
oldNormal = "black #f8cbcb"
oldHighlight = "black #ffaaaa"
newNormal = "black #cbeecb"
newHighlight = "black #aaffaa"
---------------------------------------------
Using diff-highlight as a module
--------------------------------
If you want to pre- or post- process the highlighted lines as part of
another perl script, you can use the DiffHighlight module. You can
either "require" it or just cat the module together with your script (to
avoid run-time dependencies).
Your script may set up one or more of the following variables:
- $DiffHighlight::line_cb - this should point to a function which is
called whenever DiffHighlight has lines (which may contain
highlights) to output. The default function prints each line to
stdout. Note that the function may be called with multiple lines.
- $DiffHighlight::flush_cb - this should point to a function which
flushes the output (because DiffHighlight believes it has completed
processing a logical chunk of input). The default function flushes
stdout.
The script may then feed lines, one at a time, to DiffHighlight::handle_line().
When lines are done processing, they will be fed to $line_cb. Note that
DiffHighlight may queue up many input lines (to analyze a whole hunk)
before calling $line_cb. After providing all lines, call
DiffHighlight::flush() to flush any unprocessed lines.
If you just want to process stdin, DiffHighlight::highlight_stdin()
is a convenience helper which will loop and flush for you.
Bugs
----
Because diff-highlight relies on heuristics to guess which parts of
changes are important, there are some cases where the highlighting is
more distracting than useful. Fortunately, these cases are rare in
practice, and when they do occur, the worst case is simply a little
extra highlighting. This section documents some cases known to be
sub-optimal, in case somebody feels like working on improving the
heuristics.
1. Two changes on the same line get highlighted in a blob. For example,
highlighting:
----------------------------------------------
-foo(buf, size);
+foo(obj->buf, obj->size);
----------------------------------------------
yields (where the inside of "+{}" would be highlighted):
----------------------------------------------
-foo(buf, size);
+foo(+{obj->buf, obj->}size);
----------------------------------------------
whereas a more semantically meaningful output would be:
----------------------------------------------
-foo(buf, size);
+foo(+{obj->}buf, +{obj->}size);
----------------------------------------------
Note that doing this right would probably involve a set of
content-specific boundary patterns, similar to word-diff. Otherwise
you get junk like:
-----------------------------------------------------
-this line has some -{i}nt-{ere}sti-{ng} text on it
+this line has some +{fa}nt+{a}sti+{c} text on it
-----------------------------------------------------
which is less readable than the current output.
2. The multi-line matching assumes that lines in the pre- and post-image
match by position. This is often the case, but can be fooled when a
line is removed from the top and a new one added at the bottom (or
vice versa). Unless the lines in the middle are also changed, diffs
will show this as two hunks, and it will not get highlighted at all
(which is good). But if the lines in the middle are changed, the
highlighting can be misleading. Here's a pathological case:
-----------------------------------------------------
-one
-two
-three
-four
+two 2
+three 3
+four 4
+five 5
-----------------------------------------------------
which gets highlighted as:
-----------------------------------------------------
-one
-t-{wo}
-three
-f-{our}
+two 2
+t+{hree 3}
+four 4
+f+{ive 5}
-----------------------------------------------------
because it matches "two" to "three 3", and so forth. It would be
nicer as:
-----------------------------------------------------
-one
-two
-three
-four
+two +{2}
+three +{3}
+four +{4}
+five 5
-----------------------------------------------------
which would probably involve pre-matching the lines into pairs
according to some heuristic.

View file

@ -0,0 +1,8 @@
package main;
# Some scripts may not realize that SIGPIPE is being ignored when launching the
# pager--for instance scripts written in Python.
$SIG{PIPE} = 'DEFAULT';
DiffHighlight::highlight_stdin();
exit 0;

View file

@ -0,0 +1,2 @@
/trash directory*
/test-results

View file

@ -0,0 +1,22 @@
-include ../../../config.mak.autogen
-include ../../../config.mak
# copied from ../../t/Makefile
SHELL_PATH ?= $(SHELL)
SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
T = $(wildcard t[0-9][0-9][0-9][0-9]-*.sh)
all: test
test: $(T)
.PHONY: help clean all test $(T)
help:
@echo 'Run "$(MAKE) test" to launch test scripts'
@echo 'Run "$(MAKE) clean" to remove trash folders'
$(T):
@echo "*** $@ ***"; '$(SHELL_PATH_SQ)' $@ $(GIT_TEST_OPTS)
clean:
$(RM) -r 'trash directory'.*

View file

@ -0,0 +1,341 @@
#!/bin/sh
test_description='Test diff-highlight'
CURR_DIR=$(pwd)
TEST_OUTPUT_DIRECTORY=$(pwd)
TEST_DIRECTORY="$CURR_DIR"/../../../t
DIFF_HIGHLIGHT="$CURR_DIR"/../diff-highlight
CW="$(printf "\033[7m")" # white
CR="$(printf "\033[27m")" # reset
. "$TEST_DIRECTORY"/test-lib.sh
if ! test_have_prereq PERL
then
skip_all='skipping diff-highlight tests; perl not available'
test_done
fi
# dh_test is a test helper function which takes 3 file names as parameters. The
# first 2 files are used to generate diff and commit output, which is then
# piped through diff-highlight. The 3rd file should contain the expected output
# of diff-highlight (minus the diff/commit header, ie. everything after and
# including the first @@ line).
dh_test () {
a="$1" b="$2" &&
cat >patch.exp &&
{
cat "$a" >file &&
git add file &&
git commit -m "Add a file" &&
cat "$b" >file &&
git diff file >diff.raw &&
git commit -a -m "Update a file" &&
git show >commit.raw
} >/dev/null &&
"$DIFF_HIGHLIGHT" <diff.raw | test_strip_patch_header >diff.act &&
"$DIFF_HIGHLIGHT" <commit.raw | test_strip_patch_header >commit.act &&
test_cmp patch.exp diff.act &&
test_cmp patch.exp commit.act
}
test_strip_patch_header () {
sed -n '/^@@/,$p' $*
}
# dh_test_setup_history generates a contrived graph such that we have at least
# 1 nesting (E) and 2 nestings (F).
#
# A---B master
# /
# D---E---F branch
#
# git log --all --graph
# * commit
# | B
# | * commit
# | | F
# * | commit
# | | A
# | * commit
# |/
# | E
# * commit
# D
#
dh_test_setup_history () {
echo file1 >file &&
git add file &&
test_tick &&
git commit -m "D" &&
git checkout -b branch &&
echo file2 >file &&
test_tick &&
git commit -a -m "E" &&
git checkout master &&
echo file2 >file &&
test_tick &&
git commit -a -m "A" &&
git checkout branch &&
echo file3 >file &&
test_tick &&
git commit -a -m "F" &&
git checkout master &&
echo file3 >file &&
test_tick &&
git commit -a -m "B"
}
left_trim () {
"$PERL_PATH" -pe 's/^\s+//'
}
trim_graph () {
# graphs start with * or |
# followed by a space or / or \
"$PERL_PATH" -pe 's@^((\*|\|)( |/|\\))+@@'
}
test_expect_success 'diff-highlight highlights the beginning of a line' '
cat >a <<-\EOF &&
aaa
bbb
ccc
EOF
cat >b <<-\EOF &&
aaa
0bb
ccc
EOF
dh_test a b <<-EOF
@@ -1,3 +1,3 @@
aaa
-${CW}b${CR}bb
+${CW}0${CR}bb
ccc
EOF
'
test_expect_success 'diff-highlight highlights the end of a line' '
cat >a <<-\EOF &&
aaa
bbb
ccc
EOF
cat >b <<-\EOF &&
aaa
bb0
ccc
EOF
dh_test a b <<-EOF
@@ -1,3 +1,3 @@
aaa
-bb${CW}b${CR}
+bb${CW}0${CR}
ccc
EOF
'
test_expect_success 'diff-highlight highlights the middle of a line' '
cat >a <<-\EOF &&
aaa
bbb
ccc
EOF
cat >b <<-\EOF &&
aaa
b0b
ccc
EOF
dh_test a b <<-EOF
@@ -1,3 +1,3 @@
aaa
-b${CW}b${CR}b
+b${CW}0${CR}b
ccc
EOF
'
test_expect_success 'diff-highlight does not highlight whole line' '
cat >a <<-\EOF &&
aaa
bbb
ccc
EOF
cat >b <<-\EOF &&
aaa
000
ccc
EOF
dh_test a b <<-EOF
@@ -1,3 +1,3 @@
aaa
-bbb
+000
ccc
EOF
'
test_expect_failure 'diff-highlight highlights mismatched hunk size' '
cat >a <<-\EOF &&
aaa
bbb
EOF
cat >b <<-\EOF &&
aaa
b0b
ccc
EOF
dh_test a b <<-EOF
@@ -1,3 +1,3 @@
aaa
-b${CW}b${CR}b
+b${CW}0${CR}b
+ccc
EOF
'
# These two code points share the same leading byte in UTF-8 representation;
# a naive byte-wise diff would highlight only the second byte.
#
# - U+00f3 ("o" with acute)
o_accent=$(printf '\303\263')
# - U+00f8 ("o" with stroke)
o_stroke=$(printf '\303\270')
test_expect_success 'diff-highlight treats multibyte utf-8 as a unit' '
echo "unic${o_accent}de" >a &&
echo "unic${o_stroke}de" >b &&
dh_test a b <<-EOF
@@ -1 +1 @@
-unic${CW}${o_accent}${CR}de
+unic${CW}${o_stroke}${CR}de
EOF
'
# Unlike the UTF-8 above, these are combining code points which are meant
# to modify the character preceding them:
#
# - U+0301 (combining acute accent)
combine_accent=$(printf '\314\201')
# - U+0302 (combining circumflex)
combine_circum=$(printf '\314\202')
test_expect_failure 'diff-highlight treats combining code points as a unit' '
echo "unico${combine_accent}de" >a &&
echo "unico${combine_circum}de" >b &&
dh_test a b <<-EOF
@@ -1 +1 @@
-unic${CW}o${combine_accent}${CR}de
+unic${CW}o${combine_circum}${CR}de
EOF
'
test_expect_success 'diff-highlight works with the --graph option' '
dh_test_setup_history &&
# date-order so that the commits are interleaved for both
# trim graph elements so we can do a diff
# trim leading space because our trim_graph is not perfect
git log --branches -p --date-order |
"$DIFF_HIGHLIGHT" | left_trim >graph.exp &&
git log --branches -p --date-order --graph |
"$DIFF_HIGHLIGHT" | trim_graph | left_trim >graph.act &&
test_cmp graph.exp graph.act
'
# Just reuse the previous graph test, but with --color. Our trimming
# doesn't know about color, so just sanity check that something got
# highlighted.
test_expect_success 'diff-highlight works with color graph' '
git log --branches -p --date-order --graph --color |
"$DIFF_HIGHLIGHT" | trim_graph | left_trim >graph &&
grep "\[7m" graph
'
# Most combined diffs won't meet diff-highlight's line-number filter. So we
# create one here where one side drops a line and the other modifies it. That
# should result in a diff like:
#
# - modified content
# ++resolved content
#
# which naively looks like one side added "+resolved".
test_expect_success 'diff-highlight ignores combined diffs' '
echo "content" >file &&
git add file &&
git commit -m base &&
>file &&
git commit -am master &&
git checkout -b other HEAD^ &&
echo "modified content" >file &&
git commit -am other &&
test_must_fail git merge master &&
echo "resolved content" >file &&
git commit -am resolved &&
cat >expect <<-\EOF &&
--- a/file
+++ b/file
@@@ -1,1 -1,0 +1,1 @@@
- modified content
++resolved content
EOF
git show -c | "$DIFF_HIGHLIGHT" >actual.raw &&
sed -n "/^---/,\$p" <actual.raw >actual &&
test_cmp expect actual
'
test_expect_success 'diff-highlight handles --graph with leading dash' '
cat >file <<-\EOF &&
before
the old line
-leading dash
EOF
git add file &&
git commit -m before &&
sed s/old/new/ <file >file.tmp &&
mv file.tmp file &&
git add file &&
git commit -m after &&
cat >expect <<-EOF &&
--- a/file
+++ b/file
@@ -1,3 +1,3 @@
before
-the ${CW}old${CR} line
+the ${CW}new${CR} line
-leading dash
EOF
git log --graph -p -1 | "$DIFF_HIGHLIGHT" >actual.raw &&
trim_graph <actual.raw | sed -n "/^---/,\$p" >actual &&
test_cmp expect actual
'
test_done