New --dirstat=lines mode, doing dirstat analysis based on diffstat
This patch adds an alternative implementation of show_dirstat(), called show_dirstat_by_line(), which uses the more expensive diffstat analysis (as opposed to show_dirstat()'s own (relatively inexpensive) analysis) to derive the numbers from which the --dirstat output is computed. The alternative implementation is controlled by the new "lines" parameter to the --dirstat option (or the diff.dirstat config variable). For binary files, the diffstat analysis counts bytes instead of lines, so to prevent binary files from dominating the dirstat results, the byte counts for binary files are divided by 64 before being compared to their textual/line-based counterparts. This is a stupid and ugly - but very cheap - heuristic. In linux-2.6.git, running the three different --dirstat modes: time git diff v2.6.20..v2.6.30 --dirstat=changes > /dev/null vs. time git diff v2.6.20..v2.6.30 --dirstat=lines > /dev/null vs. time git diff v2.6.20..v2.6.30 --dirstat=files > /dev/null yields the following average runtimes on my machine: - "changes" (default): ~6.0 s - "lines": ~9.6 s - "files": ~0.1 s So, as expected, there's a considerable performance hit (~60%) by going through the full diffstat analysis as compared to the default "changes" analysis (obviously, "files" is much faster than both). As such, the "lines" mode is probably only useful if you really need the --dirstat numbers to be consistent with the numbers returned from the other --*stat options. The patch also includes documentation and tests for the new dirstat mode. Improved-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Johan Herland <johan@herland.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Johan Herland committed
Apr 29, 2011 at 11:36 UTC
1c57a627bf269f3c83c48ad724cd8b14292502ef
5 files changed
+176
-2
Documentation/diff-config.txt
+8
@@ -23,6 +23,14 @@ diff.dirstat::
23
the amount of pure code movements within a file. In other words,
24
rearranging lines in a file is not counted as much as other changes.
25
This is the default behavior when no parameter is given.
26
+`lines`;;
27
+ Compute the dirstat numbers by doing the regular line-based diff
28
+ analysis, and summing the removed/added line counts. (For binary
29
+ files, count 64-byte chunks instead, since binary files have no
30
+ natural concept of lines). This is a more expensive `--dirstat`
31
+ behavior than the `changes` behavior, but it does count rearranged
32
+ lines within a file as much as other changes. The resulting output
33
+ is consistent with what you get from the other `--*stat` options.
34
`files`;;
35
Compute the dirstat numbers by counting the number of files changed.
36
Each changed file counts equally in the dirstat analysis. This is
Documentation/diff-options.txt
+8
@@ -81,6 +81,14 @@ endif::git-format-patch[]
81
the amount of pure code movements within a file. In other words,
82
rearranging lines in a file is not counted as much as other changes.
83
This is the default behavior when no parameter is given.
84
+`lines`;;
85
+ Compute the dirstat numbers by doing the regular line-based diff
86
+ analysis, and summing the removed/added line counts. (For binary
87
+ files, count 64-byte chunks instead, since binary files have no
88
+ natural concept of lines). This is a more expensive `--dirstat`
89
+ behavior than the `changes` behavior, but it does count rearranged
90
+ lines within a file as much as other changes. The resulting output
91
+ is consistent with what you get from the other `--*stat` options.
92
`files`;;
93
Compute the dirstat numbers by counting the number of files changed.
94
Each changed file counts equally in the dirstat analysis. This is
diff.c
+59
-2
@@ -73,9 +73,15 @@ static int parse_dirstat_params(struct diff_options *options, const char *params
73
while (*p) {
74
if (!prefixcmp(p, "changes")) {
75
p += 7;
76
+ DIFF_OPT_CLR(options, DIRSTAT_BY_LINE);
77
+ DIFF_OPT_CLR(options, DIRSTAT_BY_FILE);
78
+ } else if (!prefixcmp(p, "lines")) {
79
+ p += 5;
80
+ DIFF_OPT_SET(options, DIRSTAT_BY_LINE);
81
DIFF_OPT_CLR(options, DIRSTAT_BY_FILE);
82
} else if (!prefixcmp(p, "files")) {
83
p += 5;
84
+ DIFF_OPT_CLR(options, DIRSTAT_BY_LINE);
85
DIFF_OPT_SET(options, DIRSTAT_BY_FILE);
86
} else if (!prefixcmp(p, "noncumulative")) {
87
p += 13;
@@ -1671,6 +1677,50 @@ found_damage:
1677
gather_dirstat(options, &dir, changed, "", 0);
1678
}
1679
1680
+static void show_dirstat_by_line(struct diffstat_t *data, struct diff_options *options)
1681
+{
1682
+ int i;
1683
+ unsigned long changed;
1684
+ struct dirstat_dir dir;
1685
+
1686
+ if (data->nr == 0)
1687
+ return;
1688
+
1689
+ dir.files = NULL;
1690
+ dir.alloc = 0;
1691
+ dir.nr = 0;
1692
+ dir.permille = options->dirstat_permille;
1693
+ dir.cumulative = DIFF_OPT_TST(options, DIRSTAT_CUMULATIVE);
1694
+
1695
+ changed = 0;
1696
+ for (i = 0; i < data->nr; i++) {
1697
+ struct diffstat_file *file = data->files[i];
1698
+ unsigned long damage = file->added + file->deleted;
1699
+ if (file->is_binary)
1700
+ /*
1701
+ * binary files counts bytes, not lines. Must find some
1702
+ * way to normalize binary bytes vs. textual lines.
1703
+ * The following heuristic assumes that there are 64
1704
+ * bytes per "line".
1705
+ * This is stupid and ugly, but very cheap...
1706
+ */
1707
+ damage = (damage + 63) / 64;
1708
+ ALLOC_GROW(dir.files, dir.nr + 1, dir.alloc);
1709
+ dir.files[dir.nr].name = file->name;
1710
+ dir.files[dir.nr].changed = damage;
1711
+ changed += damage;
1712
+ dir.nr++;
1713
+ }
1714
+
1715
+ /* This can happen even with many files, if everything was renames */
1716
+ if (!changed)
1717
+ return;
1718
+
1719
+ /* Show all directories with more than x% of the changes */
1720
+ qsort(dir.files, dir.nr, sizeof(dir.files[0]), dirstat_compare);
1721
+ gather_dirstat(options, &dir, changed, "", 0);
1722
+}
1723
+
1724
static void free_diffstat_info(struct diffstat_t *diffstat)
1725
{
1726
int i;
@@ -4088,6 +4138,7 @@ void diff_flush(struct diff_options *options)
4138
struct diff_queue_struct *q = &diff_queued_diff;
4139
int i, output_format = options->output_format;
4140
int separator = 0;
4141
+ int dirstat_by_line = 0;
4142
4143
/*
4144
* Order: raw, stat, summary, patch
@@ -4108,7 +4159,11 @@ void diff_flush(struct diff_options *options)
4159
separator++;
4160
}
4161
4111
- if (output_format & (DIFF_FORMAT_DIFFSTAT|DIFF_FORMAT_SHORTSTAT|DIFF_FORMAT_NUMSTAT)) {
4162
+ if (output_format & DIFF_FORMAT_DIRSTAT && DIFF_OPT_TST(options, DIRSTAT_BY_LINE))
4163
+ dirstat_by_line = 1;
4164
+
4165
+ if (output_format & (DIFF_FORMAT_DIFFSTAT|DIFF_FORMAT_SHORTSTAT|DIFF_FORMAT_NUMSTAT) ||
4166
+ dirstat_by_line) {
4167
struct diffstat_t diffstat;
4168
4169
memset(&diffstat, 0, sizeof(struct diffstat_t));
@@ -4123,10 +4178,12 @@ void diff_flush(struct diff_options *options)
4178
show_stats(&diffstat, options);
4179
if (output_format & DIFF_FORMAT_SHORTSTAT)
4180
show_shortstats(&diffstat, options);
4181
+ if (output_format & DIFF_FORMAT_DIRSTAT)
4182
+ show_dirstat_by_line(&diffstat, options);
4183
free_diffstat_info(&diffstat);
4184
separator++;
4185
}
4129
- if (output_format & DIFF_FORMAT_DIRSTAT)
4186
+ if ((output_format & DIFF_FORMAT_DIRSTAT) && !dirstat_by_line)
4187
show_dirstat(options);
4188
4189
if (output_format & DIFF_FORMAT_SUMMARY && !is_summary_empty(q)) {
diff.h
+1
@@ -78,6 +78,7 @@ typedef struct strbuf *(*diff_prefix_fn_t)(struct diff_options *opt, void *data)
78
#define DIFF_OPT_IGNORE_UNTRACKED_IN_SUBMODULES (1 << 25)
79
#define DIFF_OPT_IGNORE_DIRTY_SUBMODULES (1 << 26)
80
#define DIFF_OPT_OVERRIDE_SUBMODULE_CONFIG (1 << 27)
81
+#define DIFF_OPT_DIRSTAT_BY_LINE (1 << 28)
82
83
#define DIFF_OPT_TST(opts, flag) ((opts)->flags & DIFF_OPT_##flag)
84
#define DIFF_OPT_SET(opts, flag) ((opts)->flags |= DIFF_OPT_##flag)
t/t4047-diff-dirstat.sh
+100
@@ -839,4 +839,104 @@ test_expect_success '--dirstat=files,cumulative,27.09' '
839
test_cmp expect_diff_dirstat_CC actual_diff_dirstat_CC
840
'
841
842
+cat <<EOF >expect_diff_dirstat
843
+ 10.6% dst/copy/changed/
844
+ 10.6% dst/copy/rearranged/
845
+ 10.6% dst/copy/unchanged/
846
+ 10.6% dst/move/changed/
847
+ 10.6% dst/move/rearranged/
848
+ 10.6% dst/move/unchanged/
849
+ 10.6% src/move/changed/
850
+ 10.6% src/move/rearranged/
851
+ 10.6% src/move/unchanged/
852
+EOF
853
+
854
+cat <<EOF >expect_diff_dirstat_M
855
+ 5.2% changed/
856
+ 26.3% dst/copy/changed/
857
+ 26.3% dst/copy/rearranged/
858
+ 26.3% dst/copy/unchanged/
859
+ 5.2% dst/move/changed/
860
+ 5.2% dst/move/rearranged/
861
+ 5.2% rearranged/
862
+EOF
863
+
864
+cat <<EOF >expect_diff_dirstat_CC
865
+ 16.6% changed/
866
+ 16.6% dst/copy/changed/
867
+ 16.6% dst/copy/rearranged/
868
+ 16.6% dst/move/changed/
869
+ 16.6% dst/move/rearranged/
870
+ 16.6% rearranged/
871
+EOF
872
+
873
+test_expect_success '--dirstat=lines' '
874
+ git diff --dirstat=lines HEAD^..HEAD >actual_diff_dirstat &&
875
+ test_cmp expect_diff_dirstat actual_diff_dirstat &&
876
+ git diff --dirstat=lines -M HEAD^..HEAD >actual_diff_dirstat_M &&
877
+ test_cmp expect_diff_dirstat_M actual_diff_dirstat_M &&
878
+ git diff --dirstat=lines -C -C HEAD^..HEAD >actual_diff_dirstat_CC &&
879
+ test_cmp expect_diff_dirstat_CC actual_diff_dirstat_CC
880
+'
881
+
882
+test_expect_success 'diff.dirstat=lines' '
883
+ git -c diff.dirstat=lines diff --dirstat HEAD^..HEAD >actual_diff_dirstat &&
884
+ test_cmp expect_diff_dirstat actual_diff_dirstat &&
885
+ git -c diff.dirstat=lines diff --dirstat -M HEAD^..HEAD >actual_diff_dirstat_M &&
886
+ test_cmp expect_diff_dirstat_M actual_diff_dirstat_M &&
887
+ git -c diff.dirstat=lines diff --dirstat -C -C HEAD^..HEAD >actual_diff_dirstat_CC &&
888
+ test_cmp expect_diff_dirstat_CC actual_diff_dirstat_CC
889
+'
890
+
891
+cat <<EOF >expect_diff_dirstat
892
+ 2.1% changed/
893
+ 10.6% dst/copy/changed/
894
+ 10.6% dst/copy/rearranged/
895
+ 10.6% dst/copy/unchanged/
896
+ 10.6% dst/move/changed/
897
+ 10.6% dst/move/rearranged/
898
+ 10.6% dst/move/unchanged/
899
+ 2.1% rearranged/
900
+ 10.6% src/move/changed/
901
+ 10.6% src/move/rearranged/
902
+ 10.6% src/move/unchanged/
903
+EOF
904
+
905
+cat <<EOF >expect_diff_dirstat_M
906
+ 5.2% changed/
907
+ 26.3% dst/copy/changed/
908
+ 26.3% dst/copy/rearranged/
909
+ 26.3% dst/copy/unchanged/
910
+ 5.2% dst/move/changed/
911
+ 5.2% dst/move/rearranged/
912
+ 5.2% rearranged/
913
+EOF
914
+
915
+cat <<EOF >expect_diff_dirstat_CC
916
+ 16.6% changed/
917
+ 16.6% dst/copy/changed/
918
+ 16.6% dst/copy/rearranged/
919
+ 16.6% dst/move/changed/
920
+ 16.6% dst/move/rearranged/
921
+ 16.6% rearranged/
922
+EOF
923
+
924
+test_expect_success '--dirstat=lines,0' '
925
+ git diff --dirstat=lines,0 HEAD^..HEAD >actual_diff_dirstat &&
926
+ test_cmp expect_diff_dirstat actual_diff_dirstat &&
927
+ git diff --dirstat=lines,0 -M HEAD^..HEAD >actual_diff_dirstat_M &&
928
+ test_cmp expect_diff_dirstat_M actual_diff_dirstat_M &&
929
+ git diff --dirstat=lines,0 -C -C HEAD^..HEAD >actual_diff_dirstat_CC &&
930
+ test_cmp expect_diff_dirstat_CC actual_diff_dirstat_CC
931
+'
932
+
933
+test_expect_success 'diff.dirstat=0,lines' '
934
+ git -c diff.dirstat=0,lines diff --dirstat HEAD^..HEAD >actual_diff_dirstat &&
935
+ test_cmp expect_diff_dirstat actual_diff_dirstat &&
936
+ git -c diff.dirstat=0,lines diff --dirstat -M HEAD^..HEAD >actual_diff_dirstat_M &&
937
+ test_cmp expect_diff_dirstat_M actual_diff_dirstat_M &&
938
+ git -c diff.dirstat=0,lines diff --dirstat -C -C HEAD^..HEAD >actual_diff_dirstat_CC &&
939
+ test_cmp expect_diff_dirstat_CC actual_diff_dirstat_CC
940
+'
941
+
942
test_done