git grep bug with --column and --only-matching
5 messages between Apr 21, 2026 and May 4, 2026, from Brandon Chinn, René Scharfe, Phillip Wood.
Plain Markdown or JSON for tools and agents.
I'm encountering a bug when using `git grep` with both `--column` and `--only-matching`, is this a known limitation?
Repro:
``` $ echo 'x x x' > repro.txt
$ grep -bo x repro.txt 0:x 4:x 8:x
$ git grep --no-index -o -n --column x repro.txt repro.txt:1: 1:x repro.txt:1: 2:x repro.txt:1: 6:x ```
[System Info]
git version:
git version 2.51.1
cpu: arm64
no commit associated with this build
sizeof-long: 8
sizeof-size_t: 8
shell-path: /bin/sh
feature: fsmonitor--daemon
libcurl: 8.7.1
zlib: 1.2.12
SHA-1: SHA1_DC
SHA-256: SHA256_BLK
default-ref-format: files
default-hash: sha1
uname: Darwin 25.3.0 Darwin Kernel Version 25.3.0: Wed Jan 28 20:56:34
PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T8112 arm64
compiler info: clang: 17.0.0 (clang-1700.0.13.3)
libc info: no libc information available
$SHELL (typically, interactive shell): /bin/zsh
Re: git grep bug with --column and --only-matching
On 4/21/26 7:03 AM, Brandon Chinn wrote:
Show 17 quoted lines
> I'm encountering a bug when using `git grep` with both `--column` and
> `--only-matching`, is this a known limitation?
>
> Repro:
>
> ```
> $ echo 'x x x' > repro.txt
>
> $ grep -bo x repro.txt
> 0:x
> 4:x
> 8:x
>
> $ git grep --no-index -o -n --column x repro.txt
> repro.txt:1: 1:x
> repro.txt:1: 2:x
> repro.txt:1: 6:x
The first column value matches (1-based for git grep, 0-based for grep(1)), the remaining ones are different. grep(1) shows the offset like for the first match, but git grep adds the relative position of the end of the previous match.
I don't know how to use the resulting numbers and I agree that it seems like a bug -- showing the column of each match within its line makes more sense for an option named --column.
However, the original submission of this feature in https://lore.kernel.org/git/cover.1529961706.git.me@ttaylorr.com/ gave similar examples in its commit message and its tests, called them "as one would expect" and was not challenged on that, so I may be missing something.
Here's its first example with underlined matches, a ruler and the intended output (without unnecessary headers):
(`man gitcvs-migration` or `git help cvs-migration` if git is
^^^ ^^^ ^^^
1111111111222222222233333333334444444444555555555566
123456789 123456789 123456789 123456789 123456789 123456789 1
7:git
16:git
38:gitAnd here's the last line of the test file from t7810 with underlined matches (the other four lines are very similar), a ruler and the expected output (unncessary headers removed):
foo_mmap bar mmap baz
^^^^ ^^^^
111111111122
123456789 123456789 1 5:mmap
13:mmap
If we wanted to show the column of matches 2 and beyond then we could do something like this:
diff --git a/grep.c b/grep.c
index c7e1dc1e0e..a54e5d86a9 100644
--- a/grep.c
+++ b/grep.c
@@ -1267,6 +1267,7 @@ static void show_line(struct grep_opt *opt,
regmatch_t match;
enum grep_context ctx = GREP_CONTEXT_BODY;
int eflags = 0;
+ const char *start = bol;
if (want_color(opt->color)) {
if (sign == ':')
@@ -1285,6 +1286,7 @@ static void show_line(struct grep_opt *opt,
if (match.rm_so == match.rm_eo)
break;
+ cno = bol - start + match.rm_so + 1;
if (opt->only_matching)
show_line_header(opt, name, lno, cno, sign);
else
@@ -1294,7 +1296,6 @@ static void show_line(struct grep_opt *opt,
if (opt->only_matching)
opt->output(opt, "\n", 1);
bol += match.rm_eo;
- cno += match.rm_eo;
rest -= match.rm_eo;
eflags = REG_NOTBOL;
}
diff --git a/t/t7810-grep.sh b/t/t7810-grep.sh
index 64ac4f04ee..bd439563d6 100755
--- a/t/t7810-grep.sh
+++ b/t/t7810-grep.sh
@@ -322,11 +322,11 @@ do
${HC}file:1:5:mmap
${HC}file:2:5:mmap
${HC}file:3:5:mmap
- ${HC}file:3:13:mmap
+ ${HC}file:3:14:mmap
${HC}file:4:5:mmap
- ${HC}file:4:13:mmap
+ ${HC}file:4:14:mmap
${HC}file:5:5:mmap
- ${HC}file:5:13:mmap
+ ${HC}file:5:14:mmap
EOF
git grep --column -n -o -e mmap $H >actual &&
test_cmp expected actual
Re: git grep bug with --column and --only-matching
On 21/04/2026 21:33, René Scharfe wrote:
Show 23 quoted lines
> On 4/21/26 7:03 AM, Brandon Chinn wrote:
>> I'm encountering a bug when using `git grep` with both `--column` and
>> `--only-matching`, is this a known limitation?
>>
>> Repro:
>>
>> ```
>> $ echo 'x x x' > repro.txt
>>
>> $ grep -bo x repro.txt
>> 0:x
>> 4:x
>> 8:x
>>
>> $ git grep --no-index -o -n --column x repro.txt
>> repro.txt:1: 1:x
>> repro.txt:1: 2:x
>> repro.txt:1: 6:x
>
> The first column value matches (1-based for git grep, 0-based for
> grep(1)), the remaining ones are different. grep(1) shows the offset
> like for the first match, but git grep adds the relative position of the
> end of the previous match.
I'm having a hard time understanding what git is doing. Looking at your example with the ruler
Show 8 quoted lines
> (`man gitcvs-migration` or `git help cvs-migration` if git is
> ^^^ ^^^ ^^^
> 1111111111222222222233333333334444444444555555555566
> 123456789 123456789 123456789 123456789 123456789 123456789 1
>
> 7:git
> 16:git
> 38:git
The first match ends at column 9 and the second match starts at column 29 so how do we end up with 16 as the "column" of the second match when the difference between them is 20?
> I don't know how to use the resulting numbers and I agree that it seems
> like a bug -- showing the column of each match within its line makes
> more sense for an option named --column.
Show 5 quoted lines
> However, the original submission of this feature in
> https://lore.kernel.org/git/cover.1529961706.git.me@ttaylorr.com/
> gave similar examples in its commit message and its tests, called
> them "as one would expect" and was not challenged on that, so I may be
> missing something.
I wonder how hard anyone looked at the examples and tests, the discussion seemed to have focused on other things.
Thanks
Phillip
Show 77 quoted lines
> Here's its first example with underlined matches, a ruler and the
> intended output (without unnecessary headers):
>
> (`man gitcvs-migration` or `git help cvs-migration` if git is
> ^^^ ^^^ ^^^
> 1111111111222222222233333333334444444444555555555566
> 123456789 123456789 123456789 123456789 123456789 123456789 1
>
> 7:git
> 16:git
> 38:git
>
> And here's the last line of the test file from t7810 with underlined
> matches (the other four lines are very similar), a ruler and the
> expected output (unncessary headers removed):
>
> foo_mmap bar mmap baz
> ^^^^ ^^^^
> 111111111122
> 123456789 123456789 1
>
> 5:mmap
> 13:mmap
>
> If we wanted to show the column of matches 2 and beyond then we could do
> something like this:
>
>
> diff --git a/grep.c b/grep.c
> index c7e1dc1e0e..a54e5d86a9 100644
> --- a/grep.c
> +++ b/grep.c
> @@ -1267,6 +1267,7 @@ static void show_line(struct grep_opt *opt,
> regmatch_t match;
> enum grep_context ctx = GREP_CONTEXT_BODY;
> int eflags = 0;
> + const char *start = bol;
>
> if (want_color(opt->color)) {
> if (sign == ':')
> @@ -1285,6 +1286,7 @@ static void show_line(struct grep_opt *opt,
> if (match.rm_so == match.rm_eo)
> break;
>
> + cno = bol - start + match.rm_so + 1;
> if (opt->only_matching)
> show_line_header(opt, name, lno, cno, sign);
> else
> @@ -1294,7 +1296,6 @@ static void show_line(struct grep_opt *opt,
> if (opt->only_matching)
> opt->output(opt, "\n", 1);
> bol += match.rm_eo;
> - cno += match.rm_eo;
> rest -= match.rm_eo;
> eflags = REG_NOTBOL;
> }
> diff --git a/t/t7810-grep.sh b/t/t7810-grep.sh
> index 64ac4f04ee..bd439563d6 100755
> --- a/t/t7810-grep.sh
> +++ b/t/t7810-grep.sh
> @@ -322,11 +322,11 @@ do
> ${HC}file:1:5:mmap
> ${HC}file:2:5:mmap
> ${HC}file:3:5:mmap
> - ${HC}file:3:13:mmap
> + ${HC}file:3:14:mmap
> ${HC}file:4:5:mmap
> - ${HC}file:4:13:mmap
> + ${HC}file:4:14:mmap
> ${HC}file:5:5:mmap
> - ${HC}file:5:13:mmap
> + ${HC}file:5:14:mmap
> EOF
> git grep --column -n -o -e mmap $H >actual &&
> test_cmp expected actual
>
> [PATCH] grep: fix --column --only-match for 2nd and later matches
"git grep --column --only-match" shows the 1-based column number of the first match on each line, but confusing numbers for further matches. Example:
$ echo 123456789012345678901234567890 >file
$ for d in 1 2 3 4 5 6 7 8 9 0
do
git grep --no-index --column --only-matching $d file |
awk -v FS=: -v l=$d: '{l = l sprintf("%3s", $2)} END {print l}'
done
1: 1 2 12
2: 2 4 14
3: 3 6 16
4: 4 8 18
5: 5 10 20
6: 6 12 22
7: 7 14 24
8: 8 16 26
9: 9 18 28
0: 10 20 30Report the column number of each match instead:
$ for d in 1 2 3 4 5 6 7 8 9 0
do
./git grep --no-index --column --only-matching $d file |
awk -v FS=: -v l=$d: '{l = l sprintf("%3s", $2)} END {print l}'
done
1: 1 11 21
2: 2 12 22
3: 3 13 23
4: 4 14 24
5: 5 15 25
6: 6 16 26
7: 7 17 27
8: 8 18 28
9: 9 19 29
0: 10 20 30We need to adjust the test in t7810 as well. The file it uses has the following five lines; I add a line highlighting the matches and a ruler at the bottom here, to make it easier to see that the second "mmap" indeed starts at column 14:
foo mmap bar
foo_mmap bar
foo_mmap bar mmap
foo mmap bar_mmap
foo_mmap bar mmap baz
==== ====
123456789 123456789 1Reported-by: Brandon Chinn <brandonchinn178@gmail.com>
Signed-off-by: René Scharfe <l.s.r@web.de>
---
grep.c | 3 ++-
t/t7810-grep.sh | 6 +++---
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/grep.c b/grep.c
index c7e1dc1e0e..a54e5d86a9 100644
--- a/grep.c
+++ b/grep.c
@@ -1267,6 +1267,7 @@ static void show_line(struct grep_opt *opt,
regmatch_t match;
enum grep_context ctx = GREP_CONTEXT_BODY;
int eflags = 0;
+ const char *start = bol;
if (want_color(opt->color)) {
if (sign == ':')
@@ -1285,6 +1286,7 @@ static void show_line(struct grep_opt *opt,
if (match.rm_so == match.rm_eo)
break;
+ cno = bol - start + match.rm_so + 1;
if (opt->only_matching)
show_line_header(opt, name, lno, cno, sign);
else
@@ -1294,7 +1296,6 @@ static void show_line(struct grep_opt *opt,
if (opt->only_matching)
opt->output(opt, "\n", 1);
bol += match.rm_eo;
- cno += match.rm_eo;
rest -= match.rm_eo;
eflags = REG_NOTBOL;
}
diff --git a/t/t7810-grep.sh b/t/t7810-grep.sh
index 64ac4f04ee..bd439563d6 100755
--- a/t/t7810-grep.sh
+++ b/t/t7810-grep.sh
@@ -322,11 +322,11 @@ do
${HC}file:1:5:mmap
${HC}file:2:5:mmap
${HC}file:3:5:mmap
- ${HC}file:3:13:mmap
+ ${HC}file:3:14:mmap
${HC}file:4:5:mmap
- ${HC}file:4:13:mmap
+ ${HC}file:4:14:mmap
${HC}file:5:5:mmap
- ${HC}file:5:13:mmap
+ ${HC}file:5:14:mmap
EOF
git grep --column -n -o -e mmap $H >actual &&
test_cmp expected actual
--
2.54.0
Re: [PATCH] grep: fix --column --only-match for 2nd and later matches
Hi René
On 24/04/2026 22:04, René Scharfe wrote:
Show 9 quoted lines
> diff --git a/grep.c b/grep.c
> index c7e1dc1e0e..a54e5d86a9 100644
> --- a/grep.c
> +++ b/grep.c
> @@ -1267,6 +1267,7 @@ static void show_line(struct grep_opt *opt,
> regmatch_t match;
> enum grep_context ctx = GREP_CONTEXT_BODY;
> int eflags = 0;
> + const char *start = bol;
Here we save a pointer to the start of the line
Show 7 quoted lines
> if (want_color(opt->color)) {
> if (sign == ':')
> @@ -1285,6 +1286,7 @@ static void show_line(struct grep_opt *opt,
> if (match.rm_so == match.rm_eo)
> break;
>
> + cno = bol - start + match.rm_so + 1;and then we calculate the column number relative to that.
That looks good, thanks for fixing it
Phillip
Show 30 quoted lines
> if (opt->only_matching)
> show_line_header(opt, name, lno, cno, sign);
> else
> @@ -1294,7 +1296,6 @@ static void show_line(struct grep_opt *opt,
> if (opt->only_matching)
> opt->output(opt, "\n", 1);
> bol += match.rm_eo;
> - cno += match.rm_eo;
> rest -= match.rm_eo;
> eflags = REG_NOTBOL;
> }
> diff --git a/t/t7810-grep.sh b/t/t7810-grep.sh
> index 64ac4f04ee..bd439563d6 100755
> --- a/t/t7810-grep.sh
> +++ b/t/t7810-grep.sh
> @@ -322,11 +322,11 @@ do
> ${HC}file:1:5:mmap
> ${HC}file:2:5:mmap
> ${HC}file:3:5:mmap
> - ${HC}file:3:13:mmap
> + ${HC}file:3:14:mmap
> ${HC}file:4:5:mmap
> - ${HC}file:4:13:mmap
> + ${HC}file:4:14:mmap
> ${HC}file:5:5:mmap
> - ${HC}file:5:13:mmap
> + ${HC}file:5:14:mmap
> EOF
> git grep --column -n -o -e mmap $H >actual &&
> test_cmp expected actual