# [PATCH 1/2] Make git blame's date output format configurable, like git log

6 messages from 2009-02-20 to 2009-02-24. Participants: eletuchy@gmail.com, Junio C Hamano, Jeff King, Eugene Letuchy.
Thread: https://gitlist.dev/t/17926

## eletuchy@gmail.com, 2009-02-20 22:51

Subject: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <1235170271-10694-1-git-send-email-eletuchy@gmail.com>
URL: https://gitlist.dev/e/1235170271-10694-1-git-send-email-eletuchy%40gmail.com

```
From: Eugene Letuchy <eugene@facebook.com>

Add the following:
 - git config value blame.date that expects one of the git log date
   formats ({relative,local,default,iso,rfc,short})
 - git blame command line option --date expects one of the git
   log date formats ({relative,local,default,iso,rfc,short})
 - documentation in blame-options.txt
 - git blame uses the appropriate date.c functions and enums to
   make sense of the date format and provide appropriate data
 - git blame continues to line up the output columns (by padding the
   date column up to the max width of the chosen date format)
 - the date format for git blame without both blame.date and --date
   continues to be ISO for backwards compatibility
 - git annotate ignores the date format specifiers and continues to
   uses the ISO format, as before

Signed-off-by: Eugene Letuchy <eugene@facebook.com>
---
 Documentation/blame-options.txt |    8 +++++
 builtin-blame.c                 |   62 +++++++++++++++++++++++++++++----------
 2 files changed, 54 insertions(+), 16 deletions(-)

diff --git a/Documentation/blame-options.txt b/Documentation/blame-options.txt
index 1ab1b96..ad00d36 100644
--- a/Documentation/blame-options.txt
+++ b/Documentation/blame-options.txt
@@ -63,6 +63,14 @@ of lines before or after the line given by <start>.
 	tree copy has the contents of the named file (specify
 	`-` to make the command read from the standard input).
 
+--date <format>::
+	The value is one of the following alternatives:
+	{relative,local,default,iso,rfc,short}. If --date is not
+	provided, the value of the blame.date config variable is
+	used. If the blame.date config variable is also not set, the
+	iso format is used. For more information, See the discussion
+	of the --date option at linkgit:git-log[1].
+
 -M|<num>|::
 	Detect moving lines in the file as well.  When a commit
 	moves a block of lines in a file (e.g. the original file
diff --git a/builtin-blame.c b/builtin-blame.c
index 114a214..aa5c66c 100644
--- a/builtin-blame.c
+++ b/builtin-blame.c
@@ -1,5 +1,5 @@
 /*
- * Pickaxe
+ * Blame
  *
  * Copyright (c) 2006, Junio C Hamano
  */
@@ -40,6 +40,10 @@ static int reverse;
 static int blank_boundary;
 static int incremental;
 static int xdl_opts = XDF_NEED_MINIMAL;
+
+static enum date_mode blame_date_mode = DATE_ISO8601;
+static size_t blame_date_width;
+
 static struct string_list mailmap;
 
 #ifndef DEBUG
@@ -1507,24 +1511,20 @@ static const char *format_time(unsigned long time, const char *tz_str,
 			       int show_raw_time)
 {
 	static char time_buf[128];
-	time_t t = time;
-	int minutes, tz;
-	struct tm *tm;
+	const char *time_str;
+	int time_len;
+	int tz;
 
 	if (show_raw_time) {
 		sprintf(time_buf, "%lu %s", time, tz_str);
-		return time_buf;
 	}
-
-	tz = atoi(tz_str);
-	minutes = tz < 0 ? -tz : tz;
-	minutes = (minutes / 100)*60 + (minutes % 100);
-	minutes = tz < 0 ? -minutes : minutes;
-	t = time + minutes * 60;
-	tm = gmtime(&t);
-
-	strftime(time_buf, sizeof(time_buf), "%Y-%m-%d %H:%M:%S ", tm);
-	strcat(time_buf, tz_str);
+	else {
+		tz = atoi(tz_str);
+		time_str = show_date(time, tz, blame_date_mode);
+		time_len = strlen(time_str);
+		memcpy(time_buf, time_str, time_len);
+		memset(time_buf + time_len, ' ', blame_date_width - time_len);
+	}
 	return time_buf;
 }
 
@@ -1975,6 +1975,9 @@ static int git_blame_config(const char *var, const char *value, void *cb)
 		blank_boundary = git_config_bool(var, value);
 		return 0;
 	}
+	if (!strcmp(var, "blame.date") && value[0]) {
+		blame_date_mode = parse_date_format(value);
+	}
 	return git_default_config(var, value, cb);
 }
 
@@ -2239,6 +2242,8 @@ int cmd_blame(int argc, const char **argv, const char *prefix)
 
 	git_config(git_blame_config, NULL);
 	init_revisions(&revs, NULL);
+	revs.date_mode = blame_date_mode;
+
 	save_commit_buffer = 0;
 	dashdash_pos = 0;
 
@@ -2263,8 +2268,33 @@ int cmd_blame(int argc, const char **argv, const char *prefix)
 parse_done:
 	argc = parse_options_end(&ctx);
 
-	if (cmd_is_annotate)
+	if (cmd_is_annotate) {
 		output_option |= OUTPUT_ANNOTATE_COMPAT;
+		blame_date_mode = DATE_ISO8601;
+	} else {
+		blame_date_mode = revs.date_mode;
+	}
+
+	switch (blame_date_mode) {
+	case DATE_RFC2822:
+		blame_date_width = sizeof("Thu, 19 Oct 2006 16:00:04 -0700");
+		break;
+	case DATE_ISO8601:
+		blame_date_width = sizeof("2006-10-19 16:00:04 -0700");
+		break;
+	case DATE_SHORT:
+		blame_date_width = sizeof("2006-10-19");
+		break;
+	case DATE_RELATIVE:
+		/* unfortunately "normal" is the fallback for "relative" */
+		/* blame_date_width = sizeof("14 minutes ago"); */
+		/* break; */
+	case DATE_LOCAL:
+	case DATE_NORMAL:
+		blame_date_width = sizeof("Thu Oct 19 16:00:04 2006 -0700");
+		break;
+	}
+	blame_date_width -= 1; /* strip the null */
 
 	if (DIFF_OPT_TST(&revs.diffopt, FIND_COPIES_HARDER))
 		opt |= (PICKAXE_BLAME_COPY | PICKAXE_BLAME_MOVE |
-- 
1.6.2.rc1.14.g07c3.dirty

```

## Junio C Hamano, 2009-02-22 17:23

Subject: Re: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <7vwsbiz8pg.fsf@gitster.siamese.dyndns.org>
URL: https://gitlist.dev/e/7vwsbiz8pg.fsf%40gitster.siamese.dyndns.org
In-Reply-To: <1235170271-10694-1-git-send-email-eletuchy@gmail.com>

```
Looked sensible, queued.

Thanks.

```

## Jeff King, 2009-02-22 23:03

Subject: Re: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <20090222230358.GA19011@coredump.intra.peff.net>
URL: https://gitlist.dev/e/20090222230358.GA19011%40coredump.intra.peff.net
In-Reply-To: <1235170271-10694-1-git-send-email-eletuchy@gmail.com>

```
On Fri, Feb 20, 2009 at 02:51:11PM -0800, eletuchy@gmail.com wrote:

> @@ -1975,6 +1975,9 @@ static int git_blame_config(const char *var, const char *value, void *cb)
>  		blank_boundary = git_config_bool(var, value);
>  		return 0;
>  	}
> +	if (!strcmp(var, "blame.date") && value[0]) {
> +		blame_date_mode = parse_date_format(value);
> +	}
>  	return git_default_config(var, value, cb);
>  }

When there is a config value we are expecting to have a value rather
than a boolean, we usually print an error rather than silently
discarding. IOW, something like this:

  if (!strcmp(var, "blame.date")) {
          if (!value)
                  return config_error_nonbool(var);
          blame_date_mode = parse_date_format(value);
  }

> +	switch (blame_date_mode) {
> +	case DATE_RFC2822:
> +		blame_date_width = sizeof("Thu, 19 Oct 2006 16:00:04 -0700");
> +		break;
> +	case DATE_ISO8601:
> +		blame_date_width = sizeof("2006-10-19 16:00:04 -0700");
> +		break;
> +	case DATE_SHORT:
> +		blame_date_width = sizeof("2006-10-19");
> +		break;
> +	case DATE_RELATIVE:
> +		/* unfortunately "normal" is the fallback for "relative" */
> +		/* blame_date_width = sizeof("14 minutes ago"); */
> +		/* break; */
> +	case DATE_LOCAL:
> +	case DATE_NORMAL:
> +		blame_date_width = sizeof("Thu Oct 19 16:00:04 2006 -0700");
> +		break;
> +	}
> +	blame_date_width -= 1; /* strip the null */

Maybe this should be a date_format_width() library function?


Other than that, the patch looks reasonable to me.

-Peff

```

## Eugene Letuchy, 2009-02-23 09:09

Subject: Re: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <fbb390660902230109k15ed55b7le2ab08bb7dc274b6@mail.gmail.com>
URL: https://gitlist.dev/e/fbb390660902230109k15ed55b7le2ab08bb7dc274b6%40mail.gmail.com
In-Reply-To: <20090222230358.GA19011@coredump.intra.peff.net>

```
On Sun, Feb 22, 2009 at 3:03 PM, Jeff King <peff@peff.net> wrote:
> On Fri, Feb 20, 2009 at 02:51:11PM -0800, eletuchy@gmail.com wrote:
>
>> @@ -1975,6 +1975,9 @@ static int git_blame_config(const char *var, const char *value, void *cb)
>>               blank_boundary = git_config_bool(var, value);
>>               return 0;
>>       }
>> +     if (!strcmp(var, "blame.date") && value[0]) {
>> +             blame_date_mode = parse_date_format(value);
>> +     }
>>       return git_default_config(var, value, cb);
>>  }
>
> When there is a config value we are expecting to have a value rather
> than a boolean, we usually print an error rather than silently
> discarding. IOW, something like this:
>
>  if (!strcmp(var, "blame.date")) {
>          if (!value)
>                  return config_error_nonbool(var);
>          blame_date_mode = parse_date_format(value);
>  }
>

I'll make that change to the patch.

>> +     switch (blame_date_mode) {
>> +     case DATE_RFC2822:
>> +             blame_date_width = sizeof("Thu, 19 Oct 2006 16:00:04 -0700");
>> +             break;
>> +     case DATE_ISO8601:
>> +             blame_date_width = sizeof("2006-10-19 16:00:04 -0700");
>> +             break;
>> +     case DATE_SHORT:
>> +             blame_date_width = sizeof("2006-10-19");
>> +             break;
>> +     case DATE_RELATIVE:
>> +             /* unfortunately "normal" is the fallback for "relative" */
>> +             /* blame_date_width = sizeof("14 minutes ago"); */
>> +             /* break; */
>> +     case DATE_LOCAL:
>> +     case DATE_NORMAL:
>> +             blame_date_width = sizeof("Thu Oct 19 16:00:04 2006 -0700");
>> +             break;
>> +     }
>> +     blame_date_width -= 1; /* strip the null */
>
> Maybe this should be a date_format_width() library function?
>

I think that's a possible change, but unfortunately my next two
patches would not apply cleanly with a date_format_width change.

I'm a n00b with respect to git contribution, but is there a procedure
for pushing my blame_date branch remotely so that it's possible to
track a series of patches?

>
> Other than that, the patch looks reasonable to me.
>
> -Peff
>



-- 
Eugene

```

## Junio C Hamano, 2009-02-23 16:33

Subject: Re: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <7v3ae5un84.fsf@gitster.siamese.dyndns.org>
URL: https://gitlist.dev/e/7v3ae5un84.fsf%40gitster.siamese.dyndns.org
In-Reply-To: <20090222230358.GA19011@coredump.intra.peff.net>

```
Jeff King <peff@peff.net> writes:

> On Fri, Feb 20, 2009 at 02:51:11PM -0800, eletuchy@gmail.com wrote:
>
>> @@ -1975,6 +1975,9 @@ static int git_blame_config(const char *var, const char *value, void *cb)
>>  		blank_boundary = git_config_bool(var, value);
>>  		return 0;
>>  	}
>> +	if (!strcmp(var, "blame.date") && value[0]) {
>> +		blame_date_mode = parse_date_format(value);
>> +	}
>>  	return git_default_config(var, value, cb);
>>  }
>
> When there is a config value we are expecting to have a value rather
> than a boolean, we usually print an error rather than silently
> discarding.

Oops, missed that.  Yes, this needs fixing.

Thanks.

```

## Jeff King, 2009-02-24 05:00

Subject: Re: [PATCH 1/2] Make git blame's date output format configurable, like git log
Message-ID: <20090224050028.GB4615@coredump.intra.peff.net>
URL: https://gitlist.dev/e/20090224050028.GB4615%40coredump.intra.peff.net
In-Reply-To: <fbb390660902230109k15ed55b7le2ab08bb7dc274b6@mail.gmail.com>

```
On Mon, Feb 23, 2009 at 01:09:13AM -0800, Eugene Letuchy wrote:

> > Maybe this should be a date_format_width() library function?
> 
> I think that's a possible change, but unfortunately my next two
> patches would not apply cleanly with a date_format_width change.
> 
> I'm a n00b with respect to git contribution, but is there a procedure
> for pushing my blame_date branch remotely so that it's possible to
> track a series of patches?

Updating previous work depends on whether it has been picked up in
'next' by Junio; once patches are there, they cannot be rewritten. In
that case, you can send a follow-up patch.

In your case, though, the patch is still in 'pu', so you can repost. So
I think it makes sense to use "rebase -i" (or the tool of your choice)
to make a cleaned up series, and then repost the whole thing as a
series.

-Peff

```