git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
To: Liam Beguin <liambeguin@gmail.com>
Cc: git@vger.kernel.org, gitster@pobox.com, avarab@gmail.com
Subject: Re: [PATCH 4/5] rebase -i: learn to abbreviate command names
Date: Tue, 28 Nov 2017 00:04:45 +0100 (CET)	[thread overview]
Message-ID: <alpine.DEB.2.21.1.1711272344290.6482@virtualbox> (raw)
In-Reply-To: <20171127045514.25647-5-liambeguin@gmail.com>

Hi Liam,

On Sun, 26 Nov 2017, Liam Beguin wrote:

> diff --git a/Documentation/rebase-config.txt b/Documentation/rebase-config.txt
> index 30ae08cb5a4b..0820b60f6e12 100644
> --- a/Documentation/rebase-config.txt
> +++ b/Documentation/rebase-config.txt
> @@ -30,3 +30,22 @@ rebase.instructionFormat::
>  	A format string, as specified in linkgit:git-log[1], to be used for the
>  	todo list during an interactive rebase.  The format will
>  	automatically have the long commit hash prepended to the format.
> +
> +rebase.abbreviateCommands::
> +	If set to true, `git rebase` will use abbreviated command names in the
> +	todo list resulting in something like this:
> +
> +-------------------------------------------
> +	p deadbee The oneline of the commit
> +	p fa1afe1 The oneline of the next commit
> +	...
> +-------------------------------------------

I *think* that AsciiDoc will render this in a different way from what we
want, but I am not an AsciiDoc expert. In my hands, I always had to add a
single + in an otherwise empty line to start a new indented paragraph *and
then continue with non-indented lines*.

> diff --git a/sequencer.c b/sequencer.c
> index 810b7850748e..aa01e8bd9280 100644
> --- a/sequencer.c
> +++ b/sequencer.c
> @@ -795,6 +795,13 @@ static const char *command_to_string(const enum todo_command command)
>  	die("Unknown command: %d", command);
>  }
>  
> +static const char command_to_char(const enum todo_command command)
> +{
> +	if (command < TODO_COMMENT && todo_command_info[command].c)
> +		return todo_command_info[command].c;
> +	return -1;

My initial reaction was: should we return comment_line_char instead of -1
here? Only after reading how this is called did I realize that the idea is
to use full command names if there is no abbreviation. Not sure whether
this is worth a code comment. What do you think?

> +}
> +
>  static int is_noop(const enum todo_command command)
>  {
>  	return TODO_NOOP <= command;
> @@ -1242,15 +1249,16 @@ static int parse_insn_line(struct todo_item *item, const char *bol, char *eol)
>  		return 0;
>  	}
>  
> -	for (i = 0; i < TODO_COMMENT; i++)
> +	for (i = 0; i < TODO_COMMENT; i++) {
>  		if (skip_prefix(bol, todo_command_info[i].str, &bol)) {
>  			item->command = i;
>  			break;
> -		} else if (bol[1] == ' ' && *bol == todo_command_info[i].c) {
> +		} else if (bol[1] == ' ' && *bol == command_to_char(i)) {
>  			bol++;
>  			item->command = i;
>  			break;
>  		}
> +	}
>  	if (i >= TODO_COMMENT)
>  		return -1;
>  

I would prefer this hunk to be skipped, it does not really do anything if
I understand correctly.

> @@ -2443,8 +2451,8 @@ void append_signoff(struct strbuf *msgbuf, int ignore_footer, unsigned flag)
>  	strbuf_release(&sob);
>  }
>  
> -int sequencer_make_script(int keep_empty, FILE *out,
> -		int argc, const char **argv)
> +int sequencer_make_script(int keep_empty, int abbreviate_commands, FILE *out,
> +			  int argc, const char **argv)
>  {
>  	char *format = NULL;
>  	struct pretty_print_context pp = {0};
> @@ -2483,7 +2491,9 @@ int sequencer_make_script(int keep_empty, FILE *out,
>  		strbuf_reset(&buf);
>  		if (!keep_empty && is_original_commit_empty(commit))
>  			strbuf_addf(&buf, "%c ", comment_line_char);
> -		strbuf_addf(&buf, "pick %s ", oid_to_hex(&commit->object.oid));
> +		strbuf_addf(&buf, "%s %s ",
> +			    abbreviate_commands ? "p" : "pick",
> +			    oid_to_hex(&commit->object.oid));

I guess the compiler will optimize this code so that the conditional is
evaluated only once. Not that this is performance critical ;-)

>  		pretty_print_commit(&pp, commit, &buf);
>  		strbuf_addch(&buf, '\n');
>  		fputs(buf.buf, out);
> @@ -2539,7 +2549,7 @@ int add_exec_commands(const char *command)
>  	return 0;
>  }
>  
> -int transform_todo_ids(int shorten_ids)
> +int transform_todo_ids(int shorten_ids, int abbreviate_commands)
>  {
>  	const char *todo_file = rebase_path_todo();
>  	struct todo_list todo_list = TODO_LIST_INIT;
> @@ -2575,19 +2585,33 @@ int transform_todo_ids(int shorten_ids)
>  			todo_list.items[i + 1].offset_in_buf :
>  			todo_list.buf.len;
>  
> -		if (item->command >= TODO_EXEC && item->command != TODO_DROP)
> -			fwrite(p, eol - bol, 1, out);
> -		else {
> +		if (item->command >= TODO_EXEC && item->command != TODO_DROP) {
> +			if (!abbreviate_commands || command_to_char(item->command) < 0) {
> +				fwrite(p, eol - bol, 1, out);
> +			} else {
> +				const char *end_of_line = strchrnul(p, '\n');
> +				p += strspn(p, " \t"); /* skip whitespace */
> +				p += strcspn(p, " \t"); /* skip command */
> +				fprintf(out, "%c%.*s\n",
> +					command_to_char(item->command),
> +					(int)(end_of_line - p), p);
> +			}
> +		} else {
>  			const char *id = shorten_ids ?
>  				short_commit_name(item->commit) :
>  				oid_to_hex(&item->commit->object.oid);
> -			int len;
>  
> -			p += strspn(p, " \t"); /* left-trim command */
> -			len = strcspn(p, " \t"); /* length of command */
> -
> -			fprintf(out, "%.*s %s %.*s\n",
> -				len, p, id, item->arg_len, item->arg);
> +			if (abbreviate_commands) {
> +				fprintf(out, "%c %s %.*s\n",
> +					command_to_char(item->command),
> +					id, item->arg_len, item->arg);
> +			} else {
> +				int len;
> +				p += strspn(p, " \t"); /* left-trim command */
> +				len = strcspn(p, " \t"); /* length of command */
> +				fprintf(out, "%.*s %s %.*s\n",
> +					len, p, id, item->arg_len, item->arg);
> +			}

This hunk changes indentation quite a bit, therefore it is a bit harder to
read than necessary (and the resulting code, too, as it is more smooshed
against the 80-column boundary on the right).

How about this instead:

-		if (item->command >= TODO_EXEC && item->command != TODO_DROP)
+		if (abbreviate_commands && command_to_char(item->command)) {
+			const char *id = shorten_ids ?
+				short_commit_name(item->commit) :
+				oid_to_hex(&item->commit->object.oid);
+			fprintf(out, "%c %s %.*s\n",
+				command_to_char(item->command),
+				id, item->arg_len, item->arg);
+		} else if (item->command >= TODO_EXEC &&
+			 item->command != TODO_DROP)

i.e. test first for the short and sweet case that we want (and can)
abbreviate the command, otherwise keep the code as before?

Ciao,
Dscho

  parent reply	other threads:[~2017-11-27 23:04 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-27  4:55 [PATCH 0/5] rebase -i: add config to abbreviate command names Liam Beguin
2017-11-27  4:55 ` [PATCH 1/5] Documentation: move rebase.* configs to new file Liam Beguin
2017-11-27 21:27   ` Johannes Schindelin
2017-11-27  4:55 ` [PATCH 2/5] Documentation: use preferred name for the 'todo list' script Liam Beguin
2017-11-27 21:28   ` Johannes Schindelin
2017-11-27  4:55 ` [PATCH 3/5] rebase -i: add exec commands via the rebase--helper Liam Beguin
2017-11-27  5:14   ` Junio C Hamano
2017-11-27 21:41     ` Johannes Schindelin
2017-11-27 23:45       ` Junio C Hamano
2017-11-29  2:01     ` liam Beguin
2017-11-27 22:42   ` Johannes Schindelin
2017-11-27 23:48     ` Junio C Hamano
2017-11-29  2:06     ` liam Beguin
2017-11-29 21:35       ` Johannes Schindelin
2017-11-27  4:55 ` [PATCH 4/5] rebase -i: learn to abbreviate command names Liam Beguin
2017-11-27  5:19   ` Junio C Hamano
2017-11-29  2:08     ` liam Beguin
2017-11-27 23:04   ` Johannes Schindelin [this message]
2017-11-27 23:11     ` Jeff King
2017-11-29  2:11       ` liam Beguin
2017-11-29  2:10     ` liam Beguin
2017-11-29 21:40       ` Johannes Schindelin
2017-12-03  1:18         ` Junio C Hamano
2017-11-27  4:55 ` [PATCH 5/5] t3404: add test case for abbreviated commands Liam Beguin
2017-11-27  5:28   ` Junio C Hamano
2017-11-27 23:16   ` Johannes Schindelin
2017-11-27  5:23 ` [PATCH 0/5] rebase -i: add config to abbreviate command names Junio C Hamano
2017-11-29  1:56   ` liam Beguin
2017-12-03 22:17 ` [PATCH v2 0/9] " Liam Beguin
2017-12-03 22:17   ` [PATCH v2 1/9] Documentation: move rebase.* configs to new file Liam Beguin
2017-12-03 22:17   ` [PATCH v2 2/9] Documentation: use preferred name for the 'todo list' script Liam Beguin
2017-12-03 22:17   ` [PATCH v2 3/9] rebase -i: set commit to null in exec commands Liam Beguin
2017-12-03 22:17   ` [PATCH v2 4/9] rebase -i: refactor transform_todo_ids Liam Beguin
2017-12-04 14:42     ` Johannes Schindelin
2017-12-04 16:09       ` Junio C Hamano
2017-12-05  3:36         ` liam Beguin
2017-12-05 12:35           ` Junio C Hamano
2017-12-05  3:39       ` liam Beguin
2017-12-03 22:17   ` [PATCH v2 5/9] rebase -i: replace reference to sha1 with oid Liam Beguin
2017-12-03 22:17   ` [PATCH v2 6/9] rebase -i: update functions to use a flags parameter Liam Beguin
2017-12-04 15:46     ` Johannes Schindelin
2017-12-04 16:06       ` Johannes Schindelin
2017-12-05  3:42       ` liam Beguin
2017-12-05 12:37         ` Junio C Hamano
2017-12-05 12:41           ` Kerry, Richard
2017-12-05 14:42             ` liam Beguin
2017-12-05 16:05             ` Junio C Hamano
2017-12-05 16:14               ` Kerry, Richard
2017-12-03 22:17   ` [PATCH v2 7/9] rebase -i -x: add exec commands via the rebase--helper Liam Beguin
2017-12-03 22:17   ` [PATCH v2 8/9] rebase -i: learn to abbreviate command names Liam Beguin
2017-12-25 12:48     ` Duy Nguyen
2017-12-25 15:39       ` Liam Beguin
2017-12-25 23:58         ` Duy Nguyen
2017-12-27 19:15       ` Junio C Hamano
2017-12-27 21:58         ` Liam Beguin
2017-12-28 18:55           ` Junio C Hamano
2017-12-03 22:17   ` [PATCH v2 9/9] t3404: add test case for abbreviated commands Liam Beguin
2017-12-04 16:07   ` [PATCH v2 0/9] rebase -i: add config to abbreviate command names Johannes Schindelin
2017-12-05 17:52 ` Liam Beguin
2017-12-05 17:52   ` [PATCH v3 1/9] Documentation: move rebase.* configs to new file Liam Beguin
2017-12-05 17:52   ` [PATCH v3 2/9] Documentation: use preferred name for the 'todo list' script Liam Beguin
2017-12-05 17:52   ` [PATCH v3 3/9] rebase -i: set commit to null in exec commands Liam Beguin
2017-12-05 17:52   ` [PATCH v3 4/9] rebase -i: refactor transform_todo_ids Liam Beguin
2017-12-05 17:52   ` [PATCH v3 5/9] rebase -i: replace reference to sha1 with oid Liam Beguin
2017-12-05 17:52   ` [PATCH v3 6/9] rebase -i: update functions to use a flags parameter Liam Beguin
2017-12-05 17:52   ` [PATCH v3 7/9] rebase -i -x: add exec commands via the rebase--helper Liam Beguin
2017-12-05 17:52   ` [PATCH v3 8/9] rebase -i: learn to abbreviate command names Liam Beguin
2017-12-05 17:52   ` [PATCH v3 9/9] t3404: add test case for abbreviated commands Liam Beguin
2017-12-05 22:21   ` [PATCH v2 0/9] rebase -i: add config to abbreviate command names Junio C Hamano
2017-12-06  2:42     ` liam Beguin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=alpine.DEB.2.21.1.1711272344290.6482@virtualbox \
    --to=johannes.schindelin@gmx.de \
    --cc=avarab@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=liambeguin@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).