[PATCH v5 1/4] pretty: expand tabs in indented logs to make things line up properly

[Junio C Hamano <gitster@pobox.com>]

From: Linus Torvalds <torvalds@linux-foundation.org>

A commit log message sometimes tries to line things up using tabs,
assuming fixed-width font with the standard 8-place tab settings.
Viewing such a commit however does not work well in "git log", as
we indent the lines by prefixing 4 spaces in front of them.

This should all line up:

  Column 1	Column 2
  --------	--------
  A		B
  ABCD		EFGH
  SPACES        Instead of Tabs

Even with multi-byte UTF8 characters:

  Column 1	Column 2
  --------	--------
  Ä		B
  åäö		100
  A Møøse	once bit my sister..

Tab-expand the lines in "git log --expand-tabs" output before
prefixing 4 spaces.

This is based on the patch by Linus Torvalds, but at this step, we
require an explicit command line option to enable the behaviour.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/pretty-options.txt |  5 +++
 commit.h                         |  1 +
 log-tree.c                       |  1 +
 pretty.c                         | 71 ++++++++++++++++++++++++++++++++++++++--
 revision.c                       |  2 ++
 revision.h                       |  1 +
 6 files changed, 79 insertions(+), 2 deletions(-)

diff --git a/Documentation/pretty-options.txt b/Documentation/pretty-options.txt
index 4b659ac..d820653 100644
--- a/Documentation/pretty-options.txt
+++ b/Documentation/pretty-options.txt
@@ -42,6 +42,11 @@ people using 80-column terminals.
 	verbatim; this means that invalid sequences in the original
 	commit may be copied to the output.
 
+--expand-tabs::
+	Perform a tab expansion (replace each tab with enough spaces
+	to fill to the next display column that is multiple of 8)
+	in the log message before showing it in the output.
+
 ifndef::git-rev-list[]
 --notes[=<ref>]::
 	Show the notes (see linkgit:git-notes[1]) that annotate the
diff --git a/commit.h b/commit.h
index 5d58be0..a7ef682 100644
--- a/commit.h
+++ b/commit.h
@@ -147,6 +147,7 @@ struct pretty_print_context {
 	int preserve_subject;
 	struct date_mode date_mode;
 	unsigned date_mode_explicit:1;
+	unsigned expand_tabs_in_log:1;
 	int need_8bit_cte;
 	char *notes_message;
 	struct reflog_walk_info *reflog_info;
diff --git a/log-tree.c b/log-tree.c
index 60f9839..78a5381 100644
--- a/log-tree.c
+++ b/log-tree.c
@@ -683,6 +683,7 @@ void show_log(struct rev_info *opt)
 	ctx.fmt = opt->commit_format;
 	ctx.mailmap = opt->mailmap;
 	ctx.color = opt->diffopt.use_color;
+	ctx.expand_tabs_in_log = opt->expand_tabs_in_log;
 	ctx.output_encoding = get_log_output_encoding();
 	if (opt->from_ident.mail_begin && opt->from_ident.name_begin)
 		ctx.from_ident = &opt->from_ident;
diff --git a/pretty.c b/pretty.c
index 92b2870..c8b075d 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1629,6 +1629,72 @@ void pp_title_line(struct pretty_print_context *pp,
 	strbuf_release(&title);
 }
 
+static int pp_utf8_width(const char *start, const char *end)
+{
+	int width = 0;
+	size_t remain = end - start;
+
+	while (remain) {
+		int n = utf8_width(&start, &remain);
+		if (n < 0 || !start)
+			return -1;
+		width += n;
+	}
+	return width;
+}
+
+static void strbuf_add_tabexpand(struct strbuf *sb,
+				 const char *line, int linelen)
+{
+	const char *tab;
+
+	while ((tab = memchr(line, '\t', linelen)) != NULL) {
+		int width = pp_utf8_width(line, tab);
+
+		/*
+		 * If it wasn't well-formed utf8, or it
+		 * had characters with badly defined
+		 * width (control characters etc), just
+		 * give up on trying to align things.
+		 */
+		if (width < 0)
+			break;
+
+		/* Output the data .. */
+		strbuf_add(sb, line, tab - line);
+
+		/* .. and the de-tabified tab */
+		strbuf_addchars(sb, ' ', 8 - (width % 8));
+
+		/* Skip over the printed part .. */
+		linelen -= tab + 1 - line;
+		line = tab + 1;
+	}
+
+	/*
+	 * Print out everything after the last tab without
+	 * worrying about width - there's nothing more to
+	 * align.
+	 */
+	strbuf_add(sb, line, linelen);
+}
+
+/*
+ * pp_handle_indent() prints out the intendation, and
+ * the whole line (without the final newline), after
+ * de-tabifying.
+ */
+static void pp_handle_indent(struct pretty_print_context *pp,
+			     struct strbuf *sb, int indent,
+			     const char *line, int linelen)
+{
+	strbuf_addchars(sb, ' ', indent);
+	if (pp->expand_tabs_in_log)
+		strbuf_add_tabexpand(sb, line, linelen);
+	else
+		strbuf_add(sb, line, linelen);
+}
+
 void pp_remainder(struct pretty_print_context *pp,
 		  const char **msg_p,
 		  struct strbuf *sb,
@@ -1653,8 +1719,9 @@ void pp_remainder(struct pretty_print_context *pp,
 
 		strbuf_grow(sb, linelen + indent + 20);
 		if (indent)
-			strbuf_addchars(sb, ' ', indent);
-		strbuf_add(sb, line, linelen);
+			pp_handle_indent(pp, sb, indent, line, linelen);
+		else
+			strbuf_add(sb, line, linelen);
 		strbuf_addch(sb, '\n');
 	}
 }
diff --git a/revision.c b/revision.c
index df56fce..e662230 100644
--- a/revision.c
+++ b/revision.c
@@ -1915,6 +1915,8 @@ static int handle_revision_opt(struct rev_info *revs, int argc, const char **arg
 		revs->verbose_header = 1;
 		revs->pretty_given = 1;
 		get_commit_format(arg+9, revs);
+	} else if (!strcmp(arg, "--expand-tabs")) {
+		revs->expand_tabs_in_log = 1;
 	} else if (!strcmp(arg, "--show-notes") || !strcmp(arg, "--notes")) {
 		revs->show_notes = 1;
 		revs->show_notes_given = 1;
diff --git a/revision.h b/revision.h
index 23857c0..4079753 100644
--- a/revision.h
+++ b/revision.h
@@ -133,6 +133,7 @@ struct rev_info {
 			show_notes_given:1,
 			show_signature:1,
 			pretty_given:1,
+			expand_tabs_in_log:1,
 			abbrev_commit:1,
 			abbrev_commit_given:1,
 			zero_commit:1,
-- 
2.8.1-251-g9997610