git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[RFC/PATCH 16/17] diff: buffer output in emit_line_0

From
Stefan Beller <sbeller@google.com>
Date
Sep 13, 2016, 04:46 UTC
Message-ID
<20160913044613.1037-17-sbeller@google.com>
In-Reply-To
<20160913044613.1037-1-sbeller@google.com>

emit_line_0 factors out the emission part into emit_line_emission, and depending on the diff_options->use_buffer the emission will be performed directly when calling emit_line_0 or after the whole process is done, i.e. by buffering we have add the possibility for a second pass over the whole output before doing the actual output.

In 6440d34 (2012-03-14, diff: tweak a _copy_ of diff_options with word-diff) we introduced a duplicate diff options struct for word emissions as we may have different regex settings in there.

When buffering the output, we need to operate on just one buffer, so we have to copy back the emissions of the word buffer into the main buffer.

Unconditionally enable output via buffer in this patch as it yields a great opportunity for testing, i.e. all the diff tests from the test suite pass without having reordering issues (i.e. only parts of the output got buffered, and we forgot to buffer other parts). The test suite passes, which gives confidence that we converted all functions to use emit_line_* for output.

Signed-off-by: Stefan Beller <sbeller@google.com>
---
 diff.c | 151 +++++++++++++++++++++++++++++++++++++++++++++++++++++------------
 diff.h |  18 ++++++++
 2 files changed, 142 insertions(+), 27 deletions(-)
diff --git a/diff.c b/diff.c
index 68da135..e240e89 100644
--- a/diff.c
+++ b/diff.c
@@ -449,42 +449,86 @@ static void check_blank_at_eof(mmfile_t *mf1, mmfile_t *mf2,
 	ecbdata->blank_at_eof_in_postimage = (at - l2) + 1;
 }
 
-static void emit_line_0(struct diff_options *o, const char *set, const char *reset,
-			int first, const char *line, int len)
+static void emit_line_emission(struct diff_options *o, struct line_emission *e)
 {
-	int has_trailing_newline, has_trailing_carriage_return;
-	int nofirst;
 	FILE *file = o->file;
 
 	fputs(diff_line_prefix(o), file);
 
+	if (e->len || e->first) {
+		if (e->set)
+			fputs(e->set, file);
+		if (e->first)
+			fputc(e->first, file);
+		if (e->whitespace_check) {
+			if (e->reset)
+				fputs(e->reset, file);
+			ws_check_emit(e->line, e->len, e->ws_rule,
+				      file, e->set, e->reset, e->ws);
+		} else {
+			fwrite(e->line, e->len, 1, file);
+			if (e->reset)
+				fputs(e->reset, file);
+		}
+	}
+	if (e->has_trailing_carriage_return)
+		fputc('\r', file);
+	if (e->has_trailing_newline)
+		fputc('\n', file);
+}
+
+static void free_line_emission(struct line_emission *e)
+{
+	/*
+	 * No need to free set, reset, ws as they always point to the
+	 * diff_colors[] static array. We don't own that memory.
+	 */
+	free((char*)e->line);
+}
+
+static void append_line_emission_to_buffer(struct diff_options *o,
+			     struct line_emission *e)
+{
+	struct line_emission *f;
+	ALLOC_GROW(o->line_buffer,
+		   o->line_buffer_nr + 1,
+		   o->line_buffer_alloc);
+	f = &o->line_buffer[o->line_buffer_nr++];
+	memcpy(f, e, sizeof(*e));
+
+	/* duplicate the line for now as it is not a stable pointer */
+	f->line = xmemdupz(e->line, e->len);
+}
+
+static void emit_line_0(struct diff_options *o, const char *set,
+			const char *reset, int first, const char *line, int len)
+{
+	int nofirst;
+	struct line_emission e = LINE_EMISSION_INIT;
+
 	if (len == 0) {
-		has_trailing_newline = (first == '\n');
-		has_trailing_carriage_return = (first == '\r');
-		nofirst = has_trailing_newline || has_trailing_carriage_return;
+		e.has_trailing_newline = (first == '\n');
+		e.has_trailing_carriage_return = (first == '\r');
+		nofirst = e.has_trailing_newline || e.has_trailing_carriage_return;
 	} else {
-		has_trailing_newline = (len > 0 && line[len-1] == '\n');
-		if (has_trailing_newline)
+		e.has_trailing_newline = (len > 0 && line[len-1] == '\n');
+		if (e.has_trailing_newline)
 			len--;
-		has_trailing_carriage_return = (len > 0 && line[len-1] == '\r');
-		if (has_trailing_carriage_return)
+		e.has_trailing_carriage_return = (len > 0 && line[len-1] == '\r');
+		if (e.has_trailing_carriage_return)
 			len--;
 		nofirst = 0;
 	}
 
-	if (len || !nofirst) {
-		if (set)
-			fputs(set, file);
-		if (!nofirst)
-			fputc(first, file);
-		fwrite(line, len, 1, file);
-		if (reset)
-			fputs(reset, file);
-	}
-	if (has_trailing_carriage_return)
-		fputc('\r', file);
-	if (has_trailing_newline)
-		fputc('\n', file);
+	e.set = set;
+	e.first = !nofirst ? first : 0;
+	e.line = line;
+	e.len = len;
+	e.reset = reset;
+	if (!o->use_buffer)
+		emit_line_emission(o, &e);
+	else
+		append_line_emission_to_buffer(o, &e);
 }
 
 static void emit_line(struct diff_options *o, const char *set, const char *reset,
@@ -540,9 +584,22 @@ static void emit_line_checked(const char *reset,
 		emit_line_0(ecbdata->opt, ws, reset, sign, line, len);
 	else {
 		/* Emit just the prefix, then the rest. */
-		emit_line_0(ecbdata->opt, set, reset, sign, "", 0);
-		ws_check_emit(line, len, ecbdata->ws_rule,
-			      ecbdata->opt->file, set, reset, ws);
+		if (ecbdata->opt->use_buffer) {
+			struct line_emission e;
+			e.line = line;
+			e.len = len;
+			e.ws_rule = ecbdata->ws_rule;
+			e.set = set;
+			e.reset = reset;
+			e.ws = ws;
+			e.whitespace_check = 1;
+			e.first = sign;
+			append_line_emission_to_buffer(ecbdata->opt, &e);
+		} else {
+			emit_line_0(ecbdata->opt, set, reset, sign, "", 0);
+			ws_check_emit(line, len, ecbdata->ws_rule,
+				      ecbdata->opt->file, set, reset, ws);
+		}
 	}
 }
 
@@ -1093,6 +1150,22 @@ static void diff_words_flush(struct emit_callback *ecbdata)
 	if (ecbdata->diff_words->minus.text.size ||
 	    ecbdata->diff_words->plus.text.size)
 		diff_words_show(ecbdata->diff_words);
+
+	if (ecbdata->diff_words->opt->line_buffer_nr) {
+		int i;
+		for (i = 0; i < ecbdata->diff_words->opt->line_buffer_nr; i++) {
+			struct line_emission *e =
+				&ecbdata->diff_words->opt->line_buffer[i];
+			ALLOC_GROW(ecbdata->opt->line_buffer,
+				   ecbdata->opt->line_buffer_nr + 1,
+				   ecbdata->opt->line_buffer_alloc);
+			memcpy(&ecbdata->opt->line_buffer
+					[ecbdata->opt->line_buffer_nr],
+				e, sizeof(*e));
+			ecbdata->opt->line_buffer_nr++;
+		}
+		ecbdata->diff_words->opt->line_buffer_nr = 0;
+	}
 }
 
 static void diff_filespec_load_driver(struct diff_filespec *one)
@@ -1128,6 +1201,12 @@ static void init_diff_words_data(struct emit_callback *ecbdata,
 		xcalloc(1, sizeof(struct diff_words_data));
 	ecbdata->diff_words->type = o->word_diff;
 	ecbdata->diff_words->opt = o;
+
+	/* Create our own buffer if needed. */
+	o->line_buffer = NULL;
+	o->line_buffer_nr = 0;
+	o->line_buffer_alloc = 0;
+
 	if (!o->word_regex)
 		o->word_regex = userdiff_word_regex(one);
 	if (!o->word_regex)
@@ -4649,11 +4728,29 @@ static void diff_flush_patch_all_file_pairs(struct diff_options *o)
 {
 	int i;
 	struct diff_queue_struct *q = &diff_queued_diff;
+	/*
+	 * TODO:
+	 * For testing purposes we want to make sure the diff machinery
+	 * works with the buffer. If there is anything emitted outside the
+	 * emit_line_emission, then the order is screwed up and the tests
+	 * will fail.
+	 *
+	 * We'll unset this flag in a later patch.
+	 */
+	o->use_buffer = 1;
 	for (i = 0; i < q->nr; i++) {
 		struct diff_filepair *p = q->queue[i];
 		if (check_pair_status(p))
 			diff_flush_patch(p, o);
 	}
+
+	if (o->use_buffer) {
+		for (i = 0; i < o->line_buffer_nr; i++) {
+			emit_line_emission(o, &o->line_buffer[i]);
+			free_line_emission(&o->line_buffer[i]);
+		}
+		o->line_buffer_nr = 0;
+	}
 }
 
 void diff_flush(struct diff_options *options)
diff --git a/diff.h b/diff.h
index cc5d038..4df6aa5 100644
--- a/diff.h
+++ b/diff.h
@@ -110,6 +110,20 @@ enum diff_words_type {
 	DIFF_WORDS_COLOR
 };
 
+struct line_emission {
+	const char *set;
+	const char *line;
+	const char *ws;
+	const char *reset;
+	int first;
+	int len;
+	int whitespace_check;
+	unsigned ws_rule;
+	int has_trailing_carriage_return;
+	int has_trailing_newline;
+};
+#define LINE_EMISSION_INIT {NULL, NULL, NULL, NULL, 0, 0, 0, 0, 0, 0 }
+
 struct diff_options {
 	const char *orderfile;
 	const char *pickaxe;
@@ -178,6 +192,10 @@ struct diff_options {
 	void *output_prefix_data;
 
 	int diff_path_counter;
+
+	int use_buffer;
+	struct line_emission *line_buffer;
+	int line_buffer_nr, line_buffer_alloc;
 };
 
 void emit_line_fmt(struct diff_options *o, const char *set, const char *reset,

-- 
2.10.0.21.g1da280f.dirty
Previous: Stefan BellerNext: Junio C Hamano
Message 23 of 28 in “[RFC/PATCH 00/17]”
  1. 00/17 Stefan Beller, Sep 13, 2016
  2. 01/17 diff: move line ending check into emit_hunk_headerStefan Beller, Sep 13, 2016
  3. René ScharfeSep 13, 2016
  4. Stefan BellerSep 13, 2016
  5. 02/17 diff: emit_{add, del, context}_line to increase {pre,post}image line countStefan Beller, Sep 13, 2016
  6. 03/17 diff.c: drop tautologous condition in emit_line_0Stefan Beller, Sep 13, 2016
  7. 04/17 diff.c: factor out diff_flush_patch_all_file_pairsStefan Beller, Sep 13, 2016
  8. 05/17 diff.c: emit_line_0 can handle no color settingStefan Beller, Sep 13, 2016
  9. Junio C HamanoSep 13, 2016
  10. 08/17 diff.c: convert emit_rewrite_diff to use emit_line_*Stefan Beller, Sep 13, 2016
  11. 07/17 diff.c: convert builtin_diff to use emit_line_*Stefan Beller, Sep 13, 2016
  12. 06/17 diff.c: convert fn_out_consume to use emit_line_*Stefan Beller, Sep 13, 2016
  13. Junio C HamanoSep 13, 2016
  14. 09/17 diff.c: convert emit_rewrite_lines to use emit_line_*Stefan Beller, Sep 13, 2016
  15. 13/17 diff.c: convert word diffing to use emit_line_*Stefan Beller, Sep 13, 2016
  16. 12/17 diff.c: convert show_stats to use emit_line_*Stefan Beller, Sep 13, 2016
  17. 10/17 submodule.c: convert show_submodule_summary to use emit_line_fmtStefan Beller, Sep 13, 2016
  18. Junio C HamanoSep 13, 2016
  19. Stefan BellerSep 13, 2016
  20. 14/17 diff.c: convert diff_flush to use emit_line_*Stefan Beller, Sep 13, 2016
  21. 17/17 diff.c: color moved lines differentlyStefan Beller, Sep 13, 2016
  22. 11/17 diff.c: convert emit_binary_diff_body to use emit_line_*Stefan Beller, Sep 13, 2016
  23. 16/17 diff: buffer output in emit_line_0Stefan Beller, Sep 13, 2016
  24. Junio C HamanoSep 13, 2016
  25. Stefan BellerSep 13, 2016
  26. Junio C HamanoSep 13, 2016
  27. Stefan BellerSep 13, 2016
  28. 15/17 diff.c: convert diff_summary to use emit_line_*Stefan Beller, Sep 13, 2016

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.