From: Christian Couder <christian.couder@gmail.com>
To: git@vger.kernel.org
Cc: "Junio C Hamano" <gitster@pobox.com>, "Jeff King" <peff@peff.net>,
"Ævar Arnfjörð Bjarmason" <avarab@gmail.com>,
"Karsten Blees" <karsten.blees@gmail.com>,
"Nguyen Thai Ngoc Duy" <pclouds@gmail.com>,
"Johannes Schindelin" <Johannes.Schindelin@gmx.de>,
"Stefan Beller" <sbeller@google.com>,
"Matthieu Moy" <Matthieu.Moy@grenoble-inp.fr>,
"Christian Couder" <chriscool@tuxfamily.org>
Subject: [PATCH 05/83] builtin/apply: extract line_by_line_fuzzy_match() from match_fragment()
Date: Sun, 24 Apr 2016 15:33:05 +0200 [thread overview]
Message-ID: <1461504863-15946-6-git-send-email-chriscool@tuxfamily.org> (raw)
In-Reply-To: <1461504863-15946-1-git-send-email-chriscool@tuxfamily.org>
The match_fragment() function is very big and contains a big special case
algorithm that does line by line fuzzy matching. So let's extract this
algorithm in a separate line_by_line_fuzzy_match() function.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
---
builtin/apply.c | 129 ++++++++++++++++++++++++++++++++------------------------
1 file changed, 73 insertions(+), 56 deletions(-)
diff --git a/builtin/apply.c b/builtin/apply.c
index 78849e4..02239d9 100644
--- a/builtin/apply.c
+++ b/builtin/apply.c
@@ -2242,6 +2242,74 @@ static void update_pre_post_images(struct image *preimage,
postimage->nr -= reduced;
}
+static int line_by_line_fuzzy_match(struct image *img,
+ struct image *preimage,
+ struct image *postimage,
+ unsigned long try,
+ int try_lno,
+ int preimage_limit)
+{
+ int i;
+ size_t imgoff = 0;
+ size_t preoff = 0;
+ size_t postlen = postimage->len;
+ size_t extra_chars;
+ char *buf;
+ char *preimage_eof;
+ char *preimage_end;
+ struct strbuf fixed;
+ char *fixed_buf;
+ size_t fixed_len;
+
+ for (i = 0; i < preimage_limit; i++) {
+ size_t prelen = preimage->line[i].len;
+ size_t imglen = img->line[try_lno+i].len;
+
+ if (!fuzzy_matchlines(img->buf + try + imgoff, imglen,
+ preimage->buf + preoff, prelen))
+ return 0;
+ if (preimage->line[i].flag & LINE_COMMON)
+ postlen += imglen - prelen;
+ imgoff += imglen;
+ preoff += prelen;
+ }
+
+ /*
+ * Ok, the preimage matches with whitespace fuzz.
+ *
+ * imgoff now holds the true length of the target that
+ * matches the preimage before the end of the file.
+ *
+ * Count the number of characters in the preimage that fall
+ * beyond the end of the file and make sure that all of them
+ * are whitespace characters. (This can only happen if
+ * we are removing blank lines at the end of the file.)
+ */
+ buf = preimage_eof = preimage->buf + preoff;
+ for ( ; i < preimage->nr; i++)
+ preoff += preimage->line[i].len;
+ preimage_end = preimage->buf + preoff;
+ for ( ; buf < preimage_end; buf++)
+ if (!isspace(*buf))
+ return 0;
+
+ /*
+ * Update the preimage and the common postimage context
+ * lines to use the same whitespace as the target.
+ * If whitespace is missing in the target (i.e.
+ * if the preimage extends beyond the end of the file),
+ * use the whitespace from the preimage.
+ */
+ extra_chars = preimage_end - preimage_eof;
+ strbuf_init(&fixed, imgoff + extra_chars);
+ strbuf_add(&fixed, img->buf + try, imgoff);
+ strbuf_add(&fixed, preimage_eof, extra_chars);
+ fixed_buf = strbuf_detach(&fixed, &fixed_len);
+ update_pre_post_images(preimage, postimage,
+ fixed_buf, fixed_len, postlen);
+ return 1;
+}
+
static int match_fragment(struct image *img,
struct image *preimage,
struct image *postimage,
@@ -2251,7 +2319,7 @@ static int match_fragment(struct image *img,
int match_beginning, int match_end)
{
int i;
- char *fixed_buf, *buf, *orig, *target;
+ char *fixed_buf, *orig, *target;
struct strbuf fixed;
size_t fixed_len, postlen;
int preimage_limit;
@@ -2312,6 +2380,7 @@ static int match_fragment(struct image *img,
* There must be one non-blank context line that match
* a line before the end of img.
*/
+ char *buf;
char *buf_end;
buf = preimage->buf;
@@ -2331,61 +2400,9 @@ static int match_fragment(struct image *img,
* fuzzy matching. We collect all the line length information because
* we need it to adjust whitespace if we match.
*/
- if (ws_ignore_action == ignore_ws_change) {
- size_t imgoff = 0;
- size_t preoff = 0;
- size_t postlen = postimage->len;
- size_t extra_chars;
- char *preimage_eof;
- char *preimage_end;
- for (i = 0; i < preimage_limit; i++) {
- size_t prelen = preimage->line[i].len;
- size_t imglen = img->line[try_lno+i].len;
-
- if (!fuzzy_matchlines(img->buf + try + imgoff, imglen,
- preimage->buf + preoff, prelen))
- return 0;
- if (preimage->line[i].flag & LINE_COMMON)
- postlen += imglen - prelen;
- imgoff += imglen;
- preoff += prelen;
- }
-
- /*
- * Ok, the preimage matches with whitespace fuzz.
- *
- * imgoff now holds the true length of the target that
- * matches the preimage before the end of the file.
- *
- * Count the number of characters in the preimage that fall
- * beyond the end of the file and make sure that all of them
- * are whitespace characters. (This can only happen if
- * we are removing blank lines at the end of the file.)
- */
- buf = preimage_eof = preimage->buf + preoff;
- for ( ; i < preimage->nr; i++)
- preoff += preimage->line[i].len;
- preimage_end = preimage->buf + preoff;
- for ( ; buf < preimage_end; buf++)
- if (!isspace(*buf))
- return 0;
-
- /*
- * Update the preimage and the common postimage context
- * lines to use the same whitespace as the target.
- * If whitespace is missing in the target (i.e.
- * if the preimage extends beyond the end of the file),
- * use the whitespace from the preimage.
- */
- extra_chars = preimage_end - preimage_eof;
- strbuf_init(&fixed, imgoff + extra_chars);
- strbuf_add(&fixed, img->buf + try, imgoff);
- strbuf_add(&fixed, preimage_eof, extra_chars);
- fixed_buf = strbuf_detach(&fixed, &fixed_len);
- update_pre_post_images(preimage, postimage,
- fixed_buf, fixed_len, postlen);
- return 1;
- }
+ if (ws_ignore_action == ignore_ws_change)
+ return line_by_line_fuzzy_match(img, preimage, postimage,
+ try, try_lno, preimage_limit);
if (ws_error_action != correct_ws_error)
return 0;
--
2.8.1.300.g5fed0c0
next prev parent reply other threads:[~2016-04-24 13:39 UTC|newest]
Thread overview: 156+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-24 13:33 [PATCH 00/83] libify apply and use lib in am Christian Couder
2016-04-24 13:33 ` [PATCH 01/83] builtin/apply: make gitdiff_verify_name() return void Christian Couder
2016-04-24 13:33 ` [PATCH 02/83] builtin/apply: avoid parameter shadowing 'p_value' global Christian Couder
2016-04-24 13:33 ` [PATCH 03/83] builtin/apply: avoid parameter shadowing 'linenr' global Christian Couder
2016-04-27 16:27 ` Junio C Hamano
2016-04-27 17:07 ` Christian Couder
2016-04-24 13:33 ` [PATCH 04/83] builtin/apply: avoid local variable shadowing 'len' parameter Christian Couder
2016-04-24 13:33 ` Christian Couder [this message]
2016-04-25 18:50 ` [PATCH 05/83] builtin/apply: extract line_by_line_fuzzy_match() from match_fragment() Stefan Beller
2016-04-26 16:15 ` Christian Couder
2016-04-26 20:20 ` Junio C Hamano
2016-04-27 13:31 ` Christian Couder
2016-04-24 13:33 ` [PATCH 06/83] builtin/apply: move 'options' variable into cmd_apply() Christian Couder
2016-04-24 13:33 ` [PATCH 07/83] builtin/apply: introduce 'struct apply_state' to start libifying Christian Couder
2016-04-24 13:33 ` [PATCH 08/83] builtin/apply: move 'unidiff_zero' global into 'struct apply_state' Christian Couder
2016-04-24 13:33 ` [PATCH 09/83] builtin/apply: move 'check' " Christian Couder
2016-04-25 18:57 ` Stefan Beller
2016-04-26 16:26 ` Christian Couder
2016-04-26 16:31 ` Stefan Beller
2016-04-24 13:33 ` [PATCH 10/83] builtin/apply: move 'check_index' " Christian Couder
2016-04-26 20:25 ` Junio C Hamano
2016-04-24 13:33 ` [PATCH 11/83] builtin/apply: move 'apply_in_reverse' " Christian Couder
2016-04-24 13:33 ` [PATCH 12/83] builtin/apply: move 'apply_with_reject' " Christian Couder
2016-04-24 13:33 ` [PATCH 13/83] builtin/apply: move 'apply_verbosely' " Christian Couder
2016-04-24 13:33 ` [PATCH 14/83] builtin/apply: move 'update_index' " Christian Couder
2016-04-24 13:33 ` [PATCH 15/83] builtin/apply: move 'allow_overlap' " Christian Couder
2016-04-24 13:33 ` [PATCH 16/83] builtin/apply: move 'cached' " Christian Couder
2016-04-24 13:33 ` [PATCH 17/83] builtin/apply: move 'diffstat' " Christian Couder
2016-04-24 13:33 ` [PATCH 18/83] builtin/apply: move 'numstat' " Christian Couder
2016-04-25 21:40 ` Stefan Beller
2016-04-26 16:35 ` Christian Couder
2016-04-24 13:33 ` [PATCH 19/83] builtin/apply: move 'summary' " Christian Couder
2016-04-24 13:33 ` [PATCH 20/83] builtin/apply: move 'threeway' " Christian Couder
2016-04-24 13:33 ` [PATCH 21/83] builtin/apply: move 'no-add' " Christian Couder
2016-04-24 13:33 ` [PATCH 22/83] builtin/apply: move 'unsafe_paths' " Christian Couder
2016-04-26 20:27 ` Junio C Hamano
2016-04-27 12:14 ` Christian Couder
2016-04-27 15:26 ` Junio C Hamano
2016-04-24 13:33 ` [PATCH 23/83] builtin/apply: move 'line_termination' " Christian Couder
2016-04-24 13:33 ` [PATCH 24/83] builtin/apply: move 'fake_ancestor' " Christian Couder
2016-04-24 13:33 ` [PATCH 25/83] builtin/apply: move 'p_context' " Christian Couder
2016-04-24 13:33 ` [PATCH 26/83] builtin/apply: move 'apply' " Christian Couder
2016-04-24 13:33 ` [PATCH 27/83] builtin/apply: move 'read_stdin' global into cmd_apply() Christian Couder
2016-04-26 20:28 ` Junio C Hamano
2016-04-27 13:13 ` Christian Couder
2016-04-24 13:33 ` [PATCH 28/83] builtin/apply: move 'patch_input_file' global into 'struct apply_state' Christian Couder
2016-04-24 13:33 ` [PATCH 29/83] builtin/apply: move 'limit_by_name' " Christian Couder
2016-04-24 13:33 ` [PATCH 30/83] builtin/apply: move 'has_include' " Christian Couder
2016-04-24 13:33 ` [PATCH 31/83] builtin/apply: move 'p_value' " Christian Couder
2016-04-24 13:33 ` [PATCH 32/83] builtin/apply: move 'p_value_known' " Christian Couder
2016-04-24 13:33 ` [PATCH 33/83] builtin/apply: move 'root' " Christian Couder
2016-04-25 21:50 ` Stefan Beller
2016-04-25 21:54 ` Stefan Beller
2016-04-26 20:13 ` Christian Couder
2016-04-24 13:33 ` [PATCH 34/83] builtin/apply: move 'whitespace_error' " Christian Couder
2016-04-24 13:33 ` [PATCH 35/83] builtin/apply: move 'whitespace_option' " Christian Couder
2016-04-24 13:33 ` [PATCH 36/83] builtin/apply: remove whitespace_option arg from set_default_whitespace_mode() Christian Couder
2016-04-24 13:33 ` [PATCH 37/83] builtin/apply: move 'squelch_whitespace_errors' into 'struct apply_state' Christian Couder
2016-04-24 13:33 ` [PATCH 38/83] builtin/apply: move 'applied_after_fixing_ws' " Christian Couder
2016-04-24 13:33 ` [PATCH 39/83] builtin/apply: move 'ws_error_action' " Christian Couder
2016-04-26 20:36 ` Junio C Hamano
2016-04-26 20:40 ` Junio C Hamano
2016-04-27 13:23 ` Christian Couder
2016-04-24 13:33 ` [PATCH 40/83] builtin/apply: move 'ws_ignore_action' " Christian Couder
2016-04-24 13:33 ` [PATCH 41/83] builtin/apply: move 'max_change' and 'max_len' " Christian Couder
2016-04-24 13:33 ` [PATCH 42/83] builtin/apply: move 'linenr' global " Christian Couder
2016-04-24 13:33 ` [PATCH 43/83] builtin/apply: move 'fn_table' " Christian Couder
2016-04-24 13:33 ` [PATCH 44/83] builtin/apply: move 'symlink_changes' " Christian Couder
2016-04-24 13:33 ` [PATCH 45/83] builtin/apply: move 'state' init into init_apply_state() Christian Couder
2016-04-25 7:32 ` Eric Sunshine
2016-04-27 12:00 ` Christian Couder
2016-04-24 13:33 ` [PATCH 46/83] builtin/apply: move 'state' check into check_apply_state() Christian Couder
2016-04-24 13:33 ` [PATCH 47/83] builtin/apply: move applying patches into apply_all_patches() Christian Couder
2016-04-25 22:00 ` Stefan Beller
2016-04-27 15:51 ` Christian Couder
2016-04-24 13:33 ` [PATCH 48/83] builtin/apply: rename 'prefix_' parameter to 'prefix' Christian Couder
2016-04-24 13:33 ` [PATCH 49/83] builtin/apply: move 'lock_file' global into 'struct apply_state' Christian Couder
2016-04-25 7:50 ` Eric Sunshine
2016-04-25 17:55 ` Junio C Hamano
2016-04-28 16:30 ` Christian Couder
2016-04-28 20:17 ` Junio C Hamano
2016-04-30 19:39 ` Christian Couder
2016-04-24 13:33 ` [PATCH 50/83] builtin/apply: get rid of the 'newfd' global Christian Couder
2016-04-24 13:33 ` [PATCH 51/83] builtin/apply: make apply_patch() return -1 instead of die()ing Christian Couder
2016-04-26 1:20 ` Eric Sunshine
2016-04-30 19:41 ` Christian Couder
2016-04-24 13:33 ` [PATCH 52/83] builtin/apply: read_patch_file() " Christian Couder
2016-04-24 13:33 ` [PATCH 53/83] builtin/apply: make find_header() " Christian Couder
2016-04-25 13:18 ` Duy Nguyen
2016-04-27 18:10 ` Eric Sunshine
2016-05-01 16:53 ` Christian Couder
2016-04-27 18:08 ` Eric Sunshine
2016-05-01 16:45 ` Christian Couder
2016-04-24 13:33 ` [PATCH 54/83] builtin/apply: make parse_chunk() return a negative integer on error Christian Couder
2016-05-01 19:04 ` Eric Sunshine
2016-05-01 19:51 ` Christian Couder
2016-04-24 13:33 ` [PATCH 55/83] builtin/apply: make parse_single_patch() return -1 " Christian Couder
2016-05-01 19:32 ` Eric Sunshine
2016-04-24 13:33 ` [PATCH 56/83] apply: move 'struct apply_state' to apply.h Christian Couder
2016-04-24 13:33 ` [PATCH 57/83] builtin/apply: libify parse_whitespace_option() Christian Couder
2016-04-24 13:33 ` [PATCH 58/83] builtin/apply: libify parse_ignorewhitespace_option() Christian Couder
2016-04-24 13:33 ` [PATCH 59/83] builtin/apply: move init_apply_state() to apply.c Christian Couder
2016-05-01 19:37 ` Eric Sunshine
2016-05-01 20:19 ` Christian Couder
2016-04-24 13:34 ` [PATCH 60/83] apply: libify init_apply_state() Christian Couder
2016-04-24 13:34 ` [PATCH 61/83] builtin/apply: libify check_apply_state() Christian Couder
2016-04-25 13:26 ` Duy Nguyen
2016-05-01 16:58 ` Christian Couder
2016-04-24 13:34 ` [PATCH 62/83] builtin/apply: move check_apply_state() to apply.c Christian Couder
2016-04-24 13:34 ` [PATCH 63/83] builtin/apply: make apply_all_patches() return -1 on error Christian Couder
2016-04-25 13:30 ` Duy Nguyen
2016-05-01 18:32 ` Christian Couder
2016-05-01 21:03 ` Eric Sunshine
2016-05-02 7:09 ` Johannes Schindelin
2016-05-03 13:01 ` Christian Couder
2016-04-24 13:34 ` [PATCH 64/83] builtin/apply: make parse_traditional_patch() " Christian Couder
2016-04-24 13:34 ` [PATCH 65/83] builtin/apply: make gitdiff_verify_name() " Christian Couder
2016-04-25 13:36 ` Duy Nguyen
2016-05-01 19:31 ` Christian Couder
2016-04-24 13:34 ` [PATCH 66/83] builtin/apply: change die_on_unsafe_path() to check_unsafe_path() Christian Couder
2016-04-24 13:34 ` [PATCH 67/83] builtin/apply: make build_fake_ancestor() return -1 on error Christian Couder
2016-05-02 7:32 ` Eric Sunshine
2016-05-03 13:52 ` Christian Couder
2016-04-24 13:34 ` [PATCH 68/83] builtin/apply: make remove_file() " Christian Couder
2016-04-24 13:34 ` [PATCH 69/83] builtin/apply: make add_conflicted_stages_file() " Christian Couder
2016-05-02 7:36 ` Eric Sunshine
2016-05-03 14:33 ` Christian Couder
2016-04-24 13:34 ` [PATCH 70/83] builtin/apply: make add_index_file() " Christian Couder
2016-04-24 13:34 ` [PATCH 71/83] builtin/apply: make create_file() " Christian Couder
2016-04-24 13:34 ` [PATCH 72/83] builtin/apply: make write_out_one_result() " Christian Couder
2016-04-24 13:34 ` [PATCH 73/83] builtin/apply: make write_out_results() " Christian Couder
2016-05-02 17:42 ` Eric Sunshine
2016-05-03 14:55 ` Christian Couder
2016-04-24 13:34 ` [PATCH 74/83] builtin/apply: make try_create_file() " Christian Couder
2016-05-02 18:01 ` Eric Sunshine
2016-05-03 16:24 ` Christian Couder
2016-04-24 13:34 ` [PATCH 75/83] builtin/apply: make create_one_file() " Christian Couder
2016-04-24 13:34 ` [PATCH 76/83] builtin/apply: rename option parsing functions Christian Couder
2016-04-24 13:34 ` [PATCH 77/83] apply: rename and move opt constants to apply.h Christian Couder
2016-04-24 15:23 ` [PATCH 00/83] libify apply and use lib in am Ramsay Jones
2016-04-24 16:27 ` Christian Couder
2016-04-24 16:56 ` Christian Couder
2016-04-24 17:42 ` Ramsay Jones
2016-04-25 0:14 ` Duy Nguyen
2016-04-25 9:15 ` Christian Couder
2016-04-25 15:09 ` Johannes Schindelin
2016-04-25 9:02 ` Duy Nguyen
2016-04-25 9:57 ` Christian Couder
2016-04-25 10:11 ` Duy Nguyen
[not found] ` <1461504863-15946-79-git-send-email-chriscool@tuxfamily.org>
2016-04-25 13:46 ` [PATCH 78/83] Move libified code from builtin/apply.c to apply.{c,h} Duy Nguyen
2016-05-04 10:39 ` Christian Couder
2016-05-04 11:32 ` Duy Nguyen
2016-05-05 8:50 ` Christian Couder
2016-05-06 20:18 ` Christian Couder
2016-05-06 21:07 ` Junio C Hamano
2016-05-08 20:11 ` Christian Couder
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=1461504863-15946-6-git-send-email-chriscool@tuxfamily.org \
--to=christian.couder@gmail.com \
--cc=Johannes.Schindelin@gmx.de \
--cc=Matthieu.Moy@grenoble-inp.fr \
--cc=avarab@gmail.com \
--cc=chriscool@tuxfamily.org \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=karsten.blees@gmail.com \
--cc=pclouds@gmail.com \
--cc=peff@peff.net \
--cc=sbeller@google.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).