From: Jeff Smith <whydoubt@gmail.com>
To: git@vger.kernel.org
Cc: gitster@pobox.com, peff@peff.net, Jeff Smith <whydoubt@gmail.com>
Subject: [RFC PATCH 05/10] Split blame origin into its own file
Date: Fri, 5 May 2017 00:27:24 -0500 [thread overview]
Message-ID: <20170505052729.7576-6-whydoubt@gmail.com> (raw)
In-Reply-To: <20170505052729.7576-1-whydoubt@gmail.com>
Signed-off-by: Jeff Smith <whydoubt@gmail.com>
---
Makefile | 1 +
builtin/blame.c | 151 +-------------------------------------------------------
origin.c | 62 +++++++++++++++++++++++
origin.h | 101 +++++++++++++++++++++++++++++++++++++
4 files changed, 165 insertions(+), 150 deletions(-)
create mode 100644 origin.c
create mode 100644 origin.h
diff --git a/Makefile b/Makefile
index e35542e..8cbb56c 100644
--- a/Makefile
+++ b/Makefile
@@ -791,6 +791,7 @@ LIB_OBJS += notes-merge.o
LIB_OBJS += notes-utils.o
LIB_OBJS += object.o
LIB_OBJS += oidset.o
+LIB_OBJS += origin.o
LIB_OBJS += pack-bitmap.o
LIB_OBJS += pack-bitmap-write.o
LIB_OBJS += pack-check.o
diff --git a/builtin/blame.c b/builtin/blame.c
index cc46f56..7ee84c1 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -28,6 +28,7 @@
#include "line-log.h"
#include "dir.h"
#include "progress.h"
+#include "origin.h"
static char blame_usage[] = N_("git blame [<options>] [<rev-opts>] [<rev>] [--] <file>");
@@ -84,50 +85,6 @@ static unsigned blame_copy_score;
#define METAINFO_SHOWN (1u<<12)
#define MORE_THAN_ONE_PATH (1u<<13)
-/*
- * One blob in a commit that is being suspected
- */
-struct origin {
- int refcnt;
- /* Record preceding blame record for this blob */
- struct origin *previous;
- /* origins are put in a list linked via `next' hanging off the
- * corresponding commit's util field in order to make finding
- * them fast. The presence in this chain does not count
- * towards the origin's reference count. It is tempting to
- * let it count as long as the commit is pending examination,
- * but even under circumstances where the commit will be
- * present multiple times in the priority queue of unexamined
- * commits, processing the first instance will not leave any
- * work requiring the origin data for the second instance. An
- * interspersed commit changing that would have to be
- * preexisting with a different ancestry and with the same
- * commit date in order to wedge itself between two instances
- * of the same commit in the priority queue _and_ produce
- * blame entries relevant for it. While we don't want to let
- * us get tripped up by this case, it certainly does not seem
- * worth optimizing for.
- */
- struct origin *next;
- struct commit *commit;
- /* `suspects' contains blame entries that may be attributed to
- * this origin's commit or to parent commits. When a commit
- * is being processed, all suspects will be moved, either by
- * assigning them to an origin in a different commit, or by
- * shipping them to the scoreboard's ent list because they
- * cannot be attributed to a different commit.
- */
- struct blame_entry *suspects;
- mmfile_t file;
- struct object_id blob_oid;
- unsigned mode;
- /* guilty gets set when shipping any suspects to the final
- * blame list instead of other commits
- */
- char guilty;
- char path[FLEX_ARRAY];
-};
-
struct progress_info {
struct progress *progress;
int blamed_lines;
@@ -176,39 +133,6 @@ static void fill_origin_blob(struct diff_options *opt,
*file = o->file;
}
-/*
- * Origin is refcounted and usually we keep the blob contents to be
- * reused.
- */
-static inline struct origin *origin_incref(struct origin *o)
-{
- if (o)
- o->refcnt++;
- return o;
-}
-
-static void origin_decref(struct origin *o)
-{
- if (o && --o->refcnt <= 0) {
- struct origin *p, *l = NULL;
- if (o->previous)
- origin_decref(o->previous);
- free(o->file.ptr);
- /* Should be present exactly once in commit chain */
- for (p = o->commit->util; p; l = p, p = p->next) {
- if (p == o) {
- if (l)
- l->next = p->next;
- else
- o->commit->util = p->next;
- free(o);
- return;
- }
- }
- die("internal error in blame::origin_decref");
- }
-}
-
static void drop_origin_blob(struct origin *o)
{
if (o->file.ptr) {
@@ -218,40 +142,6 @@ static void drop_origin_blob(struct origin *o)
}
/*
- * Each group of lines is described by a blame_entry; it can be split
- * as we pass blame to the parents. They are arranged in linked lists
- * kept as `suspects' of some unprocessed origin, or entered (when the
- * blame origin has been finalized) into the scoreboard structure.
- * While the scoreboard structure is only sorted at the end of
- * processing (according to final image line number), the lists
- * attached to an origin are sorted by the target line number.
- */
-struct blame_entry {
- struct blame_entry *next;
-
- /* the first line of this group in the final image;
- * internally all line numbers are 0 based.
- */
- int lno;
-
- /* how many lines this group has */
- int num_lines;
-
- /* the commit that introduced this group into the final image */
- struct origin *suspect;
-
- /* the line number of the first line of this group in the
- * suspect's file; internally all line numbers are 0 based.
- */
- int s_lno;
-
- /* how significant this entry is -- cached to avoid
- * scanning the lines over and over.
- */
- unsigned score;
-};
-
-/*
* Any merge of blames happens on lists of blames that arrived via
* different parents in a single suspect. In this case, we want to
* sort according to the suspect line numbers as opposed to the final
@@ -429,45 +319,6 @@ static void queue_blames(struct scoreboard *sb, struct origin *porigin,
}
/*
- * Given a commit and a path in it, create a new origin structure.
- * The callers that add blame to the scoreboard should use
- * get_origin() to obtain shared, refcounted copy instead of calling
- * this function directly.
- */
-static struct origin *make_origin(struct commit *commit, const char *path)
-{
- struct origin *o;
- FLEX_ALLOC_STR(o, path, path);
- o->commit = commit;
- o->refcnt = 1;
- o->next = commit->util;
- commit->util = o;
- return o;
-}
-
-/*
- * Locate an existing origin or create a new one.
- * This moves the origin to front position in the commit util list.
- */
-static struct origin *get_origin(struct commit *commit, const char *path)
-{
- struct origin *o, *l;
-
- for (o = commit->util, l = NULL; o; l = o, o = o->next) {
- if (!strcmp(o->path, path)) {
- /* bump to front */
- if (l) {
- l->next = o->next;
- o->next = commit->util;
- commit->util = o;
- }
- return origin_incref(o);
- }
- }
- return make_origin(commit, path);
-}
-
-/*
* Fill the blob_sha1 field of an origin if it hasn't, so that later
* call to fill_origin_blob() can use it to locate the data. blob_sha1
* for an origin is also used to pass the blame for the entire file to
diff --git a/origin.c b/origin.c
new file mode 100644
index 0000000..2ec25c7
--- /dev/null
+++ b/origin.c
@@ -0,0 +1,62 @@
+#include "origin.h"
+
+void origin_decref(struct origin *o)
+{
+ if (o && --o->refcnt <= 0) {
+ struct origin *p, *l = NULL;
+ if (o->previous)
+ origin_decref(o->previous);
+ free(o->file.ptr);
+ /* Should be present exactly once in commit chain */
+ for (p = o->commit->util; p; l = p, p = p->next) {
+ if (p == o) {
+ if (l)
+ l->next = p->next;
+ else
+ o->commit->util = p->next;
+ free(o);
+ return;
+ }
+ }
+ die("internal error in blame::origin_decref");
+ }
+}
+
+/*
+ * Given a commit and a path in it, create a new origin structure.
+ * The callers that add blame to the scoreboard should use
+ * get_origin() to obtain shared, refcounted copy instead of calling
+ * this function directly.
+ */
+struct origin *make_origin(struct commit *commit, const char *path)
+{
+ struct origin *o;
+ FLEX_ALLOC_STR(o, path, path);
+ o->commit = commit;
+ o->refcnt = 1;
+ o->next = commit->util;
+ commit->util = o;
+ return o;
+}
+
+/*
+ * Locate an existing origin or create a new one.
+ * This moves the origin to front position in the commit util list.
+ */
+struct origin *get_origin(struct commit *commit, const char *path)
+{
+ struct origin *o, *l;
+
+ for (o = commit->util, l = NULL; o; l = o, o = o->next) {
+ if (!strcmp(o->path, path)) {
+ /* bump to front */
+ if (l) {
+ l->next = o->next;
+ o->next = commit->util;
+ commit->util = o;
+ }
+ return origin_incref(o);
+ }
+ }
+ return make_origin(commit, path);
+}
diff --git a/origin.h b/origin.h
new file mode 100644
index 0000000..a189b78
--- /dev/null
+++ b/origin.h
@@ -0,0 +1,101 @@
+#ifndef ORIGIN_H
+#define ORIGIN_H
+
+#include "cache.h"
+#include "commit.h"
+#include "xdiff-interface.h"
+
+/*
+ * One blob in a commit that is being suspected
+ */
+struct origin {
+ int refcnt;
+ /* Record preceding blame record for this blob */
+ struct origin *previous;
+ /* origins are put in a list linked via `next' hanging off the
+ * corresponding commit's util field in order to make finding
+ * them fast. The presence in this chain does not count
+ * towards the origin's reference count. It is tempting to
+ * let it count as long as the commit is pending examination,
+ * but even under circumstances where the commit will be
+ * present multiple times in the priority queue of unexamined
+ * commits, processing the first instance will not leave any
+ * work requiring the origin data for the second instance. An
+ * interspersed commit changing that would have to be
+ * preexisting with a different ancestry and with the same
+ * commit date in order to wedge itself between two instances
+ * of the same commit in the priority queue _and_ produce
+ * blame entries relevant for it. While we don't want to let
+ * us get tripped up by this case, it certainly does not seem
+ * worth optimizing for.
+ */
+ struct origin *next;
+ struct commit *commit;
+ /* `suspects' contains blame entries that may be attributed to
+ * this origin's commit or to parent commits. When a commit
+ * is being processed, all suspects will be moved, either by
+ * assigning them to an origin in a different commit, or by
+ * shipping them to the scoreboard's ent list because they
+ * cannot be attributed to a different commit.
+ */
+ struct blame_entry *suspects;
+ mmfile_t file;
+ struct object_id blob_oid;
+ unsigned mode;
+ /* guilty gets set when shipping any suspects to the final
+ * blame list instead of other commits
+ */
+ char guilty;
+ char path[FLEX_ARRAY];
+};
+
+/*
+ * Each group of lines is described by a blame_entry; it can be split
+ * as we pass blame to the parents. They are arranged in linked lists
+ * kept as `suspects' of some unprocessed origin, or entered (when the
+ * blame origin has been finalized) into the scoreboard structure.
+ * While the scoreboard structure is only sorted at the end of
+ * processing (according to final image line number), the lists
+ * attached to an origin are sorted by the target line number.
+ */
+struct blame_entry {
+ struct blame_entry *next;
+
+ /* the first line of this group in the final image;
+ * internally all line numbers are 0 based.
+ */
+ int lno;
+
+ /* how many lines this group has */
+ int num_lines;
+
+ /* the commit that introduced this group into the final image */
+ struct origin *suspect;
+
+ /* the line number of the first line of this group in the
+ * suspect's file; internally all line numbers are 0 based.
+ */
+ int s_lno;
+
+ /* how significant this entry is -- cached to avoid
+ * scanning the lines over and over.
+ */
+ unsigned score;
+};
+
+/*
+ * Origin is refcounted and usually we keep the blob contents to be
+ * reused.
+ */
+static inline struct origin *origin_incref(struct origin *o)
+{
+ if (o)
+ o->refcnt++;
+ return o;
+}
+void origin_decref(struct origin *o);
+
+struct origin *make_origin(struct commit *commit, const char *path);
+struct origin *get_origin(struct commit *commit, const char *path);
+
+#endif /* ORIGIN_H */
--
2.9.3
next prev parent reply other threads:[~2017-05-05 5:27 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-05 5:27 [RFC PATCH 00/10] Add blame to libgit Jeff Smith
2017-05-05 5:27 ` [RFC PATCH 01/10] Remove unneeded dependency on blob.h from blame Jeff Smith
2017-05-05 7:07 ` Ævar Arnfjörð Bjarmason
[not found] ` <CAPX7N=6tQi+WidagvV1BA-CoaiNJj7OO4U7GYXNE-QzyvD=QRQ@mail.gmail.com>
2017-05-05 14:03 ` Jeffrey Smith
2017-05-05 5:27 ` [RFC PATCH 02/10] Move textconv_object to be with other textconv methods Jeff Smith
2017-05-05 17:44 ` Junio C Hamano
2017-05-05 17:50 ` Stefan Beller
2017-05-08 1:02 ` Junio C Hamano
2017-05-08 21:55 ` Jeff King
2017-05-08 22:06 ` Stefan Beller
2017-05-09 1:49 ` Junio C Hamano
2017-05-09 2:34 ` Jeff King
2017-05-05 5:27 ` [RFC PATCH 03/10] Add some missing definitions to header files Jeff Smith
2017-05-05 17:42 ` Junio C Hamano
2017-05-05 5:27 ` [RFC PATCH 04/10] Remove unused parameter from get_origin() Jeff Smith
2017-05-05 5:27 ` Jeff Smith [this message]
2017-05-05 5:27 ` [RFC PATCH 06/10] Move fake_working_tree_commit() to lib Jeff Smith
2017-05-05 5:27 ` [RFC PATCH 07/10] Break out scoreboard a little better Jeff Smith
2017-05-05 17:52 ` Junio C Hamano
2017-05-05 5:27 ` [RFC PATCH 08/10] Split blame scoreboard into its own file Jeff Smith
2017-05-05 5:27 ` [RFC PATCH 09/10] Break out scoreboard init and setup Jeff Smith
2017-05-05 5:27 ` [RFC PATCH 10/10] Move scoreboard init and setup to lib Jeff Smith
2017-05-05 17:54 ` [RFC PATCH 00/10] Add blame to libgit Junio C Hamano
2017-05-14 3:14 ` [RFC PATCH v2 00/22] " Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 01/22] blame: remove unneeded dependency on blob.h Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 02/22] blame: move textconv_object with related functions Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 03/22] blame: remove unused parameters Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 04/22] blame: move origin and entry structures to header Jeff Smith
2017-05-14 8:10 ` Junio C Hamano
2017-05-14 3:14 ` [RFC PATCH v2 05/22] blame: move scoreboard structure " Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 06/22] blame: move stat counters to scoreboard Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 07/22] blame: move copy/move thresholds " Jeff Smith
2017-05-14 3:14 ` [RFC PATCH v2 08/22] blame: move contents_from " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 09/22] blame: move reverse flag " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 10/22] blame: move show_root " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 11/22] blame: move xdl_opts flags " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 12/22] blame: move no_whole_file_rename flag " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 13/22] blame: make sanity_check use a callback in scoreboard Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 14/22] blame: move progess updates to a scoreboard callback Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 15/22] blame: wrap blame_sort and compare_blame_final Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 16/22] blame: rework methods that determine 'final' commit Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 17/22] blame: move origin-related methods to libgit Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 18/22] blame: move fake-commit-related " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 19/22] blame: move scoreboard-related " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 20/22] blame: create scoreboard init function in libgit Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 21/22] blame: create scoreboard setup " Jeff Smith
2017-05-14 3:15 ` [RFC PATCH v2 22/22] blame: create entry prepend " Jeff Smith
2017-05-15 9:24 ` [RFC PATCH v2 00/22] Add blame to libgit Junio C Hamano
2017-05-15 13:52 ` Jeffrey Smith
2017-05-16 0:23 ` Junio C Hamano
2017-05-16 2:44 ` Jeffrey Smith
2017-05-16 3:21 ` Junio C Hamano
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170505052729.7576-6-whydoubt@gmail.com \
--to=whydoubt@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=peff@peff.net \
/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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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 read-only IMAP folder(s) and NNTP newsgroup(s).