blame: add the ability to ignore commits and their changes

Commits that make formatting changes or function renames are often not
interesting when blaming a file.  A user may deem such a commit as 'not
interesting' and want to ignore and its changes it when assigning blame.

For example, say a file has the following git history / rev-list:

---O---A---X---B---C---D---Y---E---F

Commits X and Y both touch a particular line, and the other commits do
not:

X: "Take a third parameter"
-MyFunc(1, 2);
+MyFunc(1, 2, 3);

Y: "Remove camelcase"
-MyFunc(1, 2, 3);
+my_func(1, 2, 3);

git-blame will blame Y for the change.  I'd like to be able to ignore Y:
both the existence of the commit as well as any changes it made.  This
differs from -S rev-list, which specifies the list of commits to
process for the blame.  We would still process Y, but just don't let the
blame 'stick.'

This patch adds the ability for users to ignore a revision with
--ignore-rev=rev, which may be repeated.  They can specify a set of
files of full object names of revs, e.g. SHA-1 hashes, one per line.  A
single file may be specified with the blame.ignoreRevFile config option
or with --ignore-rev-file=file.  Both the config option and the command
line option may be repeated multiple times.  An empty file name "" will
clear the list of revs from previously processed files.  Config options
are processed before command line options.

For a typical use case, projects will maintain the file containing
revisions for commits that perform mass reformatting, and their users
have the option to ignore all of the commits in that file.

Additionally, a user can use the --ignore-rev option for one-off
investigation.  To go back to the example above, X was a substantive
change to the function, but not the change the user is interested in.
The user inspected X, but wanted to find the previous change to that
line - perhaps a commit that introduced that function call.

To make this work, we can't simply remove all ignored commits from the
rev-list.  We need to diff the changes introduced by Y so that we can
ignore them.  We let the blames get passed to Y, just like when
processing normally.  When Y is the target, we make sure that Y does not
*keep* any blames.  Any changes that Y is responsible for get passed to
its parent.  Note we make one pass through all of the scapegoats
(parents) to attempt to pass blame normally; we don't know if we *need*
to ignore the commit until we've checked all of the parents.

The blame_entry will get passed up the tree until we find a commit that
has a diff chunk that affects those lines.

One issue is that the ignored commit *did* make some change, and there is
no general solution to finding the line in the parent commit that
corresponds to a given line in the ignored commit.  That makes it hard
to attribute a particular line within an ignored commit's diff
correctly.

For example, the parent of an ignored commit has this, say at line 11:

commit-a 11) #include "a.h"
commit-b 12) #include "b.h"

Commit X, which we will ignore, swaps these lines:

commit-X 11) #include "b.h"
commit-X 12) #include "a.h"

We can pass that blame entry to the parent, but line 11 will be
attributed to commit A, even though "include b.h" came from commit B.
The blame mechanism will be looking at the parent's view of the file at
line number 11.

ignore_blame_entry() is set up to allow alternative algorithms for
guessing per-line blames.  Any line that is not attributed to the parent
will continue to be blamed on the ignored commit as if that commit was
not ignored.  Upcoming patches have the ability to detect these lines
and mark them in the blame output.

The existing algorithm is simple: blame each line on the corresponding
line in the parent's diff chunk.  Any lines beyond that stay with the
target.

For example, the parent of an ignored commit has this, say at line 11:

commit-a 11) void new_func_1(void *x, void *y);
commit-b 12) void new_func_2(void *x, void *y);
commit-c 13) some_line_c
commit-d 14) some_line_d

After a commit 'X', we have:

commit-X 11) void new_func_1(void *x,
commit-X 12)                 void *y);
commit-X 13) void new_func_2(void *x,
commit-X 14)                 void *y);
commit-c 15) some_line_c
commit-d 16) some_line_d

Commit X nets two additionally lines: 13 and 14.  The current
guess_line_blames() algorithm will not attribute these to the parent,
whose diff chunk is only two lines - not four.

When we ignore with the current algorithm, we get:

commit-a 11) void new_func_1(void *x,
commit-b 12)                 void *y);
commit-X 13) void new_func_2(void *x,
commit-X 14)                 void *y);
commit-c 15) some_line_c
commit-d 16) some_line_d

Note that line 12 was blamed on B, though B was the commit for
new_func_2(), not new_func_1().  Even when guess_line_blames() finds a
line in the parent, it may still be incorrect.

Signed-off-by: Barret Rhoden <brho@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

1 files changed, 167 insertions, 9 deletions

diff --git a/blame.c b/blame.c
index b6f8ba7..c57d7a3 100644
--- a/blame.c
+++ b/blame.c
@@ -859,6 +859,103 @@ static struct blame_entry *split_blame_at(struct blame_entry *e, int len,
 	return n;
 }
 
+struct blame_line_tracker {
+	int is_parent;
+	int s_lno;
+};
+
+static int are_lines_adjacent(struct blame_line_tracker *first,
+			      struct blame_line_tracker *second)
+{
+	return first->is_parent == second->is_parent &&
+	       first->s_lno + 1 == second->s_lno;
+}
+
+/*
+ * This cheap heuristic assigns lines in the chunk to their relative location in
+ * the parent's chunk.  Any additional lines are left with the target.
+ */
+static void guess_line_blames(struct blame_origin *parent,
+			      struct blame_origin *target,
+			      int tlno, int offset, int same, int parent_len,
+			      struct blame_line_tracker *line_blames)
+{
+	int i, best_idx, target_idx;
+	int parent_slno = tlno + offset;
+
+	for (i = 0; i < same - tlno; i++) {
+		target_idx = tlno + i;
+		best_idx = target_idx + offset;
+		if (best_idx < parent_slno + parent_len) {
+			line_blames[i].is_parent = 1;
+			line_blames[i].s_lno = best_idx;
+		} else {
+			line_blames[i].is_parent = 0;
+			line_blames[i].s_lno = target_idx;
+		}
+	}
+}
+
+/*
+ * This decides which parts of a blame entry go to the parent (added to the
+ * ignoredp list) and which stay with the target (added to the diffp list).  The
+ * actual decision was made in a separate heuristic function, and those answers
+ * for the lines in 'e' are in line_blames.  This consumes e, essentially
+ * putting it on a list.
+ *
+ * Note that the blame entries on the ignoredp list are not necessarily sorted
+ * with respect to the parent's line numbers yet.
+ */
+static void ignore_blame_entry(struct blame_entry *e,
+			       struct blame_origin *parent,
+			       struct blame_origin *target,
+			       struct blame_entry **diffp,
+			       struct blame_entry **ignoredp,
+			       struct blame_line_tracker *line_blames)
+{
+	int entry_len, nr_lines, i;
+
+	/*
+	 * We carve new entries off the front of e.  Each entry comes from a
+	 * contiguous chunk of lines: adjacent lines from the same origin
+	 * (either the parent or the target).
+	 */
+	entry_len = 1;
+	nr_lines = e->num_lines;	/* e changes in the loop */
+	for (i = 0; i < nr_lines; i++) {
+		struct blame_entry *next = NULL;
+
+		/*
+		 * We are often adjacent to the next line - only split the blame
+		 * entry when we have to.
+		 */
+		if (i + 1 < nr_lines) {
+			if (are_lines_adjacent(&line_blames[i],
+					       &line_blames[i + 1])) {
+				entry_len++;
+				continue;
+			}
+			next = split_blame_at(e, entry_len,
+					      blame_origin_incref(e->suspect));
+		}
+		if (line_blames[i].is_parent) {
+			blame_origin_decref(e->suspect);
+			e->suspect = blame_origin_incref(parent);
+			e->s_lno = line_blames[i - entry_len + 1].s_lno;
+			e->next = *ignoredp;
+			*ignoredp = e;
+		} else {
+			/* e->s_lno is already in the target's address space. */
+			e->next = *diffp;
+			*diffp = e;
+		}
+		assert(e->num_lines == entry_len);
+		e = next;
+		entry_len = 1;
+	}
+	assert(!e);
+}
+
 /*
  * Process one hunk from the patch between the current suspect for
  * blame_entry e and its parent.  This first blames any unfinished
@@ -868,13 +965,20 @@ static struct blame_entry *split_blame_at(struct blame_entry *e, int len,
  * -C options may lead to overlapping/duplicate source line number
  * ranges, all we can rely on from sorting/merging is the order of the
  * first suspect line number.
+ *
+ * tlno: line number in the target where this chunk begins
+ * same: line number in the target where this chunk ends
+ * offset: add to tlno to get the chunk starting point in the parent
+ * parent_len: number of lines in the parent chunk
  */
 static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq,
-			int tlno, int offset, int same,
-			struct blame_origin *parent)
+			int tlno, int offset, int same, int parent_len,
+			struct blame_origin *parent,
+			struct blame_origin *target, int ignore_diffs)
 {
 	struct blame_entry *e = **srcq;
-	struct blame_entry *samep = NULL, *diffp = NULL;
+	struct blame_entry *samep = NULL, *diffp = NULL, *ignoredp = NULL;
+	struct blame_line_tracker *line_blames = NULL;
 
 	while (e && e->s_lno < tlno) {
 		struct blame_entry *next = e->next;
@@ -923,6 +1027,14 @@ static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq,
 	 */
 	samep = NULL;
 	diffp = NULL;
+
+	if (ignore_diffs && same - tlno > 0) {
+		line_blames = xcalloc(sizeof(struct blame_line_tracker),
+				      same - tlno);
+		guess_line_blames(parent, target, tlno, offset, same,
+				  parent_len, line_blames);
+	}
+
 	while (e && e->s_lno < same) {
 		struct blame_entry *next = e->next;
 
@@ -942,10 +1054,29 @@ static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq,
 			n->next = samep;
 			samep = n;
 		}
-		e->next = diffp;
-		diffp = e;
+		if (ignore_diffs) {
+			ignore_blame_entry(e, parent, target, &diffp, &ignoredp,
+					   line_blames + e->s_lno - tlno);
+		} else {
+			e->next = diffp;
+			diffp = e;
+		}
 		e = next;
 	}
+	free(line_blames);
+	if (ignoredp) {
+		/*
+		 * Note ignoredp is not sorted yet, and thus neither is dstq.
+		 * That list must be sorted before we queue_blames().  We defer
+		 * sorting until after all diff hunks are processed, so that
+		 * guess_line_blames() can pick *any* line in the parent.  The
+		 * slight drawback is that we end up sorting all blame entries
+		 * passed to the parent, including those that are unrelated to
+		 * changes made by the ignored commit.
+		 */
+		**dstq = reverse_blame(ignoredp, **dstq);
+		*dstq = &ignoredp->next;
+	}
 	**srcq = reverse_blame(diffp, reverse_blame(samep, e));
 	/* Move across elements that are in the unblamable portion */
 	if (diffp)
@@ -954,7 +1085,9 @@ static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq,
 
 struct blame_chunk_cb_data {
 	struct blame_origin *parent;
+	struct blame_origin *target;
 	long offset;
+	int ignore_diffs;
 	struct blame_entry **dstq;
 	struct blame_entry **srcq;
 };
@@ -967,7 +1100,8 @@ static int blame_chunk_cb(long start_a, long count_a,
 	if (start_a - start_b != d->offset)
 		die("internal error in blame::blame_chunk_cb");
 	blame_chunk(&d->dstq, &d->srcq, start_b, start_a - start_b,
-		    start_b + count_b, d->parent);
+		    start_b + count_b, count_a, d->parent, d->target,
+		    d->ignore_diffs);
 	d->offset = start_a + count_a - (start_b + count_b);
 	return 0;
 }
@@ -979,7 +1113,7 @@ static int blame_chunk_cb(long start_a, long count_a,
  */
 static void pass_blame_to_parent(struct blame_scoreboard *sb,
 				 struct blame_origin *target,
-				 struct blame_origin *parent)
+				 struct blame_origin *parent, int ignore_diffs)
 {
 	mmfile_t file_p, file_o;
 	struct blame_chunk_cb_data d;
@@ -989,7 +1123,9 @@ static void pass_blame_to_parent(struct blame_scoreboard *sb,
 		return; /* nothing remains for this target */
 
 	d.parent = parent;
+	d.target = target;
 	d.offset = 0;
+	d.ignore_diffs = ignore_diffs;
 	d.dstq = &newdest; d.srcq = &target->suspects;
 
 	fill_origin_blob(&sb->revs->diffopt, parent, &file_p, &sb->num_read_blob);
@@ -1001,8 +1137,13 @@ static void pass_blame_to_parent(struct blame_scoreboard *sb,
 		    oid_to_hex(&parent->commit->object.oid),
 		    oid_to_hex(&target->commit->object.oid));
 	/* The rest are the same as the parent */
-	blame_chunk(&d.dstq, &d.srcq, INT_MAX, d.offset, INT_MAX, parent);
+	blame_chunk(&d.dstq, &d.srcq, INT_MAX, d.offset, INT_MAX, 0,
+		    parent, target, 0);
 	*d.dstq = NULL;
+	if (ignore_diffs)
+		newdest = llist_mergesort(newdest, get_next_blame,
+					  set_next_blame,
+					  compare_blame_suspect);
 	queue_blames(sb, parent, newdest);
 
 	return;
@@ -1506,12 +1647,29 @@ static void pass_blame(struct blame_scoreboard *sb, struct blame_origin *origin,
 			blame_origin_incref(porigin);
 			origin->previous = porigin;
 		}
-		pass_blame_to_parent(sb, origin, porigin);
+		pass_blame_to_parent(sb, origin, porigin, 0);
 		if (!origin->suspects)
 			goto finish;
 	}
 
 	/*
+	 * Pass remaining suspects for ignored commits to their parents.
+	 */
+	if (oidset_contains(&sb->ignore_list, &commit->object.oid)) {
+		for (i = 0, sg = first_scapegoat(revs, commit, sb->reverse);
+		     i < num_sg && sg;
+		     sg = sg->next, i++) {
+			struct blame_origin *porigin = sg_origin[i];
+
+			if (!porigin)
+				continue;
+			pass_blame_to_parent(sb, origin, porigin, 1);
+			if (!origin->suspects)
+				goto finish;
+		}
+	}
+
+	/*
 	 * Optionally find moves in parents' files.
 	 */
 	if (opt & PICKAXE_BLAME_MOVE) {