Merge branch 'jc/fix-tree-walk'

* jc/fix-tree-walk:
  read-tree --debug-unpack
  unpack-trees.c: look ahead in the index
  unpack-trees.c: prepare for looking ahead in the index
  Aggressive three-way merge: fix D/F case
  traverse_trees(): handle D/F conflict case sanely
  more D/F conflict tests
  tests: move convenience regexp to match object names to test-lib.sh

Conflicts:
	builtin-read-tree.c
	unpack-trees.c
	unpack-trees.h

builtin-read-tree.c

@@ -65,6 +65,34 @@ static int exclude_per_directory_cb(const struct option *opt, const char *arg,
 	return 0;
 }
 
+static void debug_stage(const char *label, struct cache_entry *ce,
+			struct unpack_trees_options *o)
+{
+	printf("%s ", label);
+	if (!ce)
+		printf("(missing)\n");
+	else if (ce == o->df_conflict_entry)
+		printf("(conflict)\n");
+	else
+		printf("%06o #%d %s %.8s\n",
+		       ce->ce_mode, ce_stage(ce), ce->name,
+		       sha1_to_hex(ce->sha1));
+}
+
+static int debug_merge(struct cache_entry **stages, struct unpack_trees_options *o)
+{
+	int i;
+
+	printf("* %d-way merge\n", o->merge_size);
+	debug_stage("index", stages[0], o);
+	for (i = 1; i <= o->merge_size; i++) {
+		char buf[24];
+		sprintf(buf, "ent#%d", i);
+		debug_stage(buf, stages[i], o);
+	}
+	return 0;
+}
+
 static struct lock_file lock_file;
 
 int cmd_read_tree(int argc, const char **argv, const char *unused_prefix)
@@ -101,6 +129,8 @@ int cmd_read_tree(int argc, const char **argv, const char *unused_prefix)
 			    "don't check the working tree after merging", 1),
 		OPT_SET_INT(0, "no-sparse-checkout", &opts.skip_sparse_checkout,
 			    "skip applying sparse checkout filter", 1),
+		OPT_SET_INT(0, "debug-unpack", &opts.debug_unpack,
+			    "debug unpack-trees", 1),
 		OPT_END()
 	};
 
@@ -169,6 +199,9 @@ int cmd_read_tree(int argc, const char **argv, const char *unused_prefix)
 			opts.head_idx = 1;
 	}
 
+	if (opts.debug_unpack)
+		opts.fn = debug_merge;
+
 	cache_tree_free(&active_cache_tree);
 	for (i = 0; i < nr_trees; i++) {
 		struct tree *tree = trees[i];
@@ -178,6 +211,9 @@ int cmd_read_tree(int argc, const char **argv, const char *unused_prefix)
 	if (unpack_trees(nr_trees, t, &opts))
 		return 128;
 
+	if (opts.debug_unpack)
+		return 0; /* do not write the index out */
+
 	/*
 	 * When reading only one tree (either the most basic form,
 	 * "-m ent" or "--reset ent" form), we can obtain a fully

cache.h

@@ -182,6 +182,8 @@ struct cache_entry {
 /* Only remove in work directory, not index */
 #define CE_WT_REMOVE (0x400000)
 
+#define CE_UNPACKED  (0x1000000)
+
 /*
  * Extended on-disk flags
  */

diff-lib.c

@@ -380,21 +380,6 @@ static void do_oneway_diff(struct unpack_trees_options *o,
 	show_modified(revs, tree, idx, 1, cached, match_missing);
 }
 
-static inline void skip_same_name(struct cache_entry *ce, struct unpack_trees_options *o)
-{
-	int len = ce_namelen(ce);
-	const struct index_state *index = o->src_index;
-
-	while (o->pos < index->cache_nr) {
-		struct cache_entry *next = index->cache[o->pos];
-		if (len != ce_namelen(next))
-			break;
-		if (memcmp(ce->name, next->name, len))
-			break;
-		o->pos++;
-	}
-}
-
 /*
  * The unpack_trees() interface is designed for merging, so
  * the different source entries are designed primarily for
@@ -416,9 +401,6 @@ static int oneway_diff(struct cache_entry **src, struct unpack_trees_options *o)
 	struct cache_entry *tree = src[1];
 	struct rev_info *revs = o->unpack_data;
 
-	if (idx && ce_stage(idx))
-		skip_same_name(idx, o);
-
 	/*
 	 * Unpack-trees generates a DF/conflict entry if
 	 * there was a directory in the index and a tree
@@ -464,6 +446,7 @@ int run_diff_index(struct rev_info *revs, int cached)
 		exit(128);
 
 	diff_set_mnemonic_prefix(&revs->diffopt, "c/", cached ? "i/" : "w/");
+	diffcore_fix_diff_index(&revs->diffopt);
 	diffcore_std(&revs->diffopt);
 	diff_flush(&revs->diffopt);
 	return 0;

diff.c

@@ -3678,6 +3678,23 @@ static void diffcore_skip_stat_unmatch(struct diff_options *diffopt)
 	*q = outq;
 }
 
+static int diffnamecmp(const void *a_, const void *b_)
+{
+	const struct diff_filepair *a = *((const struct diff_filepair **)a_);
+	const struct diff_filepair *b = *((const struct diff_filepair **)b_);
+	const char *name_a, *name_b;
+
+	name_a = a->one ? a->one->path : a->two->path;
+	name_b = b->one ? b->one->path : b->two->path;
+	return strcmp(name_a, name_b);
+}
+
+void diffcore_fix_diff_index(struct diff_options *options)
+{
+	struct diff_queue_struct *q = &diff_queued_diff;
+	qsort(q->queue, q->nr, sizeof(q->queue[0]), diffnamecmp);
+}
+
 void diffcore_std(struct diff_options *options)
 {
 	if (options->skip_stat_unmatch)

diff.h

@@ -210,6 +210,7 @@ extern int diff_setup_done(struct diff_options *);
 #define DIFF_PICKAXE_REGEX	2
 
 extern void diffcore_std(struct diff_options *);
+extern void diffcore_fix_diff_index(struct diff_options *);
 
 #define COMMON_DIFF_OPTIONS_HELP \
 "\ncommon diff options:\n" \

t/diff-lib.sh

@@ -1,7 +1,5 @@
 :
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 sanitize_diff_raw='/^:/s/ '"$_x40"' '"$_x40"' \([A-Z]\)[0-9]*	/ X X \1#	/'
 compare_diff_raw () {
     # When heuristics are improved, the score numbers would change.

t/t1000-read-tree-m-3way.sh

@@ -126,9 +126,6 @@ cat >expected <<\EOF
 100644 X 0	Z/NN
 EOF
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
-
 check_result () {
     git ls-files --stage | sed -e 's/ '"$_x40"' / X /' >current &&
     test_cmp expected current

t/t1001-read-tree-m-2way.sh

@@ -26,8 +26,6 @@ read_tree_twoway () {
     git read-tree -m "$1" "$2" && git ls-files --stage
 }
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 compare_change () {
 	sed -n >current \
 	    -e '/^--- /d; /^+++ /d; /^@@ /d;' \

t/t1002-read-tree-m-u-2way.sh

@@ -10,8 +10,6 @@ This is identical to t1001, but uses -u to update the work tree as well.
 '
 . ./test-lib.sh
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 compare_change () {
 	sed >current \
 	    -e '1{/^diff --git /d;}' \

t/t1012-read-tree-df.sh

@@ -0,0 +1,102 @@
+#!/bin/sh
+
+test_description='read-tree D/F conflict corner cases'
+
+. ./test-lib.sh
+
+maketree () {
+	(
+		rm -f .git/index .git/index.lock &&
+		git clean -d -f -f -q -x &&
+		name="$1" &&
+		shift &&
+		for it
+		do
+			path=$(expr "$it" : '\([^:]*\)') &&
+			mkdir -p $(dirname "$path") &&
+			echo "$it" >"$path" &&
+			git update-index --add "$path" || exit
+		done &&
+		git tag "$name" $(git write-tree)
+	)
+}
+
+settree () {
+	rm -f .git/index .git/index.lock &&
+	git clean -d -f -f -q -x &&
+	git read-tree "$1" &&
+	git checkout-index -f -q -u -a &&
+	git update-index --refresh
+}
+
+checkindex () {
+	git ls-files -s |
+	sed "s|^[0-7][0-7]* $_x40 \([0-3]\)	|\1 |" >current &&
+	cat >expect &&
+	test_cmp expect current
+}
+
+test_expect_success setup '
+	maketree O-000 a/b-2/c/d a/b/c/d a/x &&
+	maketree A-000 a/b-2/c/d a/b/c/d a/x &&
+	maketree A-001 a/b-2/c/d a/b/c/d a/b/c/e a/x &&
+	maketree B-000 a/b-2/c/d a/b     a/x &&
+
+	maketree O-010 t-0     t/1  t/2 t=3 &&
+	maketree A-010 t-0 t            t=3 &&
+	maketree B-010         t/1:     t=3: &&
+
+	maketree O-020 ds/dma/ioat.c ds/dma/ioat_dca.c &&
+	maketree A-020 ds/dma/ioat/Makefile ds/dma/ioat/registers.h &&
+	:
+'
+
+test_expect_success '3-way (1)' '
+	settree A-000 &&
+	git read-tree -m -u O-000 A-000 B-000 &&
+	checkindex <<-EOF
+	3 a/b
+	0 a/b-2/c/d
+	1 a/b/c/d
+	2 a/b/c/d
+	0 a/x
+	EOF
+'
+
+test_expect_success '3-way (2)' '
+	settree A-001 &&
+	git read-tree -m -u O-000 A-001 B-000 &&
+	checkindex <<-EOF
+	3 a/b
+	0 a/b-2/c/d
+	1 a/b/c/d
+	2 a/b/c/d
+	2 a/b/c/e
+	0 a/x
+	EOF
+'
+
+test_expect_success '3-way (3)' '
+	settree A-010 &&
+	git read-tree -m -u O-010 A-010 B-010 &&
+	checkindex <<-EOF
+	2 t
+	1 t-0
+	2 t-0
+	1 t/1
+	3 t/1
+	1 t/2
+	0 t=3
+	EOF
+'
+
+test_expect_success '2-way (1)' '
+	settree O-020 &&
+	git read-tree -m -u O-020 A-020 &&
+	checkindex <<-EOF
+	0 ds/dma/ioat/Makefile
+	0 ds/dma/ioat/registers.h
+	EOF
+'
+
+test_done

t/t3100-ls-tree-restrict.sh

@@ -43,8 +43,6 @@ test_expect_success \
      tree=`git write-tree` &&
      echo $tree'
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 test_output () {
     sed -e "s/ $_x40	/ X	/" <current >check
     test_cmp expected check

t/t3101-ls-tree-dirname.sh

@@ -39,8 +39,6 @@ test_expect_success \
      tree=`git write-tree` &&
      echo $tree'
 
-_x05='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x05$_x05$_x05$_x05$_x05$_x05$_x05$_x05"
 test_output () {
     sed -e "s/ $_x40	/ X	/" <current >check
     test_cmp expected check

t/t4006-diff-mode.sh

@@ -20,8 +20,6 @@ test_expect_success \
     'test_chmod +x rezrov &&
      git diff-index $tree >current'
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 sed -e 's/\(:100644 100755\) \('"$_x40"'\) \2 /\1 X X /' <current >check
 echo ":100644 100755 X X M	rezrov" >expected
 

t/t6012-rev-list-simplify.sh

@@ -8,9 +8,6 @@ note () {
 	git tag "$1"
 }
 
-_x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
-_x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
-
 unnote () {
 	git name-rev --tags --stdin | sed -e "s|$_x40 (tags/\([^)]*\)) |\1 |g"
 }

t/t6035-merge-dir-to-symlink.sh

@@ -48,7 +48,7 @@ test_expect_success 'setup for merge test' '
 	git tag baseline
 '
 
-test_expect_success 'do not lose a/b-2/c/d in merge (resolve)' '
+test_expect_failure 'do not lose a/b-2/c/d in merge (resolve)' '
 	git reset --hard &&
 	git checkout baseline^0 &&
 	git merge -s resolve master &&
@@ -74,7 +74,7 @@ test_expect_success 'setup a merge where dir a/b-2 changed to symlink' '
 	git tag test2
 '
 
-test_expect_failure 'merge should not have conflicts (resolve)' '
+test_expect_success 'merge should not have conflicts (resolve)' '
 	git reset --hard &&
 	git checkout baseline^0 &&
 	git merge -s resolve test2 &&

t/test-lib.sh

@@ -74,6 +74,12 @@ case $(echo $GIT_TRACE |tr "[A-Z]" "[a-z]") in
 		;;
 esac
 
+# Convenience
+#
+# A regexp to match 5 and 40 hexdigits
+_x05='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
+_x40="$_x05$_x05$_x05$_x05$_x05$_x05$_x05$_x05"
+
 # Each test should start with something like this, after copyright notices:
 #
 # test_description='Description of this test...