Add diff rename tests with partial similarity

This adds some new tests that actually exercise the similarity
metric between files to detect renames, copies, and split modified
files that are too heavily modified.

There is still more testing to do - these tests are just partially
covering the cases.

There is also one bug fix in this where a change set with only
MODIFY being broken into ADD/DELETE (due to low self-similarity)
without any additional RENAMED entries would end up not processing
the split requests (because the num_rewrites counter got reset).

1 files changed, 115 insertions, 2 deletions

diff --git a/tests-clar/diff/rename.c b/tests-clar/diff/rename.c
index 2995b4ef5..ec99e4dbe 100644
--- a/tests-clar/diff/rename.c
+++ b/tests-clar/diff/rename.c
@@ -23,8 +23,16 @@ void test_diff_rename__cleanup(void)
  *   serving.txt     -> sixserving.txt  (rename, no change, 100% match)
  *   sevencities.txt -> sevencities.txt (no change)
  *   sevencities.txt -> songofseven.txt (copy, no change, 100% match)
- *
- * TODO: add commits with various % changes of copy / rename
+ * commit 1c068dee5790ef1580cfc4cd670915b48d790084
+ *   songofseven.txt -> songofseven.txt (major rewrite, <20% match - split)
+ *   sixserving.txt  -> sixserving.txt  (indentation change)
+ *   sixserving.txt  -> ikeepsix.txt    (copy, add title, >80% match)
+ *   sevencities.txt                    (no change)
+ * commit 19dd32dfb1520a64e5bbaae8dce6ef423dfa2f13
+ *   songofseven.txt -> untimely.txt    (rename, convert to crlf)
+ *   ikeepsix.txt    -> ikeepsix.txt    (reorder sections in file)
+ *   sixserving.txt  -> sixserving.txt  (whitespace change - not just indent)
+ *   sevencities.txt -> songof7cities.txt (rename, small text changes)
  */
 
 void test_diff_rename__match_oid(void)
@@ -133,3 +141,108 @@ void test_diff_rename__checks_options_version(void)
 	git_tree_free(old_tree);
 	git_tree_free(new_tree);
 }
+
+void test_diff_rename__not_exact_match(void)
+{
+	const char *sha0 = "2bc7f351d20b53f1c72c16c4b036e491c478c49a";
+	const char *sha1 = "1c068dee5790ef1580cfc4cd670915b48d790084";
+	const char *sha2 = "19dd32dfb1520a64e5bbaae8dce6ef423dfa2f13";
+	git_tree *old_tree, *new_tree;
+	git_diff_list *diff;
+	git_diff_options diffopts = GIT_DIFF_OPTIONS_INIT;
+	git_diff_find_options opts = GIT_DIFF_FIND_OPTIONS_INIT;
+	diff_expects exp;
+
+	/* Changes:
+	 *   songofseven.txt -> songofseven.txt (major rewrite, <20% match - split)
+	 *   sixserving.txt  -> sixserving.txt  (indentation change)
+	 *   sixserving.txt  -> ikeepsix.txt    (copy, add title, >80% match)
+	 *   sevencities.txt                    (no change)
+	 */
+
+	old_tree = resolve_commit_oid_to_tree(g_repo, sha0);
+	new_tree = resolve_commit_oid_to_tree(g_repo, sha1);
+
+	/* Must pass GIT_DIFF_INCLUDE_UNMODIFIED if you expect to emulate
+	 * --find-copies-harder during rename transformion...
+	 */
+	diffopts.flags |= GIT_DIFF_INCLUDE_UNMODIFIED;
+
+	cl_git_pass(git_diff_tree_to_tree(
+		&diff, g_repo, old_tree, new_tree, &diffopts));
+
+	/* git diff --no-renames \
+	 *          2bc7f351d20b53f1c72c16c4b036e491c478c49a \
+	 *          1c068dee5790ef1580cfc4cd670915b48d790084
+	 */
+	memset(&exp, 0, sizeof(exp));
+	cl_git_pass(git_diff_foreach(
+		diff, diff_file_cb, diff_hunk_cb, diff_line_cb, &exp));
+
+	cl_assert_equal_i(4, exp.files);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_UNMODIFIED]);
+	cl_assert_equal_i(2, exp.file_status[GIT_DELTA_MODIFIED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_ADDED]);
+
+	/* git diff 31e47d8c1fa36d7f8d537b96158e3f024de0a9f2 \
+	 *          2bc7f351d20b53f1c72c16c4b036e491c478c49a
+	 */
+	cl_git_pass(git_diff_find_similar(diff, NULL));
+
+	memset(&exp, 0, sizeof(exp));
+	cl_git_pass(git_diff_foreach(
+		diff, diff_file_cb, diff_hunk_cb, diff_line_cb, &exp));
+
+	cl_assert_equal_i(4, exp.files);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_UNMODIFIED]);
+	cl_assert_equal_i(2, exp.file_status[GIT_DELTA_MODIFIED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_COPIED]);
+
+	git_diff_list_free(diff);
+
+	cl_git_pass(git_diff_tree_to_tree(
+		&diff, g_repo, old_tree, new_tree, &diffopts));
+
+	/* git diff --find-copies-harder --break-rewrites \
+	 *          2bc7f351d20b53f1c72c16c4b036e491c478c49a \
+	 *          1c068dee5790ef1580cfc4cd670915b48d790084
+	 */
+	opts.flags = GIT_DIFF_FIND_ALL;
+	cl_git_pass(git_diff_find_similar(diff, &opts));
+
+	memset(&exp, 0, sizeof(exp));
+	cl_git_pass(git_diff_foreach(
+		diff, diff_file_cb, diff_hunk_cb, diff_line_cb, &exp));
+
+	cl_assert_equal_i(5, exp.files);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_UNMODIFIED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_MODIFIED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_DELETED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_ADDED]);
+	cl_assert_equal_i(1, exp.file_status[GIT_DELTA_COPIED]);
+
+	git_diff_list_free(diff);
+
+	/* Changes:
+	 *   songofseven.txt -> untimely.txt    (rename, convert to crlf)
+	 *   ikeepsix.txt    -> ikeepsix.txt    (reorder sections in file)
+	 *   sixserving.txt  -> sixserving.txt  (whitespace - not just indent)
+	 *   sevencities.txt -> songof7cities.txt (rename, small text changes)
+	 */
+
+	git_tree_free(old_tree);
+	old_tree = new_tree;
+	new_tree = resolve_commit_oid_to_tree(g_repo, sha2);
+
+	/* moar tests needed */
+
+	git_tree_free(old_tree);
+	git_tree_free(new_tree);
+}
+
+void test_diff_rename__working_directory_changes(void)
+{
+	/* let's rewrite some files in the working directory on demand */
+
+	/* and with / without CRLF changes */
+}