[PATCH v2 2/5] lib: introduce is_weak_cmp flag for json_path_cmp

[Kirill Shcherbatov <kshcherbatov@tarantool.org>]

Introduced is_weak_cmp flag for json_path_cmp routine: when this
flag is set, strings are equal when whole b path is an a subpath
(in terms of tokens).

Needed for #1257
---
 src/box/index_def.c    |  2 +-
 src/box/key_def.c      |  4 ++--
 src/box/memtx_engine.c |  2 +-
 src/box/vinyl.c        |  2 +-
 src/lib/json/json.c    |  4 +++-
 src/lib/json/json.h    |  6 +++++-
 test/unit/json.c       | 13 ++++++++++---
 test/unit/json.result  |  4 +++-
 8 files changed, 26 insertions(+), 11 deletions(-)

diff --git a/src/box/index_def.c b/src/box/index_def.c
index 6c37f9f1d..f0c7dc4f6 100644
--- a/src/box/index_def.c
+++ b/src/box/index_def.c
@@ -285,7 +285,7 @@ index_def_is_valid(struct index_def *index_def, const char *space_name)
 			if (part_a->fieldno == part_b->fieldno &&
 			    json_path_cmp(part_a->path, part_a->path_len,
 					  part_b->path, part_b->path_len,
-					  TUPLE_INDEX_BASE) == 0) {
+					  false, TUPLE_INDEX_BASE) == 0) {
 				diag_set(ClientError, ER_MODIFY_INDEX,
 					 index_def->name, space_name,
 					 "same key part is indexed twice");
diff --git a/src/box/key_def.c b/src/box/key_def.c
index 55dcf1eb5..f9e464402 100644
--- a/src/box/key_def.c
+++ b/src/box/key_def.c
@@ -311,7 +311,7 @@ key_part_cmp(const struct key_part *parts1, uint32_t part_count1,
 			       key_part_is_nullable(part2) ? -1 : 1;
 		int rc = json_path_cmp(part1->path, part1->path_len,
 				       part2->path, part2->path_len,
-				       TUPLE_INDEX_BASE);
+				       false, TUPLE_INDEX_BASE);
 		if (rc != 0)
 			return rc;
 	}
@@ -606,7 +606,7 @@ key_def_find(const struct key_def *key_def, const struct key_part *to_find)
 		if (part->fieldno == to_find->fieldno &&
 		    json_path_cmp(part->path, part->path_len,
 				  to_find->path, to_find->path_len,
-				  TUPLE_INDEX_BASE) == 0)
+				  false, TUPLE_INDEX_BASE) == 0)
 			return part;
 	}
 	return NULL;
diff --git a/src/box/memtx_engine.c b/src/box/memtx_engine.c
index d468d1cd8..3e959ff2e 100644
--- a/src/box/memtx_engine.c
+++ b/src/box/memtx_engine.c
@@ -1317,7 +1317,7 @@ memtx_index_def_change_requires_rebuild(struct index *index,
 			return true;
 		if (json_path_cmp(old_part->path, old_part->path_len,
 				  new_part->path, new_part->path_len,
-				  TUPLE_INDEX_BASE) != 0)
+				  false, TUPLE_INDEX_BASE) != 0)
 			return true;
 	}
 	return false;
diff --git a/src/box/vinyl.c b/src/box/vinyl.c
index c2b7eb78f..78f29c624 100644
--- a/src/box/vinyl.c
+++ b/src/box/vinyl.c
@@ -1013,7 +1013,7 @@ vinyl_index_def_change_requires_rebuild(struct index *index,
 			return true;
 		if (json_path_cmp(old_part->path, old_part->path_len,
 				  new_part->path, new_part->path_len,
-				  TUPLE_INDEX_BASE) != 0)
+				  false, TUPLE_INDEX_BASE) != 0)
 			return true;
 	}
 	return false;
diff --git a/src/lib/json/json.c b/src/lib/json/json.c
index 568e9c246..4e1efb7a9 100644
--- a/src/lib/json/json.c
+++ b/src/lib/json/json.c
@@ -281,7 +281,7 @@ json_token_cmp(const struct json_token *a, const struct json_token *b)
 
 int
 json_path_cmp(const char *a, int a_len, const char *b, int b_len,
-	      int index_base)
+	      bool is_weak_cmp, int index_base)
 {
 	struct json_lexer lexer_a, lexer_b;
 	json_lexer_create(&lexer_a, a, a_len, index_base);
@@ -301,6 +301,8 @@ json_path_cmp(const char *a, int a_len, const char *b, int b_len,
 	}
 	/* Paths a and b must be valid. */
 	assert(rc_b == 0 && rc_b == 0);
+	if (is_weak_cmp)
+		return 0;
 	/*
 	 * The parser stopped because the end of one of the paths
 	 * was reached. As JSON_TOKEN_END > JSON_TOKEN_{NUM, STR},
diff --git a/src/lib/json/json.h b/src/lib/json/json.h
index 4c1bc18bc..53a0da5f9 100644
--- a/src/lib/json/json.h
+++ b/src/lib/json/json.h
@@ -245,12 +245,16 @@ json_lexer_next_token(struct json_lexer *lexer, struct json_token *token);
  * Compare two JSON paths using Lexer class.
  * - in case of paths that have same token-sequence prefix,
  *   the path having more tokens is assumed to be greater
+ *   (except the case when is_weak_cmp flag is set).
  * - both paths must be valid
  *   (may be tested with json_path_validate).
+ *
+ * When is_weak_cmp flag is set, strings are assumed to be equal
+ * when all b path tokens are present in a path.
  */
 int
 json_path_cmp(const char *a, int a_len, const char *b, int b_len,
-	      int index_base);
+	      bool is_weak_cmp, int index_base);
 
 /**
  * Check if the passed JSON path is valid.
diff --git a/test/unit/json.c b/test/unit/json.c
index 1db7e9364..10704451b 100644
--- a/test/unit/json.c
+++ b/test/unit/json.c
@@ -479,12 +479,12 @@ test_path_cmp()
 		{"Data[1][\"Info\"].fname[1]", -1},
 	};
 	header();
-	plan(lengthof(rc) + 6);
+	plan(lengthof(rc) + 8);
 	for (size_t i = 0; i < lengthof(rc); ++i) {
 		const char *path = rc[i].path;
 		int errpos = rc[i].errpos;
 		int rc = json_path_cmp(a, a_len, path, strlen(path),
-				       INDEX_BASE);
+				       false, INDEX_BASE);
 		if (rc > 0) rc = 1;
 		if (rc < 0) rc = -1;
 		is(rc, errpos, "path cmp result \"%s\" with \"%s\": "
@@ -493,7 +493,7 @@ test_path_cmp()
 	char *multikey_a = "Data[*][\"FIO\"].fname[*]";
 	char *multikey_b = "[\"Data\"][*].FIO[\"fname\"][*]";
 	int ret = json_path_cmp(multikey_a, strlen(multikey_a), multikey_b,
-				strlen(multikey_b), INDEX_BASE);
+				strlen(multikey_b), false, INDEX_BASE);
 	is(ret, 0, "path cmp result \"%s\" with \"%s\": have %d, expected %d",
 	   multikey_a, multikey_b, ret, 0);
 
@@ -510,6 +510,13 @@ test_path_cmp()
 	   "test json_path_is_multikey: multikey path");
 	is(ret, 8, "test json_path_is_multikey: multikey path %d/%d", ret, 8);
 
+	is(json_path_cmp(a, a_len, multikey_b, ret, false, INDEX_BASE) == 0, false,
+	   "test json_path_cmp  with is_weak_cmp = false: compare %s and %.*s",
+	   a, ret, multikey_b);
+	is(json_path_cmp(a, a_len, multikey_b, ret, true, INDEX_BASE) == 0, true,
+	   "test json_path_cmp with is_weak_cmp = true: compare %s and %.*s",
+	   a, ret, multikey_b);
+
 	check_plan();
 	footer();
 }
diff --git a/test/unit/json.result b/test/unit/json.result
index 0da6c8c6b..e57dc748a 100644
--- a/test/unit/json.result
+++ b/test/unit/json.result
@@ -170,7 +170,7 @@ ok 2 - subtests
 ok 3 - subtests
 	*** test_tree: done ***
 	*** test_path_cmp ***
-    1..11
+    1..13
     ok 1 - path cmp result "Data[1]["FIO"].fname" with "Data[1]["FIO"].fname": have 0, expected 0
     ok 2 - path cmp result "Data[1]["FIO"].fname" with "["Data"][1].FIO["fname"]": have 0, expected 0
     ok 3 - path cmp result "Data[1]["FIO"].fname" with "Data[1]": have 1, expected 1
@@ -182,6 +182,8 @@ ok 3 - subtests
     ok 9 - test json_path_is_multikey: non-multikey path
     ok 10 - test json_path_is_multikey: multikey path
     ok 11 - test json_path_is_multikey: multikey path 8/8
+    ok 12 - test json_path_cmp  with is_weak_cmp = false: compare Data[1]["FIO"].fname and ["Data"]
+    ok 13 - test json_path_cmp with is_weak_cmp = true: compare Data[1]["FIO"].fname and ["Data"]
 ok 4 - subtests
 	*** test_path_cmp: done ***
 	*** test_path_snprint ***
-- 
2.21.0