On Fri, Jul 19, 2024 at 5:49 AM Tomas Vondra
<tomas.von...@enterprisedb.com> wrote:
>
> Hi,
>
> If I understand correctly, the problem is that it's not clear which of
> the 'abc' substrings is matched/returned by the function, right?
>
> I wonder if this is a problem only for understanding the test, or if it
> makes the tests a bit weaker. I mean, what if the function returns the
> wrong substring? How would we know?
>
this is for understanding the test.
personally, sometimes, I feel the documentation is too dry, hard to follow.
so i can based on regress tests better understand the documentation.
that was my intention for the changes.
we have more sophisticated regex test at
https://git.postgresql.org/cgit/postgresql.git/tree/src/test/modules/test_regex
> Also, if we tweak this, shouldn't we tweak also the regext_instr() calls
> a bit earlier in the test script?
>
sure.
please check attached.
From 0b9e7a9c32d4e3163ef0cc955a0916b9d5c7f395 Mon Sep 17 00:00:00 2001
From: jian he <jian.universal...@gmail.com>
Date: Thu, 25 Jul 2024 22:46:07 +0800
Subject: [PATCH v2 1/1] refactor regex related tests
some tests for regexp_instr, regexp_substr using "abcabcabc" as the source string,
which is not good for testing.
Because in many tests the matched pattern is "abc", but in the source string we have 3 "abc"
sometimes we don't know which "abc '' refers to.
changing the source string so the results would be easier to understand.
---
src/test/regress/expected/strings.out | 42 +++++++++++++++++++--------
src/test/regress/sql/strings.sql | 21 ++++++++------
2 files changed, 42 insertions(+), 21 deletions(-)
diff --git a/src/test/regress/expected/strings.out b/src/test/regress/expected/strings.out
index 52b69a10..4471f9a2 100644
--- a/src/test/regress/expected/strings.out
+++ b/src/test/regress/expected/strings.out
@@ -790,31 +790,43 @@ SELECT regexp_instr('abcdefghi', 'd.q');
0
(1 row)
-SELECT regexp_instr('abcabcabc', 'a.c');
+SELECT regexp_instr('abcaXcaYc', 'a.c');
regexp_instr
--------------
1
(1 row)
-SELECT regexp_instr('abcabcabc', 'a.c', 2);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 2);
regexp_instr
--------------
4
(1 row)
-SELECT regexp_instr('abcabcabc', 'a.c', 1, 3);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 3);
regexp_instr
--------------
7
(1 row)
-SELECT regexp_instr('abcabcabc', 'a.c', 1, 4);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 3, 1);
+ regexp_instr
+--------------
+ 10
+(1 row)
+
+SELECT regexp_instr('abcaXcaYc', 'a.c', 2, 2, 0);
+ regexp_instr
+--------------
+ 7
+(1 row)
+
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 4);
regexp_instr
--------------
0
(1 row)
-SELECT regexp_instr('abcabcabc', 'A.C', 1, 2, 0, 'i');
+SELECT regexp_instr('abcaXcayc', 'A.C', 1, 2, 0, 'i');
regexp_instr
--------------
4
@@ -925,22 +937,22 @@ SELECT regexp_substr('abcdefghi', 'd.q') IS NULL AS t;
t
(1 row)
-SELECT regexp_substr('abcabcabc', 'a.c');
+SELECT regexp_substr('abcaXcaYc', 'a.c');
regexp_substr
---------------
abc
(1 row)
-SELECT regexp_substr('abcabcabc', 'a.c', 2);
+SELECT regexp_substr('abcaXcaYc', 'a.c', 2);
regexp_substr
---------------
- abc
+ aXc
(1 row)
-SELECT regexp_substr('abcabcabc', 'a.c', 1, 3);
+SELECT regexp_substr('abcaXcaYc', 'a.c', 1, 3);
regexp_substr
---------------
- abc
+ aYc
(1 row)
SELECT regexp_substr('abcabcabc', 'a.c', 1, 4) IS NULL AS t;
@@ -949,10 +961,16 @@ SELECT regexp_substr('abcabcabc', 'a.c', 1, 4) IS NULL AS t;
t
(1 row)
-SELECT regexp_substr('abcabcabc', 'A.C', 1, 2, 'i');
+SELECT regexp_substr('abcaXcaYc', 'A.C', 1, 2, 'i');
regexp_substr
---------------
- abc
+ aXc
+(1 row)
+
+SELECT regexp_substr('abcaXcaYc', 'A.C', 2, 2, 'i');
+ regexp_substr
+---------------
+ aYc
(1 row)
SELECT regexp_substr('1234567890', '(123)(4(56)(78))', 1, 1, 'i', 0);
diff --git a/src/test/regress/sql/strings.sql b/src/test/regress/sql/strings.sql
index 39596789..d553a98c 100644
--- a/src/test/regress/sql/strings.sql
+++ b/src/test/regress/sql/strings.sql
@@ -244,11 +244,13 @@ SELECT regexp_like('abc', 'a.c', 'g'); -- error
-- regexp_instr tests
SELECT regexp_instr('abcdefghi', 'd.f');
SELECT regexp_instr('abcdefghi', 'd.q');
-SELECT regexp_instr('abcabcabc', 'a.c');
-SELECT regexp_instr('abcabcabc', 'a.c', 2);
-SELECT regexp_instr('abcabcabc', 'a.c', 1, 3);
-SELECT regexp_instr('abcabcabc', 'a.c', 1, 4);
-SELECT regexp_instr('abcabcabc', 'A.C', 1, 2, 0, 'i');
+SELECT regexp_instr('abcaXcaYc', 'a.c');
+SELECT regexp_instr('abcaXcaYc', 'a.c', 2);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 3);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 3, 1);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 2, 2, 0);
+SELECT regexp_instr('abcaXcaYc', 'a.c', 1, 4);
+SELECT regexp_instr('abcaXcayc', 'A.C', 1, 2, 0, 'i');
SELECT regexp_instr('1234567890', '(123)(4(56)(78))', 1, 1, 0, 'i', 0);
SELECT regexp_instr('1234567890', '(123)(4(56)(78))', 1, 1, 0, 'i', 1);
SELECT regexp_instr('1234567890', '(123)(4(56)(78))', 1, 1, 0, 'i', 2);
@@ -274,11 +276,12 @@ SELECT regexp_instr('abcabcabc', 'a.c', 1, 1, 0, '', -1);
-- regexp_substr tests
SELECT regexp_substr('abcdefghi', 'd.f');
SELECT regexp_substr('abcdefghi', 'd.q') IS NULL AS t;
-SELECT regexp_substr('abcabcabc', 'a.c');
-SELECT regexp_substr('abcabcabc', 'a.c', 2);
-SELECT regexp_substr('abcabcabc', 'a.c', 1, 3);
+SELECT regexp_substr('abcaXcaYc', 'a.c');
+SELECT regexp_substr('abcaXcaYc', 'a.c', 2);
+SELECT regexp_substr('abcaXcaYc', 'a.c', 1, 3);
SELECT regexp_substr('abcabcabc', 'a.c', 1, 4) IS NULL AS t;
-SELECT regexp_substr('abcabcabc', 'A.C', 1, 2, 'i');
+SELECT regexp_substr('abcaXcaYc', 'A.C', 1, 2, 'i');
+SELECT regexp_substr('abcaXcaYc', 'A.C', 2, 2, 'i');
SELECT regexp_substr('1234567890', '(123)(4(56)(78))', 1, 1, 'i', 0);
SELECT regexp_substr('1234567890', '(123)(4(56)(78))', 1, 1, 'i', 1);
SELECT regexp_substr('1234567890', '(123)(4(56)(78))', 1, 1, 'i', 2);
--
2.34.1
