Pin two ctype-dependent test_regex_utf8 cases to pg_c_utf8

test_regex_utf8 decides whether to run by looking at the database
encoding alone, but two of its cases, [[:graph:]] and [[:print:]] over
E'xᔀሷ', depend on the ctype as well.  In a database with encoding
UTF8 and locale C they match just the x, because isgraph() and isprint()
are false for anything outside ASCII, and the file fails.

No buildfarm animal builds such a cluster, which is why this went
unnoticed, and why the to_date() crash in 18.5 went undetected for want
of exactly this coverage.  A pending buildfarm client change will let an
animal be configured that way.

Fix by giving the two cases an explicit collation, so that they exercise
a fixed Unicode ctype instead of whatever the database happened to be
initialized with.  test_regex() already passes its input collation down
to the regex compiler.  The expected results are unchanged; only the
echoed queries differ.

Backpatch-through: 17 (15 and 16 get a different fix)

Reviewed-by: Jonathan Gonzalez V. <jonathan@abdiel.eu>

2 files changed, 8 insertions, 4 deletions

diff --git a/src/test/modules/test_regex/expected/test_regex_utf8.out b/src/test/modules/test_regex/expected/test_regex_utf8.out
index befd75e96f3..23dd8b2f349 100644
--- a/src/test/modules/test_regex/expected/test_regex_utf8.out
+++ b/src/test/modules/test_regex/expected/test_regex_utf8.out
@@ -148,7 +148,9 @@ select * from test_regex('[[:digit:]]+',  E'x9\u1500\u1237', 'L');
  {9}
 (2 rows)
 
-select * from test_regex('[[:graph:]]+',  E'x\u1500\u1237', 'L');
+-- graph and print depend on the ctype and not just the encoding, so pin them
+-- to a Unicode-aware collation rather than the database's
+select * from test_regex('[[:graph:]]+',  E'x\u1500\u1237' COLLATE pg_c_utf8, 'L');
    test_regex    
 -----------------
  {0,REG_ULOCALE}
@@ -162,7 +164,7 @@ select * from test_regex('[[:lower:]]+',  E'x\u1500\u1237', 'L');
  {x}
 (2 rows)
 
-select * from test_regex('[[:print:]]+',  E'x\u1500\u1237', 'L');
+select * from test_regex('[[:print:]]+',  E'x\u1500\u1237' COLLATE pg_c_utf8, 'L');
    test_regex    
 -----------------
  {0,REG_ULOCALE}
diff --git a/src/test/modules/test_regex/sql/test_regex_utf8.sql b/src/test/modules/test_regex/sql/test_regex_utf8.sql
index 2aa3e0f1022..a208cf608d7 100644
--- a/src/test/modules/test_regex/sql/test_regex_utf8.sql
+++ b/src/test/modules/test_regex/sql/test_regex_utf8.sql
@@ -68,9 +68,11 @@ select * from test_regex('[[:ascii:]]+',  E'x\u1500\u1237', 'L');
 select * from test_regex('[[:blank:]]+',  E'x \t\u1500\u1237', 'L');
 select * from test_regex('[[:cntrl:]]+',  E'x\u1500\u1237', 'L');
 select * from test_regex('[[:digit:]]+',  E'x9\u1500\u1237', 'L');
-select * from test_regex('[[:graph:]]+',  E'x\u1500\u1237', 'L');
+-- graph and print depend on the ctype and not just the encoding, so pin them
+-- to a Unicode-aware collation rather than the database's
+select * from test_regex('[[:graph:]]+',  E'x\u1500\u1237' COLLATE pg_c_utf8, 'L');
 select * from test_regex('[[:lower:]]+',  E'x\u1500\u1237', 'L');
-select * from test_regex('[[:print:]]+',  E'x\u1500\u1237', 'L');
+select * from test_regex('[[:print:]]+',  E'x\u1500\u1237' COLLATE pg_c_utf8, 'L');
 select * from test_regex('[[:punct:]]+',  E'x.\u1500\u1237', 'L');
 select * from test_regex('[[:space:]]+',  E'x \t\u1500\u1237', 'L');
 select * from test_regex('[[:upper:]]+',  E'xX\u1500\u1237', 'L');