diff options
author | Tom Lane <tgl@sss.pgh.pa.us> | 2022-10-06 11:08:56 -0400 |
---|---|---|
committer | Tom Lane <tgl@sss.pgh.pa.us> | 2022-10-06 11:08:56 -0400 |
commit | ca71131eebba63809ceb86be033a264a5f0a1d79 (patch) | |
tree | bff861724e38a284e336b166f60ce6c341c69973 /src | |
parent | 5757141caeaff4af0c206ab672d542cbd608542c (diff) | |
download | postgresql-ca71131eebba63809ceb86be033a264a5f0a1d79.tar.gz postgresql-ca71131eebba63809ceb86be033a264a5f0a1d79.zip |
Introduce t_isalnum() to replace t_isalpha() || t_isdigit() tests.
ts_locale.c omitted support for "isalnum" tests, perhaps on the
grounds that there were initially no use-cases for that. However,
both ltree and pg_trgm need such tests, and we do also have one
use-case now in the core backend. The workaround of testing
isalpha and isdigit separately seems quite inefficient, especially
when dealing with multibyte characters; so let's fill in the
missing support.
Discussion: https://postgr.es/m/2548310.1664999615@sss.pgh.pa.us
Diffstat (limited to 'src')
-rw-r--r-- | src/backend/tsearch/ts_locale.c | 16 | ||||
-rw-r--r-- | src/backend/utils/adt/tsquery.c | 2 | ||||
-rw-r--r-- | src/include/tsearch/ts_locale.h | 1 |
3 files changed, 18 insertions, 1 deletions
diff --git a/src/backend/tsearch/ts_locale.c b/src/backend/tsearch/ts_locale.c index e0aa570bf5e..fc21bf9ee8e 100644 --- a/src/backend/tsearch/ts_locale.c +++ b/src/backend/tsearch/ts_locale.c @@ -82,6 +82,22 @@ t_isalpha(const char *ptr) } int +t_isalnum(const char *ptr) +{ + int clen = pg_mblen(ptr); + wchar_t character[WC_BUF_LEN]; + Oid collation = DEFAULT_COLLATION_OID; /* TODO */ + pg_locale_t mylocale = 0; /* TODO */ + + if (clen == 1 || lc_ctype_is_c(collation)) + return isalnum(TOUCHAR(ptr)); + + char2wchar(character, WC_BUF_LEN, ptr, clen, mylocale); + + return iswalnum((wint_t) character[0]); +} + +int t_isprint(const char *ptr) { int clen = pg_mblen(ptr); diff --git a/src/backend/utils/adt/tsquery.c b/src/backend/utils/adt/tsquery.c index f49e6bb1578..a206926042e 100644 --- a/src/backend/utils/adt/tsquery.c +++ b/src/backend/utils/adt/tsquery.c @@ -248,7 +248,7 @@ parse_or_operator(TSQueryParserState pstate) return false; /* it shouldn't be a part of any word */ - if (t_iseq(ptr, '-') || t_iseq(ptr, '_') || t_isalpha(ptr) || t_isdigit(ptr)) + if (t_iseq(ptr, '-') || t_iseq(ptr, '_') || t_isalnum(ptr)) return false; for (;;) diff --git a/src/include/tsearch/ts_locale.h b/src/include/tsearch/ts_locale.h index d14cb4ed26d..10887290dc2 100644 --- a/src/include/tsearch/ts_locale.h +++ b/src/include/tsearch/ts_locale.h @@ -42,6 +42,7 @@ typedef struct extern int t_isdigit(const char *ptr); extern int t_isspace(const char *ptr); extern int t_isalpha(const char *ptr); +extern int t_isalnum(const char *ptr); extern int t_isprint(const char *ptr); extern char *lowerstr(const char *str); |