pg_locale_builtin.c File Reference

#include "postgres.h"
#include "catalog/pg_database.h"
#include "catalog/pg_collation.h"
#include "common/unicode_case.h"
#include "common/unicode_category.h"
#include "common/unicode_limits.h"
#include "miscadmin.h"
#include "utils/builtins.h"
#include "utils/memutils.h"
#include "utils/pg_locale.h"
#include "utils/syscache.h"

Include dependency graph for pg_locale_builtin.c:

Go to the source code of this file.

Data Structures
struct	WordBoundaryState

Functions
	StaticAssertDecl (SIZE_MAX/UTF8_MAX_CASEMAP_EXPANSION > MaxAllocSize, "case mapping may overflow size_t")
pg_locale_t	create_pg_locale_builtin (Oid collid, MemoryContext context)
char *	get_collation_actual_version_builtin (const char *collcollate)
static char32_t	to_char32 (pg_wchar wc)
static pg_wchar	to_pg_wchar (char32_t c32)
static size_t	initcap_wbnext (void *state)
static size_t	strlower_builtin (char *dest, size_t destsize, const char *src, size_t srclen, pg_locale_t locale)
static size_t	strtitle_builtin (char *dest, size_t destsize, const char *src, size_t srclen, pg_locale_t locale)
static size_t	strupper_builtin (char *dest, size_t destsize, const char *src, size_t srclen, pg_locale_t locale)
static size_t	strfold_builtin (char *dest, size_t destsize, const char *src, size_t srclen, pg_locale_t locale)
static bool	wc_isdigit_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isalpha_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isalnum_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isupper_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_islower_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isgraph_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isprint_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_ispunct_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isspace_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_isxdigit_builtin (pg_wchar wc, pg_locale_t locale)
static bool	wc_iscased_builtin (pg_wchar wc, pg_locale_t locale)
static pg_wchar	wc_toupper_builtin (pg_wchar wc, pg_locale_t locale)
static pg_wchar	wc_tolower_builtin (pg_wchar wc, pg_locale_t locale)

Variables
static const struct ctype_methods	ctype_methods_builtin

Function Documentation

create_pg_locale_builtin()

pg_locale_t create_pg_locale_builtin ( Oid  collid, MemoryContext  context  )

extern

Definition at line 283 of file pg_locale_builtin.c.

{
    const char *locstr;
    pg_locale_t result;
 
    if (collid == DEFAULT_COLLATION_OID)
    {
        HeapTuple   tp;
        Datum       datum;
 
        tp = SearchSysCache1(DATABASEOID, ObjectIdGetDatum(MyDatabaseId));
        if (!HeapTupleIsValid(tp))
            elog(ERROR, "cache lookup failed for database %u", MyDatabaseId);
        datum = SysCacheGetAttrNotNull(DATABASEOID, tp,
                                       Anum_pg_database_datlocale);
        locstr = TextDatumGetCString(datum);
        ReleaseSysCache(tp);
    }
    else
    {
        HeapTuple   tp;
        Datum       datum;
 
        tp = SearchSysCache1(COLLOID, ObjectIdGetDatum(collid));
        if (!HeapTupleIsValid(tp))
            elog(ERROR, "cache lookup failed for collation %u", collid);
        datum = SysCacheGetAttrNotNull(COLLOID, tp,
                                       Anum_pg_collation_colllocale);
        locstr = TextDatumGetCString(datum);
        ReleaseSysCache(tp);
    }
 
    builtin_validate_locale(GetDatabaseEncoding(), locstr);
 
    result = MemoryContextAllocZero(context, sizeof(struct pg_locale_struct));
 
    result->builtin.locale = MemoryContextStrdup(context, locstr);
    result->builtin.casemap_full = (strcmp(locstr, "PG_UNICODE_FAST") == 0);
    result->deterministic = true;
    result->collate_is_c = true;
    result->ctype_is_c = (strcmp(locstr, "C") == 0);
    if (!result->ctype_is_c)
        result->ctype = &ctype_methods_builtin;
 
    return result;
}

References builtin_validate_locale(), collid, ctype_methods_builtin, elog, ERROR, fb(), GetDatabaseEncoding(), HeapTupleIsValid, MemoryContextAllocZero(), MemoryContextStrdup(), MyDatabaseId, ObjectIdGetDatum(), ReleaseSysCache(), result, SearchSysCache1(), SysCacheGetAttrNotNull(), and TextDatumGetCString.

Referenced by create_pg_locale(), and init_database_collation().

get_collation_actual_version_builtin()

char * get_collation_actual_version_builtin ( const char *  collcollate )

extern

Definition at line 331 of file pg_locale_builtin.c.

{
    /*
     * The only two supported locales (C and C.UTF-8) are both based on memcmp
     * and are not expected to change, but track the version anyway.
     *
     * Note that the character semantics may change for some locales, but the
     * collation version only tracks changes to sort order.
     */
    if (strcmp(collcollate, "C") == 0)
        return "1";
    else if (strcmp(collcollate, "C.UTF-8") == 0)
        return "1";
    else if (strcmp(collcollate, "PG_UNICODE_FAST") == 0)
        return "1";
    else
        ereport(ERROR,
                (errcode(ERRCODE_WRONG_OBJECT_TYPE),
                 errmsg("invalid locale name \"%s\" for builtin provider",
                        collcollate)));
 
    return NULL;                /* keep compiler quiet */
}

References ereport, errcode(), errmsg, ERROR, and fb().

Referenced by get_collation_actual_version().

initcap_wbnext()

static size_t initcap_wbnext ( void *  state )

static

Definition at line 70 of file pg_locale_builtin.c.

{
    struct WordBoundaryState *wbstate = (struct WordBoundaryState *) state;
 
    while (wbstate->offset < wbstate->len)
    {
        int         ulen = pg_utf_mblen((const unsigned char *) wbstate->str +
                                        wbstate->offset);
        char32_t    u;
        bool        curr_alnum;
        size_t      prev_offset = wbstate->offset;
 
        /* invalid UTF8 */
        if (wbstate->offset + ulen > wbstate->len)
        {
            wbstate->init = true;
            wbstate->offset = wbstate->len;
            return prev_offset;
        }
 
        u = utf8_to_unicode((const unsigned char *) wbstate->str +
                            wbstate->offset);
        curr_alnum = pg_u_isalnum(u, wbstate->posix);
 
        if (!wbstate->init || curr_alnum != wbstate->prev_alnum)
        {
            wbstate->init = true;
            wbstate->offset += ulen;
            wbstate->prev_alnum = curr_alnum;
            return prev_offset;
        }
 
        wbstate->offset += ulen;
    }
 
    return wbstate->len;
}

References fb(), pg_u_isalnum(), pg_utf_mblen, and utf8_to_unicode().

Referenced by strtitle_builtin().

StaticAssertDecl()

StaticAssertDecl	(	SIZE_MAX/	UTF8_MAX_CASEMAP_EXPANSION,
		MaxAllocSize	,
		"case mapping may overflow size_t"
	)

strfold_builtin()

static size_t strfold_builtin ( char *  dest, size_t  destsize, const char *  src, size_t  srclen, pg_locale_t  locale  )

static

Definition at line 167 of file pg_locale_builtin.c.

{
    size_t      consumed;
    size_t      result;
 
    result = unicode_strfold(dest, destsize, src, srclen, &consumed,
                             locale->builtin.casemap_full);
    if (consumed < srclen)
        report_invalid_encoding(GetDatabaseEncoding(), src + consumed,
                                srclen - consumed);
 
    return result;
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, fb(), GetDatabaseEncoding(), report_invalid_encoding(), result, and unicode_strfold().

strlower_builtin()

static size_t strlower_builtin ( char *  dest, size_t  destsize, const char *  src, size_t  srclen, pg_locale_t  locale  )

static

Definition at line 109 of file pg_locale_builtin.c.

{
    size_t      consumed;
    size_t      result;
 
    result = unicode_strlower(dest, destsize, src, srclen, &consumed,
                              locale->builtin.casemap_full);
    if (consumed < srclen)
        report_invalid_encoding(GetDatabaseEncoding(), src + consumed,
                                srclen - consumed);
 
    return result;
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, fb(), GetDatabaseEncoding(), report_invalid_encoding(), result, and unicode_strlower().

strtitle_builtin()

static size_t strtitle_builtin ( char *  dest, size_t  destsize, const char *  src, size_t  srclen, pg_locale_t  locale  )

static

Definition at line 125 of file pg_locale_builtin.c.

{
    struct WordBoundaryState wbstate = {
        .str = src,
        .len = srclen,
        .offset = 0,
        .posix = !locale->builtin.casemap_full,
        .init = false,
        .prev_alnum = false,
    };
    size_t      consumed;
    size_t      result;
 
    result = unicode_strtitle(dest, destsize, src, srclen, &consumed,
                              locale->builtin.casemap_full,
                              initcap_wbnext, &wbstate);
 
    if (consumed < srclen)
        report_invalid_encoding(GetDatabaseEncoding(), src + consumed,
                                srclen - consumed);
 
    return result;
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, fb(), GetDatabaseEncoding(), initcap_wbnext(), report_invalid_encoding(), result, WordBoundaryState::str, and unicode_strtitle().

strupper_builtin()

static size_t strupper_builtin ( char *  dest, size_t  destsize, const char *  src, size_t  srclen, pg_locale_t  locale  )

static

Definition at line 151 of file pg_locale_builtin.c.

{
    size_t      consumed;
    size_t      result;
 
    result = unicode_strupper(dest, destsize, src, srclen, &consumed,
                              locale->builtin.casemap_full);
    if (consumed < srclen)
        report_invalid_encoding(GetDatabaseEncoding(), src + consumed,
                                srclen - consumed);
 
    return result;
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, fb(), GetDatabaseEncoding(), report_invalid_encoding(), result, and unicode_strupper().

to_char32()

static char32_t to_char32 ( pg_wchar  wc )

inlinestatic

Definition at line 52 of file pg_locale_builtin.c.

{
    Assert(GetDatabaseEncoding() == PG_UTF8);
    return (char32_t) wc;
}

References Assert, GetDatabaseEncoding(), and PG_UTF8.

Referenced by wc_isalnum_builtin(), wc_isalpha_builtin(), wc_iscased_builtin(), wc_isdigit_builtin(), wc_isgraph_builtin(), wc_islower_builtin(), wc_isprint_builtin(), wc_ispunct_builtin(), wc_isspace_builtin(), wc_isupper_builtin(), wc_isxdigit_builtin(), wc_tolower_builtin(), and wc_toupper_builtin().

to_pg_wchar()

static pg_wchar to_pg_wchar ( char32_t  c32 )

inlinestatic

Definition at line 59 of file pg_locale_builtin.c.

{
    Assert(GetDatabaseEncoding() == PG_UTF8);
    return (pg_wchar) c32;
}

References Assert, fb(), GetDatabaseEncoding(), and PG_UTF8.

Referenced by wc_tolower_builtin(), and wc_toupper_builtin().

wc_isalnum_builtin()

static bool wc_isalnum_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 195 of file pg_locale_builtin.c.

{
    return pg_u_isalnum(to_char32(wc), !locale->builtin.casemap_full);
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, pg_u_isalnum(), and to_char32().

wc_isalpha_builtin()

static bool wc_isalpha_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 189 of file pg_locale_builtin.c.

{
    return pg_u_isalpha(to_char32(wc));
}

References pg_u_isalpha(), and to_char32().

wc_iscased_builtin()

static bool wc_iscased_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 243 of file pg_locale_builtin.c.

{
    return pg_u_prop_cased(to_char32(wc));
}

References pg_u_prop_cased(), and to_char32().

wc_isdigit_builtin()

static bool wc_isdigit_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 183 of file pg_locale_builtin.c.

{
    return pg_u_isdigit(to_char32(wc), !locale->builtin.casemap_full);
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, pg_u_isdigit(), and to_char32().

wc_isgraph_builtin()

static bool wc_isgraph_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 213 of file pg_locale_builtin.c.

{
    return pg_u_isgraph(to_char32(wc));
}

References pg_u_isgraph(), and to_char32().

wc_islower_builtin()

static bool wc_islower_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 207 of file pg_locale_builtin.c.

{
    return pg_u_islower(to_char32(wc));
}

References pg_u_islower(), and to_char32().

wc_isprint_builtin()

static bool wc_isprint_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 219 of file pg_locale_builtin.c.

{
    return pg_u_isprint(to_char32(wc));
}

References pg_u_isprint(), and to_char32().

wc_ispunct_builtin()

static bool wc_ispunct_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 225 of file pg_locale_builtin.c.

{
    return pg_u_ispunct(to_char32(wc), !locale->builtin.casemap_full);
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, pg_u_ispunct(), and to_char32().

wc_isspace_builtin()

static bool wc_isspace_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 231 of file pg_locale_builtin.c.

{
    return pg_u_isspace(to_char32(wc));
}

References pg_u_isspace(), and to_char32().

wc_isupper_builtin()

static bool wc_isupper_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 201 of file pg_locale_builtin.c.

{
    return pg_u_isupper(to_char32(wc));
}

References pg_u_isupper(), and to_char32().

wc_isxdigit_builtin()

static bool wc_isxdigit_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 237 of file pg_locale_builtin.c.

{
    return pg_u_isxdigit(to_char32(wc), !locale->builtin.casemap_full);
}

References pg_locale_struct::builtin, pg_locale_struct::casemap_full, pg_u_isxdigit(), and to_char32().

wc_tolower_builtin()

static pg_wchar wc_tolower_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 255 of file pg_locale_builtin.c.

{
    return to_pg_wchar(unicode_lowercase_simple(to_char32(wc)));
}

References to_char32(), to_pg_wchar(), and unicode_lowercase_simple().

wc_toupper_builtin()

static pg_wchar wc_toupper_builtin ( pg_wchar  wc, pg_locale_t  locale  )

static

Definition at line 249 of file pg_locale_builtin.c.

{
    return to_pg_wchar(unicode_uppercase_simple(to_char32(wc)));
}

References to_char32(), to_pg_wchar(), and unicode_uppercase_simple().

Variable Documentation

ctype_methods_builtin

const struct ctype_methods ctype_methods_builtin

static

Initial value:

= {
    .strlower = strlower_builtin,
    .strtitle = strtitle_builtin,
    .strupper = strupper_builtin,
    .strfold = strfold_builtin,
    
    .downcase_ident = NULL,
    .wc_isdigit = wc_isdigit_builtin,
    .wc_isalpha = wc_isalpha_builtin,
    .wc_isalnum = wc_isalnum_builtin,
    .wc_isupper = wc_isupper_builtin,
    .wc_islower = wc_islower_builtin,
    .wc_isgraph = wc_isgraph_builtin,
    .wc_isprint = wc_isprint_builtin,
    .wc_ispunct = wc_ispunct_builtin,
    .wc_isspace = wc_isspace_builtin,
    .wc_isxdigit = wc_isxdigit_builtin,
    .wc_iscased = wc_iscased_builtin,
    .wc_tolower = wc_tolower_builtin,
    .wc_toupper = wc_toupper_builtin,
}

Definition at line 260 of file pg_locale_builtin.c.

                                                          {
    .strlower = strlower_builtin,
    .strtitle = strtitle_builtin,
    .strupper = strupper_builtin,
    .strfold = strfold_builtin,
    /* uses plain ASCII semantics for historical reasons */
    .downcase_ident = NULL,
    .wc_isdigit = wc_isdigit_builtin,
    .wc_isalpha = wc_isalpha_builtin,
    .wc_isalnum = wc_isalnum_builtin,
    .wc_isupper = wc_isupper_builtin,
    .wc_islower = wc_islower_builtin,
    .wc_isgraph = wc_isgraph_builtin,
    .wc_isprint = wc_isprint_builtin,
    .wc_ispunct = wc_ispunct_builtin,
    .wc_isspace = wc_isspace_builtin,
    .wc_isxdigit = wc_isxdigit_builtin,
    .wc_iscased = wc_iscased_builtin,
    .wc_tolower = wc_tolower_builtin,
    .wc_toupper = wc_toupper_builtin,
};

Referenced by create_pg_locale_builtin().