Re: Collation rules and multi-lingual databases

Greg Stark wrote:
> Yeah I thought of that. But if making it a critical section is cheap then it's
> probably a better approach. The problem with restoring the locale for the
> palloc is that if the user is unlucky he might sort a table of thousands of
> strings that all trigger the exception case.
>

What about something like this?
8<--------------------------------

#include <setjmp.h>
#include <string.h>

#include "postgres.h"
#include "fmgr.h"
#include "tcop/tcopprot.h"
#include "utils/builtins.h"

#define GET_STR(textp) \
DatumGetCString(DirectFunctionCall1(textout, PointerGetDatum(textp)))
#define GET_BYTEA(str_) \
DatumGetTextP(DirectFunctionCall1(byteain, CStringGetDatum(str_)))
#define MAX_BYTEA_LEN 0x3fffffff

/*
* pg_strxfrm - Function to convert string similar to the strxfrm C
* function using a specified locale.
*/
extern Datum pg_strxfrm(PG_FUNCTION_ARGS);
PG_FUNCTION_INFO_V1(pg_strxfrm);

Datum
pg_strxfrm(PG_FUNCTION_ARGS)
{
char *str = GET_STR(PG_GETARG_TEXT_P(0));
size_t str_len = strlen(str);
char *localestr = GET_STR(PG_GETARG_TEXT_P(1));
size_t approx_trans_len = 4 + (str_len * 3);
char *trans = (char *) palloc(approx_trans_len + 1);
size_t actual_trans_len;
char *oldlocale;
char *newlocale;
sigjmp_buf save_restart;

if (approx_trans_len > MAX_BYTEA_LEN)
elog(ERROR, "source string too long to transform");

oldlocale = setlocale(LC_COLLATE, NULL);
if (!oldlocale)
elog(ERROR, "setlocale failed to return a locale");

/* catch elog while locale is set other than the default */
memcpy(&save_restart, &Warn_restart, sizeof(save_restart));
if (sigsetjmp(Warn_restart, 1) != 0)
{
memcpy(&Warn_restart, &save_restart, sizeof(Warn_restart));
newlocale = setlocale(LC_COLLATE, oldlocale);
if (!newlocale)
elog(PANIC, "setlocale failed to reset locale: %s", localestr);
siglongjmp(Warn_restart, 1);
}

newlocale = setlocale(LC_COLLATE, localestr);
if (!newlocale)
elog(ERROR, "setlocale failed to set a locale: %s", localestr);

actual_trans_len = strxfrm(trans, str, approx_trans_len + 1);

/* if the buffer was not large enough, resize it and try again */
if (actual_trans_len >= approx_trans_len)
{
approx_trans_len = actual_trans_len + 1;
if (approx_trans_len > MAX_BYTEA_LEN)
elog(ERROR, "source string too long to transform");

trans = (char *) repalloc(trans, approx_trans_len + 1);
actual_trans_len = strxfrm(trans, str, approx_trans_len + 1);

/* if the buffer still not large enough, punt */
if (actual_trans_len >= approx_trans_len)
elog(ERROR, "strxfrm failed, buffer insufficient");
}

newlocale = setlocale(LC_COLLATE, oldlocale);
if (!newlocale)
elog(PANIC, "setlocale failed to reset locale: %s", localestr);

PG_RETURN_BYTEA_P(GET_BYTEA(trans));
}

8<--------------------------------

Joe