oracle_compat.c

来自「postgresql8.3.4源码,开源数据库」· C语言 代码 · 共 1,498 行 · 第 1/3 页

/*-------------------------------------------------------------------------
 * oracle_compat.c
 *	Oracle compatible functions.
 *
 * Copyright (c) 1996-2008, PostgreSQL Global Development Group
 *
 *	Author: Edmund Mergl <[e-mail]>
 *	Multibyte enhancement: Tatsuo Ishii <[e-mail]>
 *
 *
 * IDENTIFICATION
 *	$PostgreSQL: pgsql/src/backend/utils/adt/oracle_compat.c,v 1.77 2008/01/01 19:45:52 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
#include "postgres.h"

#include <ctype.h>
#include <limits.h>
/*
 * towlower() and friends should be in <wctype.h>, but some pre-C99 systems
 * declare them in <wchar.h>.
 */
#ifdef HAVE_WCHAR_H
#include <wchar.h>
#endif
#ifdef HAVE_WCTYPE_H
#include <wctype.h>
#endif

#include "utils/builtins.h"
#include "utils/pg_locale.h"
#include "mb/pg_wchar.h"


/*
 * If the system provides the needed functions for wide-character manipulation
 * (which are all standardized by C99), then we implement upper/lower/initcap
 * using wide-character functions.	Otherwise we use the traditional <ctype.h>
 * functions, which of course will not work as desired in multibyte character
 * sets.  Note that in either case we are effectively assuming that the
 * database character encoding matches the encoding implied by LC_CTYPE.
 *
 * We assume if we have these two functions, we have their friends too, and
 * can use the wide-character method.
 */
#if defined(HAVE_WCSTOMBS) && defined(HAVE_TOWLOWER)
#define USE_WIDE_UPPER_LOWER
char	   *wstring_lower(char *str);
char	   *wstring_upper(char *str);
#endif

static text *dotrim(const char *string, int stringlen,
	   const char *set, int setlen,
	   bool doltrim, bool dortrim);


#ifdef USE_WIDE_UPPER_LOWER

/*
 * Convert a TEXT value into a palloc'd wchar string.
 */
static wchar_t *
texttowcs(const text *txt)
{
	int			nbytes = VARSIZE_ANY_EXHDR(txt);
	char	   *workstr;
	wchar_t    *result;
	size_t		ncodes;

	/* Overflow paranoia */
	if (nbytes < 0 ||
		nbytes > (int) (INT_MAX / sizeof(wchar_t)) - 1)
		ereport(ERROR,
				(errcode(ERRCODE_OUT_OF_MEMORY),
				 errmsg("out of memory")));

	/* Need a null-terminated version of the input */
	workstr = (char *) palloc(nbytes + 1);
	memcpy(workstr, VARDATA_ANY(txt), nbytes);
	workstr[nbytes] = '\0';

	/* Output workspace cannot have more codes than input bytes */
	result = (wchar_t *) palloc((nbytes + 1) * sizeof(wchar_t));

	/* Do the conversion */
	ncodes = mbstowcs(result, workstr, nbytes + 1);

	if (ncodes == (size_t) -1)
	{
		/*
		 * Invalid multibyte character encountered.  We try to give a useful
		 * error message by letting pg_verifymbstr check the string.  But it's
		 * possible that the string is OK to us, and not OK to mbstowcs ---
		 * this suggests that the LC_CTYPE locale is different from the
		 * database encoding.  Give a generic error message if verifymbstr
		 * can't find anything wrong.
		 */
		pg_verifymbstr(workstr, nbytes, false);
		ereport(ERROR,
				(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
				 errmsg("invalid multibyte character for locale"),
				 errhint("The server's LC_CTYPE locale is probably incompatible with the database encoding.")));
	}

	Assert(ncodes <= (size_t) nbytes);

	return result;
}


/*
 * Convert a wchar string into a palloc'd TEXT value.  The wchar string
 * must be zero-terminated, but we also require the caller to pass the string
 * length, since it will know it anyway in current uses.
 */
static text *
wcstotext(const wchar_t *str, int ncodes)
{
	text	   *result;
	size_t		nbytes;

	/* Overflow paranoia */
	if (ncodes < 0 ||
		ncodes > (int) ((INT_MAX - VARHDRSZ) / MB_CUR_MAX) - 1)
		ereport(ERROR,
				(errcode(ERRCODE_OUT_OF_MEMORY),
				 errmsg("out of memory")));

	/* Make workspace certainly large enough for result */
	result = (text *) palloc((ncodes + 1) * MB_CUR_MAX + VARHDRSZ);

	/* Do the conversion */
	nbytes = wcstombs((char *) VARDATA(result), str,
					  (ncodes + 1) * MB_CUR_MAX);

	if (nbytes == (size_t) -1)
	{
		/* Invalid multibyte character encountered ... shouldn't happen */
		ereport(ERROR,
				(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
				 errmsg("invalid multibyte character for locale")));
	}

	Assert(nbytes <= (size_t) (ncodes * MB_CUR_MAX));

	SET_VARSIZE(result, nbytes + VARHDRSZ);

	return result;
}
#endif   /* USE_WIDE_UPPER_LOWER */


/*
 * On Windows, the "Unicode" locales assume UTF16 not UTF8 encoding.
 * To make use of the upper/lower functionality, we need to map UTF8 to
 * UTF16, which for some reason mbstowcs and wcstombs won't do for us.
 * This conversion layer takes care of it.
 */

#ifdef WIN32

/* texttowcs for the case of UTF8 to UTF16 */
static wchar_t *
win32_utf8_texttowcs(const text *txt)
{
	int			nbytes = VARSIZE_ANY_EXHDR(txt);
	wchar_t    *result;
	int			r;

	/* Overflow paranoia */
	if (nbytes < 0 ||
		nbytes > (int) (INT_MAX / sizeof(wchar_t)) - 1)
		ereport(ERROR,
				(errcode(ERRCODE_OUT_OF_MEMORY),
				 errmsg("out of memory")));

	/* Output workspace cannot have more codes than input bytes */
	result = (wchar_t *) palloc((nbytes + 1) * sizeof(wchar_t));

	/* stupid Microsloth API does not work for zero-length input */
	if (nbytes == 0)
		r = 0;
	else
	{
		/* Do the conversion */
		r = MultiByteToWideChar(CP_UTF8, 0, VARDATA_ANY(txt), nbytes,
								result, nbytes);

		if (r <= 0)				/* assume it's NO_UNICODE_TRANSLATION */
		{
			/* see notes above about error reporting */
			pg_verifymbstr(VARDATA_ANY(txt), nbytes, false);
			ereport(ERROR,
					(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
					 errmsg("invalid multibyte character for locale"),
					 errhint("The server's LC_CTYPE locale is probably incompatible with the database encoding.")));
		}
	}

	/* Append trailing null wchar (MultiByteToWideChar won't have) */
	Assert(r <= nbytes);
	result[r] = 0;

	return result;
}

/* wcstotext for the case of UTF16 to UTF8 */
static text *
win32_utf8_wcstotext(const wchar_t *str)
{
	text	   *result;
	int			nbytes;
	int			r;

	/* Compute size of output string (this *will* include trailing null) */
	nbytes = WideCharToMultiByte(CP_UTF8, 0, str, -1, NULL, 0, NULL, NULL);
	if (nbytes <= 0)			/* shouldn't happen */
		ereport(ERROR,
				(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
				 errmsg("UTF-16 to UTF-8 translation failed: %lu",
						GetLastError())));

	result = palloc(nbytes + VARHDRSZ);

	r = WideCharToMultiByte(CP_UTF8, 0, str, -1, VARDATA(result), nbytes,
							NULL, NULL);
	if (r != nbytes)			/* shouldn't happen */
		ereport(ERROR,
				(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
				 errmsg("UTF-16 to UTF-8 translation failed: %lu",
						GetLastError())));

	SET_VARSIZE(result, nbytes + VARHDRSZ - 1); /* -1 to ignore null */

	return result;
}

/* interface layer to check which encoding is in use */

static wchar_t *
win32_texttowcs(const text *txt)
{
	if (GetDatabaseEncoding() == PG_UTF8)
		return win32_utf8_texttowcs(txt);
	else
		return texttowcs(txt);
}

static text *
win32_wcstotext(const wchar_t *str, int ncodes)
{
	if (GetDatabaseEncoding() == PG_UTF8)
		return win32_utf8_wcstotext(str);
	else
		return wcstotext(str, ncodes);
}

/* use macros to cause routines below to call interface layer */

#define texttowcs	win32_texttowcs
#define wcstotext	win32_wcstotext
#endif   /* WIN32 */

#ifdef USE_WIDE_UPPER_LOWER
/*
 * string_upper and string_lower are used for correct multibyte upper/lower
 * transformations localized strings. Returns pointers to transformated
 * string.
 */
char *
wstring_upper(char *str)
{
	wchar_t    *workspace;
	text	   *in_text;
	text	   *out_text;
	char	   *result;
	int			nbytes = strlen(str);
	int			i;

	in_text = palloc(nbytes + VARHDRSZ);
	memcpy(VARDATA(in_text), str, nbytes);
	SET_VARSIZE(in_text, nbytes + VARHDRSZ);

	workspace = texttowcs(in_text);

	for (i = 0; workspace[i] != 0; i++)
		workspace[i] = towupper(workspace[i]);

	out_text = wcstotext(workspace, i);

	nbytes = VARSIZE(out_text) - VARHDRSZ;
	result = palloc(nbytes + 1);
	memcpy(result, VARDATA(out_text), nbytes);

	result[nbytes] = '\0';

	pfree(workspace);
	pfree(in_text);
	pfree(out_text);

	return result;
}

char *
wstring_lower(char *str)
{
	wchar_t    *workspace;
	text	   *in_text;
	text	   *out_text;
	char	   *result;
	int			nbytes = strlen(str);
	int			i;

	in_text = palloc(nbytes + VARHDRSZ);
	memcpy(VARDATA(in_text), str, nbytes);
	SET_VARSIZE(in_text, nbytes + VARHDRSZ);

	workspace = texttowcs(in_text);

	for (i = 0; workspace[i] != 0; i++)
		workspace[i] = towlower(workspace[i]);

	out_text = wcstotext(workspace, i);

	nbytes = VARSIZE(out_text) - VARHDRSZ;
	result = palloc(nbytes + 1);
	memcpy(result, VARDATA(out_text), nbytes);

	result[nbytes] = '\0';

	pfree(workspace);
	pfree(in_text);
	pfree(out_text);

	return result;
}
#endif   /* USE_WIDE_UPPER_LOWER */

/********************************************************************
 *
 * lower
 *
 * Syntax:
 *
 *	 text lower(text string)
 *
 * Purpose:
 *
 *	 Returns string, with all letters forced to lowercase.
 *
 ********************************************************************/

Datum
lower(PG_FUNCTION_ARGS)
{
#ifdef USE_WIDE_UPPER_LOWER

	/*
	 * Use wide char code only when max encoding length > 1 and ctype != C.
	 * Some operating systems fail with multi-byte encodings and a C locale.
	 * Also, for a C locale there is no need to process as multibyte.
	 */
	if (pg_database_encoding_max_length() > 1 && !lc_ctype_is_c())
	{
		text	   *string = PG_GETARG_TEXT_PP(0);
		text	   *result;
		wchar_t    *workspace;
		int			i;

		workspace = texttowcs(string);

		for (i = 0; workspace[i] != 0; i++)
			workspace[i] = towlower(workspace[i]);

		result = wcstotext(workspace, i);

		pfree(workspace);

		PG_RETURN_TEXT_P(result);
	}
	else
#endif   /* USE_WIDE_UPPER_LOWER */
	{
		text	   *string = PG_GETARG_TEXT_P_COPY(0);
		char	   *ptr;
		int			m;

		/*
		 * Since we copied the string, we can scribble directly on the value
		 */
		ptr = VARDATA(string);
		m = VARSIZE(string) - VARHDRSZ;

		while (m-- > 0)
		{
			*ptr = tolower((unsigned char) *ptr);
			ptr++;
		}

		PG_RETURN_TEXT_P(string);
	}
}


/********************************************************************
 *
 * upper
 *
 * Syntax:
 *
 *	 text upper(text string)
 *
 * Purpose:
 *
 *	 Returns string, with all letters forced to uppercase.
 *
 ********************************************************************/

Datum
upper(PG_FUNCTION_ARGS)
{
#ifdef USE_WIDE_UPPER_LOWER

	/*
	 * Use wide char code only when max encoding length > 1 and ctype != C.
	 * Some operating systems fail with multi-byte encodings and a C locale.
	 * Also, for a C locale there is no need to process as multibyte.
	 */
	if (pg_database_encoding_max_length() > 1 && !lc_ctype_is_c())
	{
		text	   *string = PG_GETARG_TEXT_PP(0);
		text	   *result;
		wchar_t    *workspace;
		int			i;

		workspace = texttowcs(string);

		for (i = 0; workspace[i] != 0; i++)
			workspace[i] = towupper(workspace[i]);

		result = wcstotext(workspace, i);

		pfree(workspace);

		PG_RETURN_TEXT_P(result);
	}
	else
#endif   /* USE_WIDE_UPPER_LOWER */
	{
		text	   *string = PG_GETARG_TEXT_P_COPY(0);
		char	   *ptr;
		int			m;

		/*
		 * Since we copied the string, we can scribble directly on the value
		 */
		ptr = VARDATA(string);
		m = VARSIZE(string) - VARHDRSZ;

		while (m-- > 0)
		{
			*ptr = toupper((unsigned char) *ptr);
			ptr++;
		}

		PG_RETURN_TEXT_P(string);
	}
}


/********************************************************************
 *
 * initcap
 *
 * Syntax:
 *
 *	 text initcap(text string)
 *
 * Purpose:
 *
 *	 Returns string, with first letter of each word in uppercase, all
 *	 other letters in lowercase. A word is defined as a sequence of
 *	 alphanumeric characters, delimited by non-alphanumeric
 *	 characters.
 *
 ********************************************************************/

Datum
initcap(PG_FUNCTION_ARGS)
{
#ifdef USE_WIDE_UPPER_LOWER

	/*
	 * Use wide char code only when max encoding length > 1 and ctype != C.
	 * Some operating systems fail with multi-byte encodings and a C locale.
	 * Also, for a C locale there is no need to process as multibyte.
	 */
	if (pg_database_encoding_max_length() > 1 && !lc_ctype_is_c())
	{