formatting.c

PostgreSQL 8.1.4的源码 适用于Linux下的开源数据库系统

	DCH_hh24,
	DCH_hh12,
	DCH_hh,
	DCH_iw,
	DCH_iyyy,
	DCH_iyy,
	DCH_iy,
	DCH_i,
	DCH_j,
	DCH_mi,
	DCH_mm,
	DCH_month,
	DCH_mon,
	DCH_ms,
	DCH_p_m,
	DCH_pm,
	DCH_q,
	DCH_rm,
	DCH_ssss,
	DCH_ss,
	DCH_tz,
	DCH_us,
	DCH_ww,
	DCH_w,
	DCH_y_yyy,
	DCH_yyyy,
	DCH_yyy,
	DCH_yy,
	DCH_y,

	/* last */
	_DCH_last_
} DCH_poz;

typedef enum
{
	NUM_COMMA,
	NUM_DEC,
	NUM_0,
	NUM_9,
	NUM_B,
	NUM_C,
	NUM_D,
	NUM_E,
	NUM_FM,
	NUM_G,
	NUM_L,
	NUM_MI,
	NUM_PL,
	NUM_PR,
	NUM_RN,
	NUM_SG,
	NUM_SP,
	NUM_S,
	NUM_TH,
	NUM_V,
	NUM_b,
	NUM_c,
	NUM_d,
	NUM_e,
	NUM_fm,
	NUM_g,
	NUM_l,
	NUM_mi,
	NUM_pl,
	NUM_pr,
	NUM_rn,
	NUM_sg,
	NUM_sp,
	NUM_s,
	NUM_th,
	NUM_v,

	/* last */
	_NUM_last_
} NUM_poz;

/* ----------
 * KeyWords for DATE-TIME version
 * ----------
 */
static const KeyWord DCH_keywords[] = {
/*	keyword, len, func, type, isitdigit			 is in Index */
	{"A.D.", 4, dch_date, DCH_A_D, FALSE},		/* A */
	{"A.M.", 4, dch_time, DCH_A_M, FALSE},
	{"AD", 2, dch_date, DCH_AD, FALSE},
	{"AM", 2, dch_time, DCH_AM, FALSE},
	{"B.C.", 4, dch_date, DCH_B_C, FALSE},		/* B */
	{"BC", 2, dch_date, DCH_BC, FALSE},
	{"CC", 2, dch_date, DCH_CC, TRUE},	/* C */
	{"DAY", 3, dch_date, DCH_DAY, FALSE},		/* D */
	{"DDD", 3, dch_date, DCH_DDD, TRUE},
	{"DD", 2, dch_date, DCH_DD, TRUE},
	{"DY", 2, dch_date, DCH_DY, FALSE},
	{"Day", 3, dch_date, DCH_Day, FALSE},
	{"Dy", 2, dch_date, DCH_Dy, FALSE},
	{"D", 1, dch_date, DCH_D, TRUE},
	{"FX", 2, dch_global, DCH_FX, FALSE},		/* F */
	{"HH24", 4, dch_time, DCH_HH24, TRUE},		/* H */
	{"HH12", 4, dch_time, DCH_HH12, TRUE},
	{"HH", 2, dch_time, DCH_HH, TRUE},
	{"IW", 2, dch_date, DCH_IW, TRUE},	/* I */
	{"IYYY", 4, dch_date, DCH_IYYY, TRUE},
	{"IYY", 3, dch_date, DCH_IYY, TRUE},
	{"IY", 2, dch_date, DCH_IY, TRUE},
	{"I", 1, dch_date, DCH_I, TRUE},
	{"J", 1, dch_date, DCH_J, TRUE},	/* J */
	{"MI", 2, dch_time, DCH_MI, TRUE},
	{"MM", 2, dch_date, DCH_MM, TRUE},
	{"MONTH", 5, dch_date, DCH_MONTH, FALSE},
	{"MON", 3, dch_date, DCH_MON, FALSE},
	{"MS", 2, dch_time, DCH_MS, TRUE},
	{"Month", 5, dch_date, DCH_Month, FALSE},
	{"Mon", 3, dch_date, DCH_Mon, FALSE},
	{"P.M.", 4, dch_time, DCH_P_M, FALSE},		/* P */
	{"PM", 2, dch_time, DCH_PM, FALSE},
	{"Q", 1, dch_date, DCH_Q, TRUE},	/* Q */
	{"RM", 2, dch_date, DCH_RM, FALSE}, /* R */
	{"SSSS", 4, dch_time, DCH_SSSS, TRUE},		/* S */
	{"SS", 2, dch_time, DCH_SS, TRUE},
	{"TZ", 2, dch_time, DCH_TZ, FALSE}, /* T */
	{"US", 2, dch_time, DCH_US, TRUE},	/* U */
	{"WW", 2, dch_date, DCH_WW, TRUE},	/* W */
	{"W", 1, dch_date, DCH_W, TRUE},
	{"Y,YYY", 5, dch_date, DCH_Y_YYY, TRUE},	/* Y */
	{"YYYY", 4, dch_date, DCH_YYYY, TRUE},
	{"YYY", 3, dch_date, DCH_YYY, TRUE},
	{"YY", 2, dch_date, DCH_YY, TRUE},
	{"Y", 1, dch_date, DCH_Y, TRUE},
	{"a.d.", 4, dch_date, DCH_a_d, FALSE},		/* a */
	{"a.m.", 4, dch_time, DCH_a_m, FALSE},
	{"ad", 2, dch_date, DCH_ad, FALSE},
	{"am", 2, dch_time, DCH_am, FALSE},
	{"b.c.", 4, dch_date, DCH_b_c, FALSE},		/* b */
	{"bc", 2, dch_date, DCH_bc, FALSE},
	{"cc", 2, dch_date, DCH_CC, TRUE},	/* c */
	{"day", 3, dch_date, DCH_day, FALSE},		/* d */
	{"ddd", 3, dch_date, DCH_DDD, TRUE},
	{"dd", 2, dch_date, DCH_DD, TRUE},
	{"dy", 2, dch_date, DCH_dy, FALSE},
	{"d", 1, dch_date, DCH_D, TRUE},
	{"fx", 2, dch_global, DCH_FX, FALSE},		/* f */
	{"hh24", 4, dch_time, DCH_HH24, TRUE},		/* h */
	{"hh12", 4, dch_time, DCH_HH12, TRUE},
	{"hh", 2, dch_time, DCH_HH, TRUE},
	{"iw", 2, dch_date, DCH_IW, TRUE},	/* i */
	{"iyyy", 4, dch_date, DCH_IYYY, TRUE},
	{"iyy", 3, dch_date, DCH_IYY, TRUE},
	{"iy", 2, dch_date, DCH_IY, TRUE},
	{"i", 1, dch_date, DCH_I, TRUE},
	{"j", 1, dch_time, DCH_J, TRUE},	/* j */
	{"mi", 2, dch_time, DCH_MI, TRUE},	/* m */
	{"mm", 2, dch_date, DCH_MM, TRUE},
	{"month", 5, dch_date, DCH_month, FALSE},
	{"mon", 3, dch_date, DCH_mon, FALSE},
	{"ms", 2, dch_time, DCH_MS, TRUE},
	{"p.m.", 4, dch_time, DCH_p_m, FALSE},		/* p */
	{"pm", 2, dch_time, DCH_pm, FALSE},
	{"q", 1, dch_date, DCH_Q, TRUE},	/* q */
	{"rm", 2, dch_date, DCH_rm, FALSE}, /* r */
	{"ssss", 4, dch_time, DCH_SSSS, TRUE},		/* s */
	{"ss", 2, dch_time, DCH_SS, TRUE},
	{"tz", 2, dch_time, DCH_tz, FALSE}, /* t */
	{"us", 2, dch_time, DCH_US, TRUE},	/* u */
	{"ww", 2, dch_date, DCH_WW, TRUE},	/* w */
	{"w", 1, dch_date, DCH_W, TRUE},
	{"y,yyy", 5, dch_date, DCH_Y_YYY, TRUE},	/* y */
	{"yyyy", 4, dch_date, DCH_YYYY, TRUE},
	{"yyy", 3, dch_date, DCH_YYY, TRUE},
	{"yy", 2, dch_date, DCH_YY, TRUE},
	{"y", 1, dch_date, DCH_Y, TRUE},
/* last */
{NULL, 0, NULL, 0}};

/* ----------
 * KeyWords for NUMBER version (now, isitdigit info is not needful here..)
 * ----------
 */
static const KeyWord NUM_keywords[] = {
/*	keyword,	len, func.	type			   is in Index */
	{",", 1, NULL, NUM_COMMA},	/* , */
	{".", 1, NULL, NUM_DEC},	/* . */
	{"0", 1, NULL, NUM_0},		/* 0 */
	{"9", 1, NULL, NUM_9},		/* 9 */
	{"B", 1, NULL, NUM_B},		/* B */
	{"C", 1, NULL, NUM_C},		/* C */
	{"D", 1, NULL, NUM_D},		/* D */
	{"E", 1, NULL, NUM_E},		/* E */
	{"FM", 2, NULL, NUM_FM},	/* F */
	{"G", 1, NULL, NUM_G},		/* G */
	{"L", 1, NULL, NUM_L},		/* L */
	{"MI", 2, NULL, NUM_MI},	/* M */
	{"PL", 2, NULL, NUM_PL},	/* P */
	{"PR", 2, NULL, NUM_PR},
	{"RN", 2, NULL, NUM_RN},	/* R */
	{"SG", 2, NULL, NUM_SG},	/* S */
	{"SP", 2, NULL, NUM_SP},
	{"S", 1, NULL, NUM_S},
	{"TH", 2, NULL, NUM_TH},	/* T */
	{"V", 1, NULL, NUM_V},		/* V */
	{"b", 1, NULL, NUM_B},		/* b */
	{"c", 1, NULL, NUM_C},		/* c */
	{"d", 1, NULL, NUM_D},		/* d */
	{"e", 1, NULL, NUM_E},		/* e */
	{"fm", 2, NULL, NUM_FM},	/* f */
	{"g", 1, NULL, NUM_G},		/* g */
	{"l", 1, NULL, NUM_L},		/* l */
	{"mi", 2, NULL, NUM_MI},	/* m */
	{"pl", 2, NULL, NUM_PL},	/* p */
	{"pr", 2, NULL, NUM_PR},
	{"rn", 2, NULL, NUM_rn},	/* r */
	{"sg", 2, NULL, NUM_SG},	/* s */
	{"sp", 2, NULL, NUM_SP},
	{"s", 1, NULL, NUM_S},
	{"th", 2, NULL, NUM_th},	/* t */
	{"v", 1, NULL, NUM_V},		/* v */

/* last */
{NULL, 0, NULL, 0}};


/* ----------
 * KeyWords index for DATE-TIME version
 * ----------
 */
static const int DCH_index[KeyWord_INDEX_SIZE] = {
/*
0	1	2	3	4	5	6	7	8	9
*/
	/*---- first 0..31 chars are skipped ----*/

	-1, -1, -1, -1, -1, -1, -1, -1,
	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
	-1, -1, -1, -1, -1, DCH_A_D, DCH_B_C, DCH_CC, DCH_DAY, -1,
	DCH_FX, -1, DCH_HH24, DCH_IW, DCH_J, -1, -1, DCH_MI, -1, -1,
	DCH_P_M, DCH_Q, DCH_RM, DCH_SSSS, DCH_TZ, DCH_US, -1, DCH_WW, -1, DCH_Y_YYY,
	-1, -1, -1, -1, -1, -1, -1, DCH_a_d, DCH_b_c, DCH_cc,
	DCH_day, -1, DCH_fx, -1, DCH_hh24, DCH_iw, DCH_j, -1, -1, DCH_mi,
	-1, -1, DCH_p_m, DCH_q, DCH_rm, DCH_ssss, DCH_tz, DCH_us, -1, DCH_ww,
	-1, DCH_y_yyy, -1, -1, -1, -1

	/*---- chars over 126 are skiped ----*/
};

/* ----------
 * KeyWords index for NUMBER version
 * ----------
 */
static const int NUM_index[KeyWord_INDEX_SIZE] = {
/*
0	1	2	3	4	5	6	7	8	9
*/
	/*---- first 0..31 chars are skiped ----*/

	-1, -1, -1, -1, -1, -1, -1, -1,
	-1, -1, -1, -1, NUM_COMMA, -1, NUM_DEC, -1, NUM_0, -1,
	-1, -1, -1, -1, -1, -1, -1, NUM_9, -1, -1,
	-1, -1, -1, -1, -1, -1, NUM_B, NUM_C, NUM_D, NUM_E,
	NUM_FM, NUM_G, -1, -1, -1, -1, NUM_L, NUM_MI, -1, -1,
	NUM_PL, -1, NUM_RN, NUM_SG, NUM_TH, -1, NUM_V, -1, -1, -1,
	-1, -1, -1, -1, -1, -1, -1, -1, NUM_b, NUM_c,
	NUM_d, NUM_e, NUM_fm, NUM_g, -1, -1, -1, -1, NUM_l, NUM_mi,
	-1, -1, NUM_pl, -1, NUM_rn, NUM_sg, NUM_th, -1, NUM_v, -1,
	-1, -1, -1, -1, -1, -1

	/*---- chars over 126 are skiped ----*/
};

/* ----------
 * Number processor struct
 * ----------
 */
typedef struct NUMProc
{
	bool		is_to_char;
	NUMDesc    *Num;			/* number description		*/

	int			sign,			/* '-' or '+'			*/
				sign_wrote,		/* was sign write		*/
				num_count,		/* number of write digits	*/
				num_in,			/* is inside number		*/
				num_curr,		/* current position in number	*/
				num_pre,		/* space before first number	*/

				read_dec,		/* to_number - was read dec. point	*/
				read_post,		/* to_number - number of dec. digit */
				read_pre;		/* to_number - number non-dec. digit */

	char	   *number,			/* string with number	*/
			   *number_p,		/* pointer to current number position */
			   *inout,			/* in / out buffer	*/
			   *inout_p,		/* pointer to current inout position */
			   *last_relevant,	/* last relevant number after decimal point */

			   *L_negative_sign,	/* Locale */
			   *L_positive_sign,
			   *decimal,
			   *L_thousands_sep,
			   *L_currency_symbol;
} NUMProc;


/* ----------
 * Functions
 * ----------
 */
static const KeyWord *index_seq_search(char *str, const KeyWord *kw,
				 const int *index);
static KeySuffix *suff_search(char *str, KeySuffix *suf, int type);
static void NUMDesc_prepare(NUMDesc *num, FormatNode *n);
static void parse_format(FormatNode *node, char *str, const KeyWord *kw,
			 KeySuffix *suf, const int *index, int ver, NUMDesc *Num);
static char *DCH_processor(FormatNode *node, char *inout, bool is_to_char,
			  bool is_interval, void *data);

#ifdef DEBUG_TO_FROM_CHAR
static void dump_index(const KeyWord *k, const int *index);
static void dump_node(FormatNode *node, int max);
#endif

static char *get_th(char *num, int type);
static char *str_numth(char *dest, char *num, int type);
static int	strdigits_len(char *str);
static char *str_toupper(char *buff);
static char *str_tolower(char *buff);

/* static int is_acdc(char *str, int *len); */
static int	seq_search(char *name, char **array, int type, int max, int *len);
static void do_to_timestamp(text *date_txt, text *fmt,
				struct pg_tm * tm, fsec_t *fsec);
static char *fill_str(char *str, int c, int max);
static FormatNode *NUM_cache(int len, NUMDesc *Num, char *pars_str, bool *shouldFree);
static char *int_to_roman(int number);
static void NUM_prepare_locale(NUMProc *Np);
static char *get_last_relevant_decnum(char *num);
static void NUM_numpart_from_char(NUMProc *Np, int id, int plen);
static void NUM_numpart_to_char(NUMProc *Np, int id);
static char *NUM_processor(FormatNode *node, NUMDesc *Num, char *inout, char *number,
			  int plen, int sign, bool is_to_char);
static DCHCacheEntry *DCH_cache_search(char *str);
static DCHCacheEntry *DCH_cache_getnew(char *str);

static NUMCacheEntry *NUM_cache_search(char *str);
static NUMCacheEntry *NUM_cache_getnew(char *str);
static void NUM_cache_remove(NUMCacheEntry *ent);


/* ----------
 * Fast sequential search, use index for data selection which
 * go to seq. cycle (it is very fast for unwanted strings)
 * (can't be used binary search in format parsing)
 * ----------
 */
static const KeyWord *
index_seq_search(char *str, const KeyWord *kw, const int *index)
{
	int			poz;

	if (!KeyWord_INDEX_FILTER(*str))
		return NULL;

	if ((poz = *(index + (*str - ' '))) > -1)
	{
		const KeyWord *k = kw + poz;

		do
		{
			if (!strncmp(str, k->name, k->len))
				return k;
			k++;
			if (!k->name)
				return NULL;
		} while (*str == *k->name);
	}
	return NULL;
}

static KeySuffix *
suff_search(char *str, KeySuffix *suf, int type)
{
	KeySuffix  *s;

	for (s = suf; s->name != NULL; s++)
	{
		if (s->type != type)
			continue;

		if (!strncmp(str, s->name, s->len))
			return s;
	}
	return NULL;
}

/* ----------
 * Prepare NUMDesc (number description struct) via FormatNode struct
 * ----------
 */
static void
NUMDesc_prepare(NUMDesc *num, FormatNode *n)
{

	if (n->type != NODE_TYPE_ACTION)
		return;

	switch (n->key->id)
	{
		case NUM_9:
			if (IS_BRACKET(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("\"9\" must be ahead of \"PR\"")));
			}
			if (IS_MULTI(num))
			{
				++num->multi;
				break;
			}
			if (IS_DECIMAL(num))
				++num->post;
			else
				++num->pre;
			break;

		case NUM_0:
			if (IS_BRACKET(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("\"0\" must be ahead of \"PR\"")));
			}
			if (!IS_ZERO(num) && !IS_DECIMAL(num))
			{
				num->flag |= NUM_F_ZERO;
				num->zero_start = num->pre + 1;
			}
			if (!IS_DECIMAL(num))
				++num->pre;
			else
				++num->post;

			num->zero_end = num->pre + num->post;
			break;

		case NUM_B:
			if (num->pre == 0 && num->post == 0 && (!IS_ZERO(num)))
				num->flag |= NUM_F_BLANK;
			break;

		case NUM_D:
			num->flag |= NUM_F_LDECIMAL;
			num->need_locale = TRUE;
		case NUM_DEC:
			if (IS_DECIMAL(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("multiple decimal points")));
			}
			if (IS_MULTI(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
					 errmsg("cannot use \"V\" and decimal point together")));
			}
			num->flag |= NUM_F_DECIMAL;
			break;

		case NUM_FM:
			num->flag |= NUM_F_FILLMODE;
			break;

		case NUM_S:
			if (IS_LSIGN(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("not unique \"S\"")));
			}
			if (IS_PLUS(num) || IS_MINUS(num) || IS_BRACKET(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("cannot use \"S\" and \"PL\"/\"MI\"/\"SG\"/\"PR\" together")));
			}
			if (!IS_DECIMAL(num))
			{
				num->lsign = NUM_LSIGN_PRE;
				num->pre_lsign_num = num->pre;
				num->need_locale = TRUE;
				num->flag |= NUM_F_LSIGN;

			}
			else if (num->lsign == NUM_LSIGN_NONE)
			{
				num->lsign = NUM_LSIGN_POST;
				num->need_locale = TRUE;
				num->flag |= NUM_F_LSIGN;
			}
			break;

		case NUM_MI:
			if (IS_LSIGN(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("cannot use \"S\" and \"MI\" together")));
			}
			num->flag |= NUM_F_MINUS;
			if (IS_DECIMAL(num))
				num->flag |= NUM_F_MINUS_POST;
			break;

		case NUM_PL:
			if (IS_LSIGN(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("cannot use \"S\" and \"PL\" together")));
			}
			num->flag |= NUM_F_PLUS;
			if (IS_DECIMAL(num))
				num->flag |= NUM_F_PLUS_POST;
			break;

		case NUM_SG:
			if (IS_LSIGN(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("cannot use \"S\" and \"SG\" together")));
			}
			num->flag |= NUM_F_MINUS;
			num->flag |= NUM_F_PLUS;
			break;

		case NUM_PR:
			if (IS_LSIGN(num) || IS_PLUS(num) || IS_MINUS(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
						 errmsg("cannot use \"PR\" and \"S\"/\"PL\"/\"MI\"/\"SG\" together")));
			}
			num->flag |= NUM_F_BRACKET;
			break;

		case NUM_rn:
		case NUM_RN:
			num->flag |= NUM_F_ROMAN;
			break;

		case NUM_L:
		case NUM_G:
			num->need_locale = TRUE;
			break;

		case NUM_V:
			if (IS_DECIMAL(num))
			{
				NUM_cache_remove(last_NUMCacheEntry);
				ereport(ERROR,
						(errcode(ERRCODE_SYNTAX_ERROR),
					 errmsg("cannot use \"V\" and decimal point together")));
			}
			num->flag |= NUM_F_MULTI;
			break;

		case NUM_E:
			NUM_cache_remove(last_NUMCacheEntry);
			ereport(ERROR,
					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
					 errmsg("\"E\" is not supported")));
	}