#include "postgres.h"
#include <ctype.h>
#include <limits.h>
#include <math.h>
#include "access/htup_details.h"
#include "access/xact.h"
#include "catalog/pg_type.h"
#include "common/int.h"
#include "common/string.h"
#include "funcapi.h"
#include "miscadmin.h"
#include "nodes/nodeFuncs.h"
#include "parser/scansup.h"
#include "utils/builtins.h"
#include "utils/date.h"
#include "utils/datetime.h"
#include "utils/guc.h"
#include "utils/memutils.h"
#include "utils/tzparser.h"

Include dependency graph for datetime.c:

Macros
#define	APPEND_CHAR(bufptr, end, newchar)

Functions
static int	DecodeNumber (int flen, char str, bool haveTextMonth, int fmask, int tmask, struct pg_tm tm, fsec_t fsec, bool *is2digits)

static int	DecodeNumberField (int len, char str, int fmask, int tmask, struct pg_tm tm, fsec_t fsec, bool *is2digits)

static int	DecodeTimeCommon (char str, int fmask, int range, int tmask, struct pg_itm *itm)

static int	DecodeTime (char str, int fmask, int range, int tmask, struct pg_tm tm, fsec_t fsec)

static int	DecodeTimeForInterval (char str, int fmask, int range, int tmask, struct pg_itm_in *itm_in)

static const datetkn *	datebsearch (const char key, const datetkn base, int nel)

static int	DecodeDate (char str, int fmask, int tmask, bool is2digits, struct pg_tm tm)

static char *	AppendSeconds (char *cp, int sec, fsec_t fsec, int precision, bool fillzeros)

static bool	int64_multiply_add (int64 val, int64 multiplier, int64 *sum)

static bool	AdjustFractMicroseconds (double frac, int64 scale, struct pg_itm_in *itm_in)

static bool	AdjustFractDays (double frac, int scale, struct pg_itm_in *itm_in)

static bool	AdjustFractYears (double frac, int scale, struct pg_itm_in *itm_in)

static bool	AdjustMicroseconds (int64 val, double fval, int64 scale, struct pg_itm_in *itm_in)

static bool	AdjustDays (int64 val, int scale, struct pg_itm_in *itm_in)

static bool	AdjustMonths (int64 val, struct pg_itm_in *itm_in)

static bool	AdjustYears (int64 val, int scale, struct pg_itm_in *itm_in)

static int	DetermineTimeZoneOffsetInternal (struct pg_tm tm, pg_tz tzp, pg_time_t *tp)

static bool	DetermineTimeZoneAbbrevOffsetInternal (pg_time_t t, const char abbr, pg_tz tzp, int offset, int isdst)

static pg_tz *	FetchDynamicTimeZone (TimeZoneAbbrevTable tbl, const datetkn tp, DateTimeErrorExtra *extra)

int	date2j (int year, int month, int day)

void	j2date (int jd, int year, int month, int *day)

int	j2day (int date)

void	GetCurrentDateTime (struct pg_tm *tm)

void	GetCurrentTimeUsec (struct pg_tm tm, fsec_t fsec, int *tzp)

static char *	AppendTimestampSeconds (char cp, struct pg_tm tm, fsec_t fsec)

static int	ParseFraction (char cp, double frac)

static int	ParseFractionalSecond (char cp, fsec_t fsec)

int	ParseDateTime (const char timestr, char workbuf, size_t buflen, char *field, int ftype, int maxfields, int *numfields)

int	DecodeDateTime (char *field, int ftype, int nf, int dtype, struct pg_tm tm, fsec_t fsec, int tzp, DateTimeErrorExtra *extra)

int	DetermineTimeZoneOffset (struct pg_tm tm, pg_tz tzp)

int	DetermineTimeZoneAbbrevOffset (struct pg_tm tm, const char abbr, pg_tz *tzp)

int	DetermineTimeZoneAbbrevOffsetTS (TimestampTz ts, const char abbr, pg_tz tzp, int *isdst)

int	DecodeTimeOnly (char *field, int ftype, int nf, int dtype, struct pg_tm tm, fsec_t fsec, int tzp, DateTimeErrorExtra *extra)

int	ValidateDate (int fmask, bool isjulian, bool is2digits, bool bc, struct pg_tm *tm)

int	DecodeTimezone (const char str, int tzp)

int	DecodeTimezoneAbbrev (int field, const char lowtoken, int ftype, int offset, pg_tz tz, DateTimeErrorExtra extra)

int	DecodeSpecial (int field, const char lowtoken, int val)

int	DecodeTimezoneName (const char tzname, int offset, pg_tz **tz)

pg_tz *	DecodeTimezoneNameToTz (const char *tzname)

static void	ClearPgItmIn (struct pg_itm_in *itm_in)

int	DecodeInterval (char *field, int ftype, int nf, int range, int dtype, struct pg_itm_in itm_in)

static int	ParseISO8601Number (char str, char endptr, int64 ipart, double *fpart)

static int	ISO8601IntegerWidth (char *fieldstart)

int	DecodeISO8601Interval (char str, int dtype, struct pg_itm_in *itm_in)

int	DecodeUnits (int field, const char lowtoken, int val)

void	DateTimeParseError (int dterr, DateTimeErrorExtra extra, const char str, const char datatype, Node escontext)

static char *	EncodeTimezone (char *str, int tz, int style)

void	EncodeDateOnly (struct pg_tm tm, int style, char str)

void	EncodeTimeOnly (struct pg_tm tm, fsec_t fsec, bool print_tz, int tz, int style, char str)

void	EncodeDateTime (struct pg_tm tm, fsec_t fsec, bool print_tz, int tz, const char tzn, int style, char *str)

static char *	AddISO8601IntPart (char *cp, int64 value, char units)

static char *	AddPostgresIntPart (char cp, int64 value, const char units, bool is_zero, bool is_before)

static char *	AddVerboseIntPart (char cp, int64 value, const char units, bool is_zero, bool is_before)

void	EncodeInterval (struct pg_itm itm, int style, char str)

static bool	CheckDateTokenTable (const char tablename, const datetkn base, int nel)

bool	CheckDateTokenTables (void)

Node *	TemporalSimplify (int32 max_precis, Node *node)

TimeZoneAbbrevTable *	ConvertTimeZoneAbbrevs (struct tzEntry *abbrevs, int n)

void	InstallTimeZoneAbbrevs (TimeZoneAbbrevTable *tbl)

Datum	pg_timezone_abbrevs (PG_FUNCTION_ARGS)

Datum	pg_timezone_names (PG_FUNCTION_ARGS)

Variables
const int	day_tab [2][13]

const char *const	months []

const char *const	days []

static const datetkn	datetktbl []

static const int	szdatetktbl = sizeof datetktbl / sizeof datetktbl[0]

static const datetkn	deltatktbl []

static const int	szdeltatktbl = sizeof deltatktbl / sizeof deltatktbl[0]

static TimeZoneAbbrevTable *	zoneabbrevtbl = NULL

static const datetkn *	datecache [MAXDATEFIELDS] = {NULL}

static const datetkn *	deltacache [MAXDATEFIELDS] = {NULL}

static const datetkn *	abbrevcache [MAXDATEFIELDS] = {NULL}

Macro Definition Documentation

◆ APPEND_CHAR

#define APPEND_CHAR	(	bufptr,
		end,
		newchar
	)

Value:

    do                                          \
    {                                           \
        if (((bufptr) + 1) >= (end))            \
            return DTERR_BAD_FORMAT;            \
        *(bufptr)++ = newchar;                  \
    } while (0)

Function Documentation

◆ AddISO8601IntPart()

static char* AddISO8601IntPart	(	char *	cp,
		int64	value,
		char	units
	)

static

Definition at line 4402 of file datetime.c.

 {
     if (value == 0)
         return cp;
     sprintf(cp, "%lld%c", (long long) value, units);
     return cp + strlen(cp);
 }

References sprintf, and value.

Referenced by EncodeInterval().

◆ AddPostgresIntPart()

static char* AddPostgresIntPart	(	char *	cp,
		int64	value,
		const char *	units,
		bool *	is_zero,
		bool *	is_before
	)

static

Definition at line 4412 of file datetime.c.

 {
     if (value == 0)
         return cp;
     sprintf(cp, "%s%s%lld %s%s",
             (!*is_zero) ? " " : "",
             (*is_before && value > 0) ? "+" : "",
             (long long) value,
             units,
             (value != 1) ? "s" : "");
  
     /*
      * Each nonzero field sets is_before for (only) the next one.  This is a
      * tad bizarre but it's how it worked before...
      */
     *is_before = (value < 0);
     *is_zero = false;
     return cp + strlen(cp);
 }

References sprintf, and value.

Referenced by EncodeInterval().

◆ AddVerboseIntPart()

static char* AddVerboseIntPart	(	char *	cp,
		int64	value,
		const char *	units,
		bool *	is_zero,
		bool *	is_before
	)

static

Definition at line 4435 of file datetime.c.

 {
     if (value == 0)
         return cp;
     /* first nonzero value sets is_before */
     if (*is_zero)
     {
         *is_before = (value < 0);
         value = i64abs(value);
     }
     else if (*is_before)
         value = -value;
     sprintf(cp, " %lld %s%s", (long long) value, units, (value == 1) ? "" : "s");
     *is_zero = false;
     return cp + strlen(cp);
 }

References i64abs, sprintf, and value.

Referenced by EncodeInterval().

◆ AdjustDays()

static bool AdjustDays	(	int64	val,
		int	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 635 of file datetime.c.

 {
     int         days;
  
     if (val < INT_MIN || val > INT_MAX)
         return false;
     return !pg_mul_s32_overflow((int32) val, scale, &days) &&
         !pg_add_s32_overflow(itm_in->tm_mday, days, &itm_in->tm_mday);
 }

References days, pg_add_s32_overflow(), pg_mul_s32_overflow(), scale, pg_itm_in::tm_mday, and val.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustFractDays()

static bool AdjustFractDays	(	double	frac,
		int	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 571 of file datetime.c.

 {
     int         extra_days;
  
     /* Fast path for common case */
     if (frac == 0)
         return true;
  
     /*
      * We assume the input frac has abs value less than 1, so overflow of frac
      * or extra_days is not an issue.
      */
     frac *= scale;
     extra_days = (int) frac;
  
     /* ... but this could overflow, if tm_mday is already nonzero */
     if (pg_add_s32_overflow(itm_in->tm_mday, extra_days, &itm_in->tm_mday))
         return false;
  
     /* Handle any fractional day */
     frac -= extra_days;
     return AdjustFractMicroseconds(frac, USECS_PER_DAY, itm_in);
 }

References AdjustFractMicroseconds(), pg_add_s32_overflow(), scale, pg_itm_in::tm_mday, and USECS_PER_DAY.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustFractMicroseconds()

static bool AdjustFractMicroseconds	(	double	frac,
		int64	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 539 of file datetime.c.

 {
     int64       usec;
  
     /* Fast path for common case */
     if (frac == 0)
         return true;
  
     /*
      * We assume the input frac has abs value less than 1, so overflow of frac
      * or usec is not an issue for interesting values of scale.
      */
     frac *= scale;
     usec = (int64) frac;
  
     /* Round off any fractional microsecond */
     frac -= usec;
     if (frac > 0.5)
         usec++;
     else if (frac < -0.5)
         usec--;
  
     return !pg_add_s64_overflow(itm_in->tm_usec, usec, &itm_in->tm_usec);
 }

References pg_add_s64_overflow(), scale, and pg_itm_in::tm_usec.

Referenced by AdjustFractDays(), AdjustMicroseconds(), DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustFractYears()

static bool AdjustFractYears	(	double	frac,
		int	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 603 of file datetime.c.

 {
     /*
      * As above, we assume abs(frac) < 1, so this can't overflow for any
      * interesting value of scale.
      */
     int         extra_months = (int) rint(frac * scale * MONTHS_PER_YEAR);
  
     return !pg_add_s32_overflow(itm_in->tm_mon, extra_months, &itm_in->tm_mon);
 }

References MONTHS_PER_YEAR, pg_add_s32_overflow(), scale, and pg_itm_in::tm_mon.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustMicroseconds()

static bool AdjustMicroseconds	(	int64	val,
		double	fval,
		int64	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 620 of file datetime.c.

 {
     /* Handle the integer part */
     if (!int64_multiply_add(val, scale, &itm_in->tm_usec))
         return false;
     /* Handle the float part */
     return AdjustFractMicroseconds(fval, scale, itm_in);
 }

References AdjustFractMicroseconds(), int64_multiply_add(), scale, pg_itm_in::tm_usec, and val.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustMonths()

static bool AdjustMonths	(	int64	val,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 651 of file datetime.c.

 {
     if (val < INT_MIN || val > INT_MAX)
         return false;
     return !pg_add_s32_overflow(itm_in->tm_mon, (int32) val, &itm_in->tm_mon);
 }

References pg_add_s32_overflow(), pg_itm_in::tm_mon, and val.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AdjustYears()

static bool AdjustYears	(	int64	val,
		int	scale,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 663 of file datetime.c.

 {
     int         years;
  
     if (val < INT_MIN || val > INT_MAX)
         return false;
     return !pg_mul_s32_overflow((int32) val, scale, &years) &&
         !pg_add_s32_overflow(itm_in->tm_year, years, &itm_in->tm_year);
 }

References pg_add_s32_overflow(), pg_mul_s32_overflow(), scale, pg_itm_in::tm_year, and val.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ AppendSeconds()

static char * AppendSeconds	(	char *	cp,
		int	sec,
		fsec_t	fsec,
		int	precision,
		bool	fillzeros
	)

static

Definition at line 450 of file datetime.c.

 {
     Assert(precision >= 0);
  
     if (fillzeros)
         cp = pg_ultostr_zeropad(cp, abs(sec), 2);
     else
         cp = pg_ultostr(cp, abs(sec));
  
     /* fsec_t is just an int32 */
     if (fsec != 0)
     {
         int32       value = abs(fsec);
         char       *end = &cp[precision + 1];
         bool        gotnonzero = false;
  
         *cp++ = '.';
  
         /*
          * Append the fractional seconds part.  Note that we don't want any
          * trailing zeros here, so since we're building the number in reverse
          * we'll skip appending zeros until we've output a non-zero digit.
          */
         while (precision--)
         {
             int32       oldval = value;
             int32       remainder;
  
             value /= 10;
             remainder = oldval - value * 10;
  
             /* check if we got a non-zero */
             if (remainder)
                 gotnonzero = true;
  
             if (gotnonzero)
                 cp[precision] = '0' + remainder;
             else
                 end = &cp[precision];
         }
  
         /*
          * If we still have a non-zero value then precision must have not been
          * enough to print the number.  We punt the problem to pg_ultostr(),
          * which will generate a correct answer in the minimum valid width.
          */
         if (value)
             return pg_ultostr(cp, abs(fsec));
  
         return end;
     }
     else
         return cp;
 }

References Assert(), pg_ultostr(), pg_ultostr_zeropad(), and value.

Referenced by AppendTimestampSeconds(), EncodeInterval(), and EncodeTimeOnly().

◆ AppendTimestampSeconds()

static char* AppendTimestampSeconds	(	char *	cp,
		struct pg_tm *	tm,
		fsec_t	fsec
	)

static

Definition at line 513 of file datetime.c.

 {
     return AppendSeconds(cp, tm->tm_sec, fsec, MAX_TIMESTAMP_PRECISION, true);
 }

References AppendSeconds(), MAX_TIMESTAMP_PRECISION, tm, and pg_tm::tm_sec.

Referenced by EncodeDateTime().

◆ CheckDateTokenTable()

static bool CheckDateTokenTable	(	const char *	tablename,
		const datetkn *	base,
		int	nel
	)

static

Definition at line 4668 of file datetime.c.

 {
     bool        ok = true;
     int         i;
  
     for (i = 0; i < nel; i++)
     {
         /* check for token strings that don't fit */
         if (strlen(base[i].token) > TOKMAXLEN)
         {
             /* %.*s is safe since all our tokens are ASCII */
             elog(LOG, "token too long in %s table: \"%.*s\"",
                  tablename,
                  TOKMAXLEN + 1, base[i].token);
             ok = false;
             break;              /* don't risk applying strcmp */
         }
         /* check for out of order */
         if (i > 0 &&
             strcmp(base[i - 1].token, base[i].token) >= 0)
         {
             elog(LOG, "ordering error in %s table: \"%s\" >= \"%s\"",
                  tablename,
                  base[i - 1].token,
                  base[i].token);
             ok = false;
         }
     }
     return ok;
 }

References elog(), i, LOG, token, and TOKMAXLEN.

Referenced by CheckDateTokenTables(), and ConvertTimeZoneAbbrevs().

◆ CheckDateTokenTables()

bool CheckDateTokenTables ( void )

Definition at line 4700 of file datetime.c.

 {
     bool        ok = true;
  
     Assert(UNIX_EPOCH_JDATE == date2j(1970, 1, 1));
     Assert(POSTGRES_EPOCH_JDATE == date2j(2000, 1, 1));
  
     ok &= CheckDateTokenTable("datetktbl", datetktbl, szdatetktbl);
     ok &= CheckDateTokenTable("deltatktbl", deltatktbl, szdeltatktbl);
     return ok;
 }

References Assert(), CheckDateTokenTable(), date2j(), datetktbl, deltatktbl, POSTGRES_EPOCH_JDATE, szdatetktbl, szdeltatktbl, and UNIX_EPOCH_JDATE.

Referenced by PostmasterMain().

◆ ClearPgItmIn()

static void ClearPgItmIn ( struct pg_itm_in * itm_in )

inlinestatic

Definition at line 3255 of file datetime.c.

 {
     itm_in->tm_usec = 0;
     itm_in->tm_mday = 0;
     itm_in->tm_mon = 0;
     itm_in->tm_year = 0;
 }

References pg_itm_in::tm_mday, pg_itm_in::tm_mon, pg_itm_in::tm_usec, and pg_itm_in::tm_year.

Referenced by DecodeInterval(), and DecodeISO8601Interval().

◆ ConvertTimeZoneAbbrevs()

TimeZoneAbbrevTable* ConvertTimeZoneAbbrevs	(	struct tzEntry *	abbrevs,
		int	n
	)

Definition at line 4762 of file datetime.c.

 {
     TimeZoneAbbrevTable *tbl;
     Size        tbl_size;
     int         i;
  
     /* Space for fixed fields and datetkn array */
     tbl_size = offsetof(TimeZoneAbbrevTable, abbrevs) +
         n * sizeof(datetkn);
     tbl_size = MAXALIGN(tbl_size);
     /* Count up space for dynamic abbreviations */
     for (i = 0; i < n; i++)
     {
         struct tzEntry *abbr = abbrevs + i;
  
         if (abbr->zone != NULL)
         {
             Size        dsize;
  
             dsize = offsetof(DynamicZoneAbbrev, zone) +
                 strlen(abbr->zone) + 1;
             tbl_size += MAXALIGN(dsize);
         }
     }
  
     /* Alloc the result ... */
     tbl = guc_malloc(LOG, tbl_size);
     if (!tbl)
         return NULL;
  
     /* ... and fill it in */
     tbl->tblsize = tbl_size;
     tbl->numabbrevs = n;
     /* in this loop, tbl_size reprises the space calculation above */
     tbl_size = offsetof(TimeZoneAbbrevTable, abbrevs) +
         n * sizeof(datetkn);
     tbl_size = MAXALIGN(tbl_size);
     for (i = 0; i < n; i++)
     {
         struct tzEntry *abbr = abbrevs + i;
         datetkn    *dtoken = tbl->abbrevs + i;
  
         /* use strlcpy to truncate name if necessary */
         strlcpy(dtoken->token, abbr->abbrev, TOKMAXLEN + 1);
         if (abbr->zone != NULL)
         {
             /* Allocate a DynamicZoneAbbrev for this abbreviation */
             DynamicZoneAbbrev *dtza;
             Size        dsize;
  
             dtza = (DynamicZoneAbbrev *) ((char *) tbl + tbl_size);
             dtza->tz = NULL;
             strcpy(dtza->zone, abbr->zone);
  
             dtoken->type = DYNTZ;
             /* value is offset from table start to DynamicZoneAbbrev */
             dtoken->value = (int32) tbl_size;
  
             dsize = offsetof(DynamicZoneAbbrev, zone) +
                 strlen(abbr->zone) + 1;
             tbl_size += MAXALIGN(dsize);
         }
         else
         {
             dtoken->type = abbr->is_dst ? DTZ : TZ;
             dtoken->value = abbr->offset;
         }
     }
  
     /* Assert the two loops above agreed on size calculations */
     Assert(tbl->tblsize == tbl_size);
  
     /* Check the ordering, if testing */
     Assert(CheckDateTokenTable("timezone abbreviations", tbl->abbrevs, n));
  
     return tbl;
 }

References tzEntry::abbrev, TimeZoneAbbrevTable::abbrevs, Assert(), CheckDateTokenTable(), DTZ, DYNTZ, guc_malloc(), i, tzEntry::is_dst, LOG, MAXALIGN, TimeZoneAbbrevTable::numabbrevs, tzEntry::offset, strlcpy(), TimeZoneAbbrevTable::tblsize, datetkn::token, TOKMAXLEN, datetkn::type, TZ, DynamicZoneAbbrev::tz, datetkn::value, DynamicZoneAbbrev::zone, and tzEntry::zone.

Referenced by load_tzoffsets().

◆ date2j()

int date2j	(	int	year,
		int	month,
		int	day
	)

Definition at line 288 of file datetime.c.

 {
     int         julian;
     int         century;
  
     if (month > 2)
     {
         month += 1;
         year += 4800;
     }
     else
     {
         month += 13;
         year += 4799;
     }
  
     century = year / 100;
     julian = year * 365 - 32167;
     julian += year / 4 - century + century / 4;
     julian += 7834 * month / 256 + day;
  
     return julian;
 }                               /* date2j() */

Referenced by CheckDateTokenTables(), date2isoweek(), date2isoyear(), date2isoyearday(), date_in(), DCH_to_char(), DecodeDateTime(), DetermineTimeZoneOffsetInternal(), EncodeDateTime(), extract_date(), GetSQLCurrentDate(), isoweek2j(), make_date(), make_timestamp_internal(), parse_datetime(), PGTYPESdate_dayofweek(), PGTYPESdate_defmt_asc(), PGTYPESdate_fmt_asc(), PGTYPESdate_from_asc(), PGTYPESdate_julmdy(), PGTYPESdate_mdyjul(), PGTYPESdate_to_asc(), PGTYPESdate_today(), timestamp2tm(), timestamp_date(), timestamp_part_common(), timestamp_pl_interval(), timestamp_to_char(), timestamptz_date(), timestamptz_part_common(), timestamptz_pl_interval_internal(), timestamptz_to_char(), tm2timestamp(), to_date(), and ValidateDate().

◆ datebsearch()

static const datetkn * datebsearch	(	const char *	key,
		const datetkn *	base,
		int	nel
	)

static

Definition at line 4042 of file datetime.c.

 {
     if (nel > 0)
     {
         const datetkn *last = base + nel - 1,
                    *position;
         int         result;
  
         while (last >= base)
         {
             position = base + ((last - base) >> 1);
             /* precheck the first character for a bit of extra speed */
             result = (int) key[0] - (int) position->token[0];
             if (result == 0)
             {
                 /* use strncmp so that we match truncated tokens */
                 result = strncmp(key, position->token, TOKMAXLEN);
                 if (result == 0)
                     return position;
             }
             if (result < 0)
                 last = position - 1;
             else
                 base = position + 1;
         }
     }
     return NULL;
 }

References sort-test::key, and TOKMAXLEN.

Referenced by DecodeSpecial(), DecodeTimezoneAbbrev(), DecodeUnits(), and ParseDateTime().

◆ DateTimeParseError()

void DateTimeParseError	(	int	dterr,
		DateTimeErrorExtra *	extra,
		const char *	str,
		const char *	datatype,
		Node *	escontext
	)

Definition at line 3981 of file datetime.c.

 {
     switch (dterr)
     {
         case DTERR_FIELD_OVERFLOW:
             errsave(escontext,
                     (errcode(ERRCODE_DATETIME_FIELD_OVERFLOW),
                      errmsg("date/time field value out of range: \"%s\"",
                             str)));
             break;
         case DTERR_MD_FIELD_OVERFLOW:
             /* <nanny>same as above, but add hint about DateStyle</nanny> */
             errsave(escontext,
                     (errcode(ERRCODE_DATETIME_FIELD_OVERFLOW),
                      errmsg("date/time field value out of range: \"%s\"",
                             str),
                      errhint("Perhaps you need a different \"datestyle\" setting.")));
             break;
         case DTERR_INTERVAL_OVERFLOW:
             errsave(escontext,
                     (errcode(ERRCODE_INTERVAL_FIELD_OVERFLOW),
                      errmsg("interval field value out of range: \"%s\"",
                             str)));
             break;
         case DTERR_TZDISP_OVERFLOW:
             errsave(escontext,
                     (errcode(ERRCODE_INVALID_TIME_ZONE_DISPLACEMENT_VALUE),
                      errmsg("time zone displacement out of range: \"%s\"",
                             str)));
             break;
         case DTERR_BAD_TIMEZONE:
             errsave(escontext,
                     (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
                      errmsg("time zone \"%s\" not recognized",
                             extra->dtee_timezone)));
             break;
         case DTERR_BAD_ZONE_ABBREV:
             errsave(escontext,
                     (errcode(ERRCODE_CONFIG_FILE_ERROR),
                      errmsg("time zone \"%s\" not recognized",
                             extra->dtee_timezone),
                      errdetail("This time zone name appears in the configuration file for time zone abbreviation \"%s\".",
                                extra->dtee_abbrev)));
             break;
         case DTERR_BAD_FORMAT:
         default:
             errsave(escontext,
                     (errcode(ERRCODE_INVALID_DATETIME_FORMAT),
                      errmsg("invalid input syntax for type %s: \"%s\"",
                             datatype, str)));
             break;
     }
 }

References DateTimeErrorExtra::dtee_abbrev, DateTimeErrorExtra::dtee_timezone, DTERR_BAD_FORMAT, DTERR_BAD_TIMEZONE, DTERR_BAD_ZONE_ABBREV, DTERR_FIELD_OVERFLOW, DTERR_INTERVAL_OVERFLOW, DTERR_MD_FIELD_OVERFLOW, DTERR_TZDISP_OVERFLOW, errcode(), errdetail(), errhint(), errmsg(), errsave, and generate_unaccent_rules::str.

Referenced by date_in(), DecodeTimezoneName(), do_to_timestamp(), interval_in(), parse_datetime(), pg_timezone_abbrevs(), time_in(), timestamp_in(), timestamptz_in(), timetz_in(), and to_timestamp().

◆ DecodeDate()

static int DecodeDate	(	char *	str,
		int	fmask,
		int *	tmask,
		bool *	is2digits,
		struct pg_tm *	tm
	)

static

Definition at line 2389 of file datetime.c.

 {
     fsec_t      fsec;
     int         nf = 0;
     int         i,
                 len;
     int         dterr;
     bool        haveTextMonth = false;
     int         type,
                 val,
                 dmask = 0;
     char       *field[MAXDATEFIELDS];
  
     *tmask = 0;
  
     /* parse this string... */
     while (*str != '\0' && nf < MAXDATEFIELDS)
     {
         /* skip field separators */
         while (*str != '\0' && !isalnum((unsigned char) *str))
             str++;
  
         if (*str == '\0')
             return DTERR_BAD_FORMAT;    /* end of string after separator */
  
         field[nf] = str;
         if (isdigit((unsigned char) *str))
         {
             while (isdigit((unsigned char) *str))
                 str++;
         }
         else if (isalpha((unsigned char) *str))
         {
             while (isalpha((unsigned char) *str))
                 str++;
         }
  
         /* Just get rid of any non-digit, non-alpha characters... */
         if (*str != '\0')
             *str++ = '\0';
         nf++;
     }
  
     /* look first for text fields, since that will be unambiguous month */
     for (i = 0; i < nf; i++)
     {
         if (isalpha((unsigned char) *field[i]))
         {
             type = DecodeSpecial(i, field[i], &val);
             if (type == IGNORE_DTF)
                 continue;
  
             dmask = DTK_M(type);
             switch (type)
             {
                 case MONTH:
                     tm->tm_mon = val;
                     haveTextMonth = true;
                     break;
  
                 default:
                     return DTERR_BAD_FORMAT;
             }
             if (fmask & dmask)
                 return DTERR_BAD_FORMAT;
  
             fmask |= dmask;
             *tmask |= dmask;
  
             /* mark this field as being completed */
             field[i] = NULL;
         }
     }
  
     /* now pick up remaining numeric fields */
     for (i = 0; i < nf; i++)
     {
         if (field[i] == NULL)
             continue;
  
         if ((len = strlen(field[i])) <= 0)
             return DTERR_BAD_FORMAT;
  
         dterr = DecodeNumber(len, field[i], haveTextMonth, fmask,
                              &dmask, tm,
                              &fsec, is2digits);
         if (dterr)
             return dterr;
  
         if (fmask & dmask)
             return DTERR_BAD_FORMAT;
  
         fmask |= dmask;
         *tmask |= dmask;
     }
  
     if ((fmask & ~(DTK_M(DOY) | DTK_M(TZ))) != DTK_DATE_M)
         return DTERR_BAD_FORMAT;
  
     /* validation of the field values must wait until ValidateDate() */
  
     return 0;
 }

References DecodeNumber(), DecodeSpecial(), DOY, DTERR_BAD_FORMAT, DTK_DATE_M, DTK_M, i, IGNORE_DTF, len, MAXDATEFIELDS, MONTH, generate_unaccent_rules::str, tm, pg_tm::tm_mon, type, TZ, and val.

Referenced by DecodeDateTime(), and DecodeTimeOnly().

◆ DecodeDateTime()

int DecodeDateTime	(	char **	field,
		int *	ftype,
		int	nf,
		int *	dtype,
		struct pg_tm *	tm,
		fsec_t *	fsec,
		int *	tzp,
		DateTimeErrorExtra *	extra
	)

Definition at line 980 of file datetime.c.

 {
     int         fmask = 0,
                 tmask,
                 type;
     int         ptype = 0;      /* "prefix type" for ISO and Julian formats */
     int         i;
     int         val;
     int         dterr;
     int         mer = HR24;
     bool        haveTextMonth = false;
     bool        isjulian = false;
     bool        is2digits = false;
     bool        bc = false;
     pg_tz      *namedTz = NULL;
     pg_tz      *abbrevTz = NULL;
     pg_tz      *valtz;
     char       *abbrev = NULL;
     struct pg_tm cur_tm;
  
     /*
      * We'll insist on at least all of the date fields, but initialize the
      * remaining fields in case they are not set later...
      */
     *dtype = DTK_DATE;
     tm->tm_hour = 0;
     tm->tm_min = 0;
     tm->tm_sec = 0;
     *fsec = 0;
     /* don't know daylight savings time status apriori */
     tm->tm_isdst = -1;
     if (tzp != NULL)
         *tzp = 0;
  
     for (i = 0; i < nf; i++)
     {
         switch (ftype[i])
         {
             case DTK_DATE:
  
                 /*
                  * Integral julian day with attached time zone? All other
                  * forms with JD will be separated into distinct fields, so we
                  * handle just this case here.
                  */
                 if (ptype == DTK_JULIAN)
                 {
                     char       *cp;
                     int         jday;
  
                     if (tzp == NULL)
                         return DTERR_BAD_FORMAT;
  
                     errno = 0;
                     jday = strtoint(field[i], &cp, 10);
                     if (errno == ERANGE || jday < 0)
                         return DTERR_FIELD_OVERFLOW;
  
                     j2date(jday, &tm->tm_year, &tm->tm_mon, &tm->tm_mday);
                     isjulian = true;
  
                     /* Get the time zone from the end of the string */
                     dterr = DecodeTimezone(cp, tzp);
                     if (dterr)
                         return dterr;
  
                     tmask = DTK_DATE_M | DTK_TIME_M | DTK_M(TZ);
                     ptype = 0;
                     break;
                 }
  
                 /*
                  * Already have a date? Then this might be a time zone name
                  * with embedded punctuation (e.g. "America/New_York") or a
                  * run-together time with trailing time zone (e.g. hhmmss-zz).
                  * - thomas 2001-12-25
                  *
                  * We consider it a time zone if we already have month & day.
                  * This is to allow the form "mmm dd hhmmss tz year", which
                  * we've historically accepted.
                  */
                 else if (ptype != 0 ||
                          ((fmask & (DTK_M(MONTH) | DTK_M(DAY))) ==
                           (DTK_M(MONTH) | DTK_M(DAY))))
                 {
                     /* No time zone accepted? Then quit... */
                     if (tzp == NULL)
                         return DTERR_BAD_FORMAT;
  
                     if (isdigit((unsigned char) *field[i]) || ptype != 0)
                     {
                         char       *cp;
  
                         /*
                          * Allow a preceding "t" field, but no other units.
                          */
                         if (ptype != 0)
                         {
                             /* Sanity check; should not fail this test */
                             if (ptype != DTK_TIME)
                                 return DTERR_BAD_FORMAT;
                             ptype = 0;
                         }
  
                         /*
                          * Starts with a digit but we already have a time
                          * field? Then we are in trouble with a date and time
                          * already...
                          */
                         if ((fmask & DTK_TIME_M) == DTK_TIME_M)
                             return DTERR_BAD_FORMAT;
  
                         if ((cp = strchr(field[i], '-')) == NULL)
                             return DTERR_BAD_FORMAT;
  
                         /* Get the time zone from the end of the string */
                         dterr = DecodeTimezone(cp, tzp);
                         if (dterr)
                             return dterr;
                         *cp = '\0';
  
                         /*
                          * Then read the rest of the field as a concatenated
                          * time
                          */
                         dterr = DecodeNumberField(strlen(field[i]), field[i],
                                                   fmask,
                                                   &tmask, tm,
                                                   fsec, &is2digits);
                         if (dterr < 0)
                             return dterr;
  
                         /*
                          * modify tmask after returning from
                          * DecodeNumberField()
                          */
                         tmask |= DTK_M(TZ);
                     }
                     else
                     {
                         namedTz = pg_tzset(field[i]);
                         if (!namedTz)
                         {
                             extra->dtee_timezone = field[i];
                             return DTERR_BAD_TIMEZONE;
                         }
                         /* we'll apply the zone setting below */
                         tmask = DTK_M(TZ);
                     }
                 }
                 else
                 {
                     dterr = DecodeDate(field[i], fmask,
                                        &tmask, &is2digits, tm);
                     if (dterr)
                         return dterr;
                 }
                 break;
  
             case DTK_TIME:
  
                 /*
                  * This might be an ISO time following a "t" field.
                  */
                 if (ptype != 0)
                 {
                     /* Sanity check; should not fail this test */
                     if (ptype != DTK_TIME)
                         return DTERR_BAD_FORMAT;
                     ptype = 0;
                 }
                 dterr = DecodeTime(field[i], fmask, INTERVAL_FULL_RANGE,
                                    &tmask, tm, fsec);
                 if (dterr)
                     return dterr;
  
                 /* check for time overflow */
                 if (time_overflows(tm->tm_hour, tm->tm_min, tm->tm_sec,
                                    *fsec))
                     return DTERR_FIELD_OVERFLOW;
                 break;
  
             case DTK_TZ:
                 {
                     int         tz;
  
                     if (tzp == NULL)
                         return DTERR_BAD_FORMAT;
  
                     dterr = DecodeTimezone(field[i], &tz);
                     if (dterr)
                         return dterr;
                     *tzp = tz;
                     tmask = DTK_M(TZ);
                 }
                 break;
  
             case DTK_NUMBER:
  
                 /*
                  * Deal with cases where previous field labeled this one
                  */
                 if (ptype != 0)
                 {
                     char       *cp;
                     int         value;
  
                     errno = 0;
                     value = strtoint(field[i], &cp, 10);
                     if (errno == ERANGE)
                         return DTERR_FIELD_OVERFLOW;
                     if (*cp != '.' && *cp != '\0')
                         return DTERR_BAD_FORMAT;
  
                     switch (ptype)
                     {
                         case DTK_JULIAN:
                             /* previous field was a label for "julian date" */
                             if (value < 0)
                                 return DTERR_FIELD_OVERFLOW;
                             tmask = DTK_DATE_M;
                             j2date(value, &tm->tm_year, &tm->tm_mon, &tm->tm_mday);
                             isjulian = true;
  
                             /* fractional Julian Day? */
                             if (*cp == '.')
                             {
                                 double      time;
  
                                 dterr = ParseFraction(cp, &time);
                                 if (dterr)
                                     return dterr;
                                 time *= USECS_PER_DAY;
                                 dt2time(time,
                                         &tm->tm_hour, &tm->tm_min,
                                         &tm->tm_sec, fsec);
                                 tmask |= DTK_TIME_M;
                             }
                             break;
  
                         case DTK_TIME:
                             /* previous field was "t" for ISO time */
                             dterr = DecodeNumberField(strlen(field[i]), field[i],
                                                       (fmask | DTK_DATE_M),
                                                       &tmask, tm,
                                                       fsec, &is2digits);
                             if (dterr < 0)
                                 return dterr;
                             if (tmask != DTK_TIME_M)
                                 return DTERR_BAD_FORMAT;
                             break;
  
                         default:
                             return DTERR_BAD_FORMAT;
                             break;
                     }
  
                     ptype = 0;
                     *dtype = DTK_DATE;
                 }
                 else
                 {
                     char       *cp;
                     int         flen;
  
                     flen = strlen(field[i]);
                     cp = strchr(field[i], '.');
  
                     /* Embedded decimal and no date yet? */
                     if (cp != NULL && !(fmask & DTK_DATE_M))
                     {
                         dterr = DecodeDate(field[i], fmask,
                                            &tmask, &is2digits, tm);
                         if (dterr)
                             return dterr;
                     }
                     /* embedded decimal and several digits before? */
                     else if (cp != NULL && flen - strlen(cp) > 2)
                     {
                         /*
                          * Interpret as a concatenated date or time Set the
                          * type field to allow decoding other fields later.
                          * Example: 20011223 or 040506
                          */
                         dterr = DecodeNumberField(flen, field[i], fmask,
                                                   &tmask, tm,
                                                   fsec, &is2digits);
                         if (dterr < 0)
                             return dterr;
                     }
  
                     /*
                      * Is this a YMD or HMS specification, or a year number?
                      * YMD and HMS are required to be six digits or more, so
                      * if it is 5 digits, it is a year.  If it is six or more
                      * digits, we assume it is YMD or HMS unless no date and
                      * no time values have been specified.  This forces 6+
                      * digit years to be at the end of the string, or to use
                      * the ISO date specification.
                      */
                     else if (flen >= 6 && (!(fmask & DTK_DATE_M) ||
                                            !(fmask & DTK_TIME_M)))
                     {
                         dterr = DecodeNumberField(flen, field[i], fmask,
                                                   &tmask, tm,
                                                   fsec, &is2digits);
                         if (dterr < 0)
                             return dterr;
                     }
                     /* otherwise it is a single date/time field... */
                     else
                     {
                         dterr = DecodeNumber(flen, field[i],
                                              haveTextMonth, fmask,
                                              &tmask, tm,
                                              fsec, &is2digits);
                         if (dterr)
                             return dterr;
                     }
                 }
                 break;
  
             case DTK_STRING:
             case DTK_SPECIAL:
                 /* timezone abbrevs take precedence over built-in tokens */
                 dterr = DecodeTimezoneAbbrev(i, field[i],
                                              &type, &val, &valtz, extra);
                 if (dterr)
                     return dterr;
                 if (type == UNKNOWN_FIELD)
                     type = DecodeSpecial(i, field[i], &val);
                 if (type == IGNORE_DTF)
                     continue;
  
                 tmask = DTK_M(type);
                 switch (type)
                 {
                     case RESERV:
                         switch (val)
                         {
                             case DTK_NOW:
                                 tmask = (DTK_DATE_M | DTK_TIME_M | DTK_M(TZ));
                                 *dtype = DTK_DATE;
                                 GetCurrentTimeUsec(tm, fsec, tzp);
                                 break;
  
                             case DTK_YESTERDAY:
                                 tmask = DTK_DATE_M;
                                 *dtype = DTK_DATE;
                                 GetCurrentDateTime(&cur_tm);
                                 j2date(date2j(cur_tm.tm_year, cur_tm.tm_mon, cur_tm.tm_mday) - 1,
                                        &tm->tm_year, &tm->tm_mon, &tm->tm_mday);
                                 break;
  
                             case DTK_TODAY:
                                 tmask = DTK_DATE_M;
                                 *dtype = DTK_DATE;
                                 GetCurrentDateTime(&cur_tm);
                                 tm->tm_year = cur_tm.tm_year;
                                 tm->tm_mon = cur_tm.tm_mon;
                                 tm->tm_mday = cur_tm.tm_mday;
                                 break;
  
                             case DTK_TOMORROW:
                                 tmask = DTK_DATE_M;
                                 *dtype = DTK_DATE;
                                 GetCurrentDateTime(&cur_tm);
                                 j2date(date2j(cur_tm.tm_year, cur_tm.tm_mon, cur_tm.tm_mday) + 1,
                                        &tm->tm_year, &tm->tm_mon, &tm->tm_mday);
                                 break;
  
                             case DTK_ZULU:
                                 tmask = (DTK_TIME_M | DTK_M(TZ));
                                 *dtype = DTK_DATE;
                                 tm->tm_hour = 0;
                                 tm->tm_min = 0;
                                 tm->tm_sec = 0;
                                 if (tzp != NULL)
                                     *tzp = 0;
                                 break;
  
                             case DTK_EPOCH:
                             case DTK_LATE:
                             case DTK_EARLY:
                                 tmask = (DTK_DATE_M | DTK_TIME_M | DTK_M(TZ));
                                 *dtype = val;
                                 /* caller ignores tm for these dtype codes */
                                 break;
  
                             default:
                                 elog(ERROR, "unrecognized RESERV datetime token: %d",
                                      val);
                         }
  
                         break;
  
                     case MONTH:
  
                         /*
                          * already have a (numeric) month? then see if we can
                          * substitute...
                          */
                         if ((fmask & DTK_M(MONTH)) && !haveTextMonth &&
                             !(fmask & DTK_M(DAY)) && tm->tm_mon >= 1 &&
                             tm->tm_mon <= 31)
                         {
                             tm->tm_mday = tm->tm_mon;
                             tmask = DTK_M(DAY);
                         }
                         haveTextMonth = true;
                         tm->tm_mon = val;
                         break;
  
                     case DTZMOD:
  
                         /*
                          * daylight savings time modifier (solves "MET DST"
                          * syntax)
                          */
                         tmask |= DTK_M(DTZ);
                         tm->tm_isdst = 1;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp -= val;
                         break;
  
                     case DTZ:
  
                         /*
                          * set mask for TZ here _or_ check for DTZ later when
                          * getting default timezone
                          */
                         tmask |= DTK_M(TZ);
                         tm->tm_isdst = 1;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp = -val;
                         break;
  
                     case TZ:
                         tm->tm_isdst = 0;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp = -val;
                         break;
  
                     case DYNTZ:
                         tmask |= DTK_M(TZ);
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         /* we'll determine the actual offset later */
                         abbrevTz = valtz;
                         abbrev = field[i];
                         break;
  
                     case AMPM:
                         mer = val;
                         break;
  
                     case ADBC:
                         bc = (val == BC);
                         break;
  
                     case DOW:
                         tm->tm_wday = val;
                         break;
  
                     case UNITS:
                         tmask = 0;
                         /* reject consecutive unhandled units */
                         if (ptype != 0)
                             return DTERR_BAD_FORMAT;
                         ptype = val;
                         break;
  
                     case ISOTIME:
  
                         /*
                          * This is a filler field "t" indicating that the next
                          * field is time. Try to verify that this is sensible.
                          */
                         tmask = 0;
  
                         /* No preceding date? Then quit... */
                         if ((fmask & DTK_DATE_M) != DTK_DATE_M)
                             return DTERR_BAD_FORMAT;
  
                         /* reject consecutive unhandled units */
                         if (ptype != 0)
                             return DTERR_BAD_FORMAT;
                         ptype = val;
                         break;
  
                     case UNKNOWN_FIELD:
  
                         /*
                          * Before giving up and declaring error, check to see
                          * if it is an all-alpha timezone name.
                          */
                         namedTz = pg_tzset(field[i]);
                         if (!namedTz)
                             return DTERR_BAD_FORMAT;
                         /* we'll apply the zone setting below */
                         tmask = DTK_M(TZ);
                         break;
  
                     default:
                         return DTERR_BAD_FORMAT;
                 }
                 break;
  
             default:
                 return DTERR_BAD_FORMAT;
         }
  
         if (tmask & fmask)
             return DTERR_BAD_FORMAT;
         fmask |= tmask;
     }                           /* end loop over fields */
  
     /* reject if prefix type appeared and was never handled */
     if (ptype != 0)
         return DTERR_BAD_FORMAT;
  
     /* do additional checking for normal date specs (but not "infinity" etc) */
     if (*dtype == DTK_DATE)
     {
         /* do final checking/adjustment of Y/M/D fields */
         dterr = ValidateDate(fmask, isjulian, is2digits, bc, tm);
         if (dterr)
             return dterr;
  
         /* handle AM/PM */
         if (mer != HR24 && tm->tm_hour > HOURS_PER_DAY / 2)
             return DTERR_FIELD_OVERFLOW;
         if (mer == AM && tm->tm_hour == HOURS_PER_DAY / 2)
             tm->tm_hour = 0;
         else if (mer == PM && tm->tm_hour != HOURS_PER_DAY / 2)
             tm->tm_hour += HOURS_PER_DAY / 2;
  
         /* check for incomplete input */
         if ((fmask & DTK_DATE_M) != DTK_DATE_M)
         {
             if ((fmask & DTK_TIME_M) == DTK_TIME_M)
                 return 1;
             return DTERR_BAD_FORMAT;
         }
  
         /*
          * If we had a full timezone spec, compute the offset (we could not do
          * it before, because we need the date to resolve DST status).
          */
         if (namedTz != NULL)
         {
             /* daylight savings time modifier disallowed with full TZ */
             if (fmask & DTK_M(DTZMOD))
                 return DTERR_BAD_FORMAT;
  
             *tzp = DetermineTimeZoneOffset(tm, namedTz);
         }
  
         /*
          * Likewise, if we had a dynamic timezone abbreviation, resolve it
          * now.
          */
         if (abbrevTz != NULL)
         {
             /* daylight savings time modifier disallowed with dynamic TZ */
             if (fmask & DTK_M(DTZMOD))
                 return DTERR_BAD_FORMAT;
  
             *tzp = DetermineTimeZoneAbbrevOffset(tm, abbrev, abbrevTz);
         }
  
         /* timezone not specified? then use session timezone */
         if (tzp != NULL && !(fmask & DTK_M(TZ)))
         {
             /*
              * daylight savings time modifier but no standard timezone? then
              * error
              */
             if (fmask & DTK_M(DTZMOD))
                 return DTERR_BAD_FORMAT;
  
             *tzp = DetermineTimeZoneOffset(tm, session_timezone);
         }
     }
  
     return 0;
 }

References ADBC, AM, AMPM, BC, date2j(), DAY, DecodeDate(), DecodeNumber(), DecodeNumberField(), DecodeSpecial(), DecodeTime(), DecodeTimezone(), DecodeTimezoneAbbrev(), DetermineTimeZoneAbbrevOffset(), DetermineTimeZoneOffset(), DOW, dt2time(), DateTimeErrorExtra::dtee_timezone, DTERR_BAD_FORMAT, DTERR_BAD_TIMEZONE, DTERR_FIELD_OVERFLOW, DTK_DATE, DTK_DATE_M, DTK_EARLY, DTK_EPOCH, DTK_JULIAN, DTK_LATE, DTK_M, DTK_NOW, DTK_NUMBER, DTK_SPECIAL, DTK_STRING, DTK_TIME, DTK_TIME_M, DTK_TODAY, DTK_TOMORROW, DTK_TZ, DTK_YESTERDAY, DTK_ZULU, DTZ, DTZMOD, DYNTZ, elog(), ERROR, GetCurrentDateTime(), GetCurrentTimeUsec(), HOURS_PER_DAY, HR24, i, IGNORE_DTF, INTERVAL_FULL_RANGE, ISOTIME, j2date(), MONTH, ParseFraction(), pg_tzset(), PM, RESERV, session_timezone, strtoint(), time_overflows(), tm, pg_tm::tm_hour, pg_tm::tm_isdst, pg_tm::tm_mday, pg_tm::tm_min, pg_tm::tm_mon, pg_tm::tm_sec, pg_tm::tm_wday, pg_tm::tm_year, type, TZ, UNITS, UNKNOWN_FIELD, USECS_PER_DAY, val, ValidateDate(), and value.

Referenced by check_recovery_target_time(), date_in(), pg_logdir_ls_internal(), PGTYPESdate_from_asc(), PGTYPEStimestamp_from_asc(), timestamp_in(), and timestamptz_in().

◆ DecodeInterval()

int DecodeInterval	(	char **	field,
		int *	ftype,
		int	nf,
		int	range,
		int *	dtype,
		struct pg_itm_in *	itm_in
	)

Definition at line 3276 of file datetime.c.

 {
     bool        force_negative = false;
     bool        is_before = false;
     bool        parsing_unit_val = false;
     char       *cp;
     int         fmask = 0,
                 tmask,
                 type,
                 uval;
     int         i;
     int         dterr;
     int64       val;
     double      fval;
  
     *dtype = DTK_DELTA;
     type = IGNORE_DTF;
     ClearPgItmIn(itm_in);
  
     /*----------
      * The SQL standard defines the interval literal
      *   '-1 1:00:00'
      * to mean "negative 1 days and negative 1 hours", while Postgres
      * traditionally treats this as meaning "negative 1 days and positive
      * 1 hours".  In SQL_STANDARD intervalstyle, we apply the leading sign
      * to all fields if there are no other explicit signs.
      *
      * We leave the signs alone if there are additional explicit signs.
      * This protects us against misinterpreting postgres-style dump output,
      * since the postgres-style output code has always put an explicit sign on
      * all fields following a negative field.  But note that SQL-spec output
      * is ambiguous and can be misinterpreted on load!  (So it's best practice
      * to dump in postgres style, not SQL style.)
      *----------
      */
     if (IntervalStyle == INTSTYLE_SQL_STANDARD && nf > 0 && *field[0] == '-')
     {
         force_negative = true;
         /* Check for additional explicit signs */
         for (i = 1; i < nf; i++)
         {
             if (*field[i] == '-' || *field[i] == '+')
             {
                 force_negative = false;
                 break;
             }
         }
     }
  
     /* read through list backwards to pick up units before values */
     for (i = nf - 1; i >= 0; i--)
     {
         switch (ftype[i])
         {
             case DTK_TIME:
                 dterr = DecodeTimeForInterval(field[i], fmask, range,
                                               &tmask, itm_in);
                 if (dterr)
                     return dterr;
                 if (force_negative &&
                     itm_in->tm_usec > 0)
                     itm_in->tm_usec = -itm_in->tm_usec;
                 type = DTK_DAY;
                 parsing_unit_val = false;
                 break;
  
             case DTK_TZ:
  
                 /*
                  * Timezone means a token with a leading sign character and at
                  * least one digit; there could be ':', '.', '-' embedded in
                  * it as well.
                  */
                 Assert(*field[i] == '-' || *field[i] == '+');
  
                 /*
                  * Check for signed hh:mm or hh:mm:ss.  If so, process exactly
                  * like DTK_TIME case above, plus handling the sign.
                  */
                 if (strchr(field[i] + 1, ':') != NULL &&
                     DecodeTimeForInterval(field[i] + 1, fmask, range,
                                           &tmask, itm_in) == 0)
                 {
                     if (*field[i] == '-')
                     {
                         /* flip the sign on time field */
                         if (itm_in->tm_usec == PG_INT64_MIN)
                             return DTERR_FIELD_OVERFLOW;
                         itm_in->tm_usec = -itm_in->tm_usec;
                     }
  
                     if (force_negative &&
                         itm_in->tm_usec > 0)
                         itm_in->tm_usec = -itm_in->tm_usec;
  
                     /*
                      * Set the next type to be a day, if units are not
                      * specified. This handles the case of '1 +02:03' since we
                      * are reading right to left.
                      */
                     type = DTK_DAY;
                     parsing_unit_val = false;
                     break;
                 }
  
                 /*
                  * Otherwise, fall through to DTK_NUMBER case, which can
                  * handle signed float numbers and signed year-month values.
                  */
  
                 /* FALLTHROUGH */
  
             case DTK_DATE:
             case DTK_NUMBER:
                 if (type == IGNORE_DTF)
                 {
                     /* use typmod to decide what rightmost field is */
                     switch (range)
                     {
                         case INTERVAL_MASK(YEAR):
                             type = DTK_YEAR;
                             break;
                         case INTERVAL_MASK(MONTH):
                         case INTERVAL_MASK(YEAR) | INTERVAL_MASK(MONTH):
                             type = DTK_MONTH;
                             break;
                         case INTERVAL_MASK(DAY):
                             type = DTK_DAY;
                             break;
                         case INTERVAL_MASK(HOUR):
                         case INTERVAL_MASK(DAY) | INTERVAL_MASK(HOUR):
                             type = DTK_HOUR;
                             break;
                         case INTERVAL_MASK(MINUTE):
                         case INTERVAL_MASK(HOUR) | INTERVAL_MASK(MINUTE):
                         case INTERVAL_MASK(DAY) | INTERVAL_MASK(HOUR) | INTERVAL_MASK(MINUTE):
                             type = DTK_MINUTE;
                             break;
                         case INTERVAL_MASK(SECOND):
                         case INTERVAL_MASK(MINUTE) | INTERVAL_MASK(SECOND):
                         case INTERVAL_MASK(HOUR) | INTERVAL_MASK(MINUTE) | INTERVAL_MASK(SECOND):
                         case INTERVAL_MASK(DAY) | INTERVAL_MASK(HOUR) | INTERVAL_MASK(MINUTE) | INTERVAL_MASK(SECOND):
                             type = DTK_SECOND;
                             break;
                         default:
                             type = DTK_SECOND;
                             break;
                     }
                 }
  
                 errno = 0;
                 val = strtoi64(field[i], &cp, 10);
                 if (errno == ERANGE)
                     return DTERR_FIELD_OVERFLOW;
  
                 if (*cp == '-')
                 {
                     /* SQL "years-months" syntax */
                     int         val2;
  
                     val2 = strtoint(cp + 1, &cp, 10);
                     if (errno == ERANGE || val2 < 0 || val2 >= MONTHS_PER_YEAR)
                         return DTERR_FIELD_OVERFLOW;
                     if (*cp != '\0')
                         return DTERR_BAD_FORMAT;
                     type = DTK_MONTH;
                     if (*field[i] == '-')
                         val2 = -val2;
                     if (pg_mul_s64_overflow(val, MONTHS_PER_YEAR, &val))
                         return DTERR_FIELD_OVERFLOW;
                     if (pg_add_s64_overflow(val, val2, &val))
                         return DTERR_FIELD_OVERFLOW;
                     fval = 0;
                 }
                 else if (*cp == '.')
                 {
                     dterr = ParseFraction(cp, &fval);
                     if (dterr)
                         return dterr;
                     if (*field[i] == '-')
                         fval = -fval;
                 }
                 else if (*cp == '\0')
                     fval = 0;
                 else
                     return DTERR_BAD_FORMAT;
  
                 tmask = 0;      /* DTK_M(type); */
  
                 if (force_negative)
                 {
                     /* val and fval should be of same sign, but test anyway */
                     if (val > 0)
                         val = -val;
                     if (fval > 0)
                         fval = -fval;
                 }
  
                 switch (type)
                 {
                     case DTK_MICROSEC:
                         if (!AdjustMicroseconds(val, fval, 1, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(MICROSECOND);
                         break;
  
                     case DTK_MILLISEC:
                         if (!AdjustMicroseconds(val, fval, 1000, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(MILLISECOND);
                         break;
  
                     case DTK_SECOND:
                         if (!AdjustMicroseconds(val, fval, USECS_PER_SEC, itm_in))
                             return DTERR_FIELD_OVERFLOW;
  
                         /*
                          * If any subseconds were specified, consider this
                          * microsecond and millisecond input as well.
                          */
                         if (fval == 0)
                             tmask = DTK_M(SECOND);
                         else
                             tmask = DTK_ALL_SECS_M;
                         break;
  
                     case DTK_MINUTE:
                         if (!AdjustMicroseconds(val, fval, USECS_PER_MINUTE, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(MINUTE);
                         break;
  
                     case DTK_HOUR:
                         if (!AdjustMicroseconds(val, fval, USECS_PER_HOUR, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(HOUR);
                         type = DTK_DAY; /* set for next field */
                         break;
  
                     case DTK_DAY:
                         if (!AdjustDays(val, 1, itm_in) ||
                             !AdjustFractMicroseconds(fval, USECS_PER_DAY, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(DAY);
                         break;
  
                     case DTK_WEEK:
                         if (!AdjustDays(val, 7, itm_in) ||
                             !AdjustFractDays(fval, 7, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(WEEK);
                         break;
  
                     case DTK_MONTH:
                         if (!AdjustMonths(val, itm_in) ||
                             !AdjustFractDays(fval, DAYS_PER_MONTH, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(MONTH);
                         break;
  
                     case DTK_YEAR:
                         if (!AdjustYears(val, 1, itm_in) ||
                             !AdjustFractYears(fval, 1, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(YEAR);
                         break;
  
                     case DTK_DECADE:
                         if (!AdjustYears(val, 10, itm_in) ||
                             !AdjustFractYears(fval, 10, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(DECADE);
                         break;
  
                     case DTK_CENTURY:
                         if (!AdjustYears(val, 100, itm_in) ||
                             !AdjustFractYears(fval, 100, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(CENTURY);
                         break;
  
                     case DTK_MILLENNIUM:
                         if (!AdjustYears(val, 1000, itm_in) ||
                             !AdjustFractYears(fval, 1000, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         tmask = DTK_M(MILLENNIUM);
                         break;
  
                     default:
                         return DTERR_BAD_FORMAT;
                 }
                 parsing_unit_val = false;
                 break;
  
             case DTK_STRING:
             case DTK_SPECIAL:
                 /* reject consecutive unhandled units */
                 if (parsing_unit_val)
                     return DTERR_BAD_FORMAT;
                 type = DecodeUnits(i, field[i], &uval);
                 if (type == IGNORE_DTF)
                     continue;
  
                 tmask = 0;      /* DTK_M(type); */
                 switch (type)
                 {
                     case UNITS:
                         type = uval;
                         parsing_unit_val = true;
                         break;
  
                     case AGO:
  
                         /*
                          * "ago" is only allowed to appear at the end of the
                          * interval.
                          */
                         if (i != nf - 1)
                             return DTERR_BAD_FORMAT;
                         is_before = true;
                         type = uval;
                         break;
  
                     default:
                         return DTERR_BAD_FORMAT;
                 }
                 break;
  
             default:
                 return DTERR_BAD_FORMAT;
         }
  
         if (tmask & fmask)
             return DTERR_BAD_FORMAT;
         fmask |= tmask;
     }
  
     /* ensure that at least one time field has been found */
     if (fmask == 0)
         return DTERR_BAD_FORMAT;
  
     /* reject if unit appeared and was never handled */
     if (parsing_unit_val)
         return DTERR_BAD_FORMAT;
  
     /* finally, AGO negates everything */
     if (is_before)
     {
         if (itm_in->tm_usec == PG_INT64_MIN ||
             itm_in->tm_mday == INT_MIN ||
             itm_in->tm_mon == INT_MIN ||
             itm_in->tm_year == INT_MIN)
             return DTERR_FIELD_OVERFLOW;
  
         itm_in->tm_usec = -itm_in->tm_usec;
         itm_in->tm_mday = -itm_in->tm_mday;
         itm_in->tm_mon = -itm_in->tm_mon;
         itm_in->tm_year = -itm_in->tm_year;
     }
  
     return 0;
 }

References AdjustDays(), AdjustFractDays(), AdjustFractMicroseconds(), AdjustFractYears(), AdjustMicroseconds(), AdjustMonths(), AdjustYears(), AGO, Assert(), CENTURY, ClearPgItmIn(), DAY, DAYS_PER_MONTH, DECADE, DecodeTimeForInterval(), DecodeUnits(), DTERR_BAD_FORMAT, DTERR_FIELD_OVERFLOW, DTK_ALL_SECS_M, DTK_CENTURY, DTK_DATE, DTK_DAY, DTK_DECADE, DTK_DELTA, DTK_HOUR, DTK_M, DTK_MICROSEC, DTK_MILLENNIUM, DTK_MILLISEC, DTK_MINUTE, DTK_MONTH, DTK_NUMBER, DTK_SECOND, DTK_SPECIAL, DTK_STRING, DTK_TIME, DTK_TZ, DTK_WEEK, DTK_YEAR, HOUR, i, IGNORE_DTF, INTERVAL_MASK, IntervalStyle, INTSTYLE_SQL_STANDARD, MICROSECOND, MILLENNIUM, MILLISECOND, MINUTE, MONTH, MONTHS_PER_YEAR, ParseFraction(), pg_add_s64_overflow(), PG_INT64_MIN, pg_mul_s64_overflow(), range(), SECOND, strtoi64, strtoint(), pg_itm_in::tm_mday, pg_itm_in::tm_mon, pg_itm_in::tm_usec, pg_itm_in::tm_year, type, UNITS, USECS_PER_DAY, USECS_PER_HOUR, USECS_PER_MINUTE, USECS_PER_SEC, val, WEEK, and YEAR.

Referenced by interval_in().

◆ DecodeISO8601Interval()

int DecodeISO8601Interval	(	char *	str,
		int *	dtype,
		struct pg_itm_in *	itm_in
	)

Definition at line 3718 of file datetime.c.

 {
     bool        datepart = true;
     bool        havefield = false;
  
     *dtype = DTK_DELTA;
     ClearPgItmIn(itm_in);
  
     if (strlen(str) < 2 || str[0] != 'P')
         return DTERR_BAD_FORMAT;
  
     str++;
     while (*str)
     {
         char       *fieldstart;
         int64       val;
         double      fval;
         char        unit;
         int         dterr;
  
         if (*str == 'T')        /* T indicates the beginning of the time part */
         {
             datepart = false;
             havefield = false;
             str++;
             continue;
         }
  
         fieldstart = str;
         dterr = ParseISO8601Number(str, &str, &val, &fval);
         if (dterr)
             return dterr;
  
         /*
          * Note: we could step off the end of the string here.  Code below
          * *must* exit the loop if unit == '\0'.
          */
         unit = *str++;
  
         if (datepart)
         {
             switch (unit)       /* before T: Y M W D */
             {
                 case 'Y':
                     if (!AdjustYears(val, 1, itm_in) ||
                         !AdjustFractYears(fval, 1, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'M':
                     if (!AdjustMonths(val, itm_in) ||
                         !AdjustFractDays(fval, DAYS_PER_MONTH, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'W':
                     if (!AdjustDays(val, 7, itm_in) ||
                         !AdjustFractDays(fval, 7, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'D':
                     if (!AdjustDays(val, 1, itm_in) ||
                         !AdjustFractMicroseconds(fval, USECS_PER_DAY, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'T':       /* ISO 8601 4.4.3.3 Alternative Format / Basic */
                 case '\0':
                     if (ISO8601IntegerWidth(fieldstart) == 8 && !havefield)
                     {
                         if (!AdjustYears(val / 10000, 1, itm_in) ||
                             !AdjustMonths((val / 100) % 100, itm_in) ||
                             !AdjustDays(val % 100, 1, itm_in) ||
                             !AdjustFractMicroseconds(fval, USECS_PER_DAY, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         if (unit == '\0')
                             return 0;
                         datepart = false;
                         havefield = false;
                         continue;
                     }
                     /* Else fall through to extended alternative format */
                     /* FALLTHROUGH */
                 case '-':       /* ISO 8601 4.4.3.3 Alternative Format,
                                  * Extended */
                     if (havefield)
                         return DTERR_BAD_FORMAT;
  
                     if (!AdjustYears(val, 1, itm_in) ||
                         !AdjustFractYears(fval, 1, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (unit == '\0')
                         return 0;
                     if (unit == 'T')
                     {
                         datepart = false;
                         havefield = false;
                         continue;
                     }
  
                     dterr = ParseISO8601Number(str, &str, &val, &fval);
                     if (dterr)
                         return dterr;
                     if (!AdjustMonths(val, itm_in) ||
                         !AdjustFractDays(fval, DAYS_PER_MONTH, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (*str == '\0')
                         return 0;
                     if (*str == 'T')
                     {
                         datepart = false;
                         havefield = false;
                         continue;
                     }
                     if (*str != '-')
                         return DTERR_BAD_FORMAT;
                     str++;
  
                     dterr = ParseISO8601Number(str, &str, &val, &fval);
                     if (dterr)
                         return dterr;
                     if (!AdjustDays(val, 1, itm_in) ||
                         !AdjustFractMicroseconds(fval, USECS_PER_DAY, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (*str == '\0')
                         return 0;
                     if (*str == 'T')
                     {
                         datepart = false;
                         havefield = false;
                         continue;
                     }
                     return DTERR_BAD_FORMAT;
                 default:
                     /* not a valid date unit suffix */
                     return DTERR_BAD_FORMAT;
             }
         }
         else
         {
             switch (unit)       /* after T: H M S */
             {
                 case 'H':
                     if (!AdjustMicroseconds(val, fval, USECS_PER_HOUR, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'M':
                     if (!AdjustMicroseconds(val, fval, USECS_PER_MINUTE, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case 'S':
                     if (!AdjustMicroseconds(val, fval, USECS_PER_SEC, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     break;
                 case '\0':      /* ISO 8601 4.4.3.3 Alternative Format */
                     if (ISO8601IntegerWidth(fieldstart) == 6 && !havefield)
                     {
                         if (!AdjustMicroseconds(val / 10000, 0, USECS_PER_HOUR, itm_in) ||
                             !AdjustMicroseconds((val / 100) % 100, 0, USECS_PER_MINUTE, itm_in) ||
                             !AdjustMicroseconds(val % 100, 0, USECS_PER_SEC, itm_in) ||
                             !AdjustFractMicroseconds(fval, 1, itm_in))
                             return DTERR_FIELD_OVERFLOW;
                         return 0;
                     }
                     /* Else fall through to extended alternative format */
                     /* FALLTHROUGH */
                 case ':':       /* ISO 8601 4.4.3.3 Alternative Format,
                                  * Extended */
                     if (havefield)
                         return DTERR_BAD_FORMAT;
  
                     if (!AdjustMicroseconds(val, fval, USECS_PER_HOUR, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (unit == '\0')
                         return 0;
  
                     dterr = ParseISO8601Number(str, &str, &val, &fval);
                     if (dterr)
                         return dterr;
                     if (!AdjustMicroseconds(val, fval, USECS_PER_MINUTE, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (*str == '\0')
                         return 0;
                     if (*str != ':')
                         return DTERR_BAD_FORMAT;
                     str++;
  
                     dterr = ParseISO8601Number(str, &str, &val, &fval);
                     if (dterr)
                         return dterr;
                     if (!AdjustMicroseconds(val, fval, USECS_PER_SEC, itm_in))
                         return DTERR_FIELD_OVERFLOW;
                     if (*str == '\0')
                         return 0;
                     return DTERR_BAD_FORMAT;
  
                 default:
                     /* not a valid time unit suffix */
                     return DTERR_BAD_FORMAT;
             }
         }
  
         havefield = true;
     }
  
     return 0;
 }

References AdjustDays(), AdjustFractDays(), AdjustFractMicroseconds(), AdjustFractYears(), AdjustMicroseconds(), AdjustMonths(), AdjustYears(), ClearPgItmIn(), DAYS_PER_MONTH, DTERR_BAD_FORMAT, DTERR_FIELD_OVERFLOW, DTK_DELTA, ISO8601IntegerWidth(), ParseISO8601Number(), generate_unaccent_rules::str, USECS_PER_DAY, USECS_PER_HOUR, USECS_PER_MINUTE, USECS_PER_SEC, and val.

Referenced by interval_in().

◆ DecodeNumber()

static int DecodeNumber	(	int	flen,
		char *	str,
		bool	haveTextMonth,
		int	fmask,
		int *	tmask,
		struct pg_tm *	tm,
		fsec_t *	fsec,
		bool *	is2digits
	)

static

Definition at line 2718 of file datetime.c.

 {
     int         val;
     char       *cp;
     int         dterr;
  
     *tmask = 0;
  
     errno = 0;
     val = strtoint(str, &cp, 10);
     if (errno == ERANGE)
         return DTERR_FIELD_OVERFLOW;
     if (cp == str)
         return DTERR_BAD_FORMAT;
  
     if (*cp == '.')
     {
         /*
          * More than two digits before decimal point? Then could be a date or
          * a run-together time: 2001.360 20011225 040506.789
          */
         if (cp - str > 2)
         {
             dterr = DecodeNumberField(flen, str,
                                       (fmask | DTK_DATE_M),
                                       tmask, tm,
                                       fsec, is2digits);
             if (dterr < 0)
                 return dterr;
             return 0;
         }
  
         dterr = ParseFractionalSecond(cp, fsec);
         if (dterr)
             return dterr;
     }
     else if (*cp != '\0')
         return DTERR_BAD_FORMAT;
  
     /* Special case for day of year */
     if (flen == 3 && (fmask & DTK_DATE_M) == DTK_M(YEAR) && val >= 1 &&
         val <= 366)
     {
         *tmask = (DTK_M(DOY) | DTK_M(MONTH) | DTK_M(DAY));
         tm->tm_yday = val;
         /* tm_mon and tm_mday can't actually be set yet ... */
         return 0;
     }
  
     /* Switch based on what we have so far */
     switch (fmask & DTK_DATE_M)
     {
         case 0:
  
             /*
              * Nothing so far; make a decision about what we think the input
              * is.  There used to be lots of heuristics here, but the
              * consensus now is to be paranoid.  It *must* be either
              * YYYY-MM-DD (with a more-than-two-digit year field), or the
              * field order defined by DateOrder.
              */
             if (flen >= 3 || DateOrder == DATEORDER_YMD)
             {
                 *tmask = DTK_M(YEAR);
                 tm->tm_year = val;
             }
             else if (DateOrder == DATEORDER_DMY)
             {
                 *tmask = DTK_M(DAY);
                 tm->tm_mday = val;
             }
             else
             {
                 *tmask = DTK_M(MONTH);
                 tm->tm_mon = val;
             }
             break;
  
         case (DTK_M(YEAR)):
             /* Must be at second field of YY-MM-DD */
             *tmask = DTK_M(MONTH);
             tm->tm_mon = val;
             break;
  
         case (DTK_M(MONTH)):
             if (haveTextMonth)
             {
                 /*
                  * We are at the first numeric field of a date that included a
                  * textual month name.  We want to support the variants
                  * MON-DD-YYYY, DD-MON-YYYY, and YYYY-MON-DD as unambiguous
                  * inputs.  We will also accept MON-DD-YY or DD-MON-YY in
                  * either DMY or MDY modes, as well as YY-MON-DD in YMD mode.
                  */
                 if (flen >= 3 || DateOrder == DATEORDER_YMD)
                 {
                     *tmask = DTK_M(YEAR);
                     tm->tm_year = val;
                 }
                 else
                 {
                     *tmask = DTK_M(DAY);
                     tm->tm_mday = val;
                 }
             }
             else
             {
                 /* Must be at second field of MM-DD-YY */
                 *tmask = DTK_M(DAY);
                 tm->tm_mday = val;
             }
             break;
  
         case (DTK_M(YEAR) | DTK_M(MONTH)):
             if (haveTextMonth)
             {
                 /* Need to accept DD-MON-YYYY even in YMD mode */
                 if (flen >= 3 && *is2digits)
                 {
                     /* Guess that first numeric field is day was wrong */
                     *tmask = DTK_M(DAY);    /* YEAR is already set */
                     tm->tm_mday = tm->tm_year;
                     tm->tm_year = val;
                     *is2digits = false;
                 }
                 else
                 {
                     *tmask = DTK_M(DAY);
                     tm->tm_mday = val;
                 }
             }
             else
             {
                 /* Must be at third field of YY-MM-DD */
                 *tmask = DTK_M(DAY);
                 tm->tm_mday = val;
             }
             break;
  
         case (DTK_M(DAY)):
             /* Must be at second field of DD-MM-YY */
             *tmask = DTK_M(MONTH);
             tm->tm_mon = val;
             break;
  
         case (DTK_M(MONTH) | DTK_M(DAY)):
             /* Must be at third field of DD-MM-YY or MM-DD-YY */
             *tmask = DTK_M(YEAR);
             tm->tm_year = val;
             break;
  
         case (DTK_M(YEAR) | DTK_M(MONTH) | DTK_M(DAY)):
             /* we have all the date, so it must be a time field */
             dterr = DecodeNumberField(flen, str, fmask,
                                       tmask, tm,
                                       fsec, is2digits);
             if (dterr < 0)
                 return dterr;
             return 0;
  
         default:
             /* Anything else is bogus input */
             return DTERR_BAD_FORMAT;
     }
  
     /*
      * When processing a year field, mark it for adjustment if it's only one
      * or two digits.
      */
     if (*tmask == DTK_M(YEAR))
         *is2digits = (flen <= 2);
  
     return 0;
 }

References DateOrder, DATEORDER_DMY, DATEORDER_YMD, DAY, DecodeNumberField(), DOY, DTERR_BAD_FORMAT, DTERR_FIELD_OVERFLOW, DTK_DATE_M, DTK_M, MONTH, ParseFractionalSecond(), generate_unaccent_rules::str, strtoint(), tm, pg_tm::tm_mday, pg_tm::tm_mon, pg_tm::tm_yday, pg_tm::tm_year, val, and YEAR.

Referenced by DecodeDate(), DecodeDateTime(), and DecodeTimeOnly().

◆ DecodeNumberField()

static int DecodeNumberField	(	int	len,
		char *	str,
		int	fmask,
		int *	tmask,
		struct pg_tm *	tm,
		fsec_t *	fsec,
		bool *	is2digits
	)

static

Definition at line 2903 of file datetime.c.

 {
     char       *cp;
  
     /*
      * Have a decimal point? Then this is a date or something with a seconds
      * field...
      */
     if ((cp = strchr(str, '.')) != NULL)
     {
         /*
          * Can we use ParseFractionalSecond here?  Not clear whether trailing
          * junk should be rejected ...
          */
         if (cp[1] == '\0')
         {
             /* avoid assuming that strtod will accept "." */
             *fsec = 0;
         }
         else
         {
             double      frac;
  
             errno = 0;
             frac = strtod(cp, NULL);
             if (errno != 0)
                 return DTERR_BAD_FORMAT;
             *fsec = rint(frac * 1000000);
         }
         /* Now truncate off the fraction for further processing */
         *cp = '\0';
         len = strlen(str);
     }
     /* No decimal point and no complete date yet? */
     else if ((fmask & DTK_DATE_M) != DTK_DATE_M)
     {
         if (len >= 6)
         {
             *tmask = DTK_DATE_M;
  
             /*
              * Start from end and consider first 2 as Day, next 2 as Month,
              * and the rest as Year.
              */
             tm->tm_mday = atoi(str + (len - 2));
             *(str + (len - 2)) = '\0';
             tm->tm_mon = atoi(str + (len - 4));
             *(str + (len - 4)) = '\0';
             tm->tm_year = atoi(str);
             if ((len - 4) == 2)
                 *is2digits = true;
  
             return DTK_DATE;
         }
     }
  
     /* not all time fields are specified? */
     if ((fmask & DTK_TIME_M) != DTK_TIME_M)
     {
         /* hhmmss */
         if (len == 6)
         {
             *tmask = DTK_TIME_M;
             tm->tm_sec = atoi(str + 4);
             *(str + 4) = '\0';
             tm->tm_min = atoi(str + 2);
             *(str + 2) = '\0';
             tm->tm_hour = atoi(str);
  
             return DTK_TIME;
         }
         /* hhmm? */
         else if (len == 4)
         {
             *tmask = DTK_TIME_M;
             tm->tm_sec = 0;
             tm->tm_min = atoi(str + 2);
             *(str + 2) = '\0';
             tm->tm_hour = atoi(str);
  
             return DTK_TIME;
         }
     }
  
     return DTERR_BAD_FORMAT;
 }

References DTERR_BAD_FORMAT, DTK_DATE, DTK_DATE_M, DTK_TIME, DTK_TIME_M, len, generate_unaccent_rules::str, tm, pg_tm::tm_hour, pg_tm::tm_mday, pg_tm::tm_min, pg_tm::tm_mon, pg_tm::tm_sec, and pg_tm::tm_year.

Referenced by DecodeDateTime(), DecodeNumber(), and DecodeTimeOnly().

◆ DecodeSpecial()

int DecodeSpecial	(	int	field,
		const char *	lowtoken,
		int *	val
	)

Definition at line 3139 of file datetime.c.

 {
     int         type;
     const datetkn *tp;
  
     tp = datecache[field];
     /* use strncmp so that we match truncated tokens */
     if (tp == NULL || strncmp(lowtoken, tp->token, TOKMAXLEN) != 0)
     {
         tp = datebsearch(lowtoken, datetktbl, szdatetktbl);
     }
     if (tp == NULL)
     {
         type = UNKNOWN_FIELD;
         *val = 0;
     }
     else
     {
         datecache[field] = tp;
         type = tp->type;
         *val = tp->value;
     }
  
     return type;
 }

References datebsearch(), datecache, datetktbl, szdatetktbl, datetkn::token, TOKMAXLEN, type, datetkn::type, UNKNOWN_FIELD, val, and datetkn::value.

Referenced by DecodeDate(), DecodeDateTime(), DecodeTimeOnly(), extract_date(), interval_part_common(), time_part_common(), timestamp_part_common(), timestamptz_part_common(), and timetz_part_common().

◆ DecodeTime()

static int DecodeTime	(	char *	str,
		int	fmask,
		int	range,
		int *	tmask,
		struct pg_tm *	tm,
		fsec_t *	fsec
	)

static

Definition at line 2663 of file datetime.c.

 {
     struct pg_itm itm;
     int         dterr;
  
     dterr = DecodeTimeCommon(str, fmask, range,
                              tmask, &itm);
     if (dterr)
         return dterr;
  
     if (itm.tm_hour > INT_MAX)
         return DTERR_FIELD_OVERFLOW;
     tm->tm_hour = (int) itm.tm_hour;
     tm->tm_min = itm.tm_min;
     tm->tm_sec = itm.tm_sec;
     *fsec = itm.tm_usec;
  
     return 0;
 }

References DecodeTimeCommon(), DTERR_FIELD_OVERFLOW, range(), generate_unaccent_rules::str, tm, pg_itm::tm_hour, pg_tm::tm_hour, pg_itm::tm_min, pg_tm::tm_min, pg_itm::tm_sec, pg_tm::tm_sec, and pg_itm::tm_usec.

Referenced by DecodeDateTime(), DecodeInterval(), and DecodeTimeOnly().

◆ DecodeTimeCommon()

static int DecodeTimeCommon	(	char *	str,
		int	fmask,
		int	range,
		int *	tmask,
		struct pg_itm *	itm
	)

static

Definition at line 2581 of file datetime.c.

 {
     char       *cp;
     int         dterr;
     fsec_t      fsec = 0;
  
     *tmask = DTK_TIME_M;
  
     errno = 0;
     itm->tm_hour = strtoi64(str, &cp, 10);
     if (errno == ERANGE)
         return DTERR_FIELD_OVERFLOW;
     if (*cp != ':')
         return DTERR_BAD_FORMAT;
     errno = 0;
     itm->tm_min = strtoint(cp + 1, &cp, 10);
     if (errno == ERANGE)
         return DTERR_FIELD_OVERFLOW;
     if (*cp == '\0')
     {
         itm->tm_sec = 0;
         /* If it's a MINUTE TO SECOND interval, take 2 fields as being mm:ss */
         if (range == (INTERVAL_MASK(MINUTE) | INTERVAL_MASK(SECOND)))
         {
             if (itm->tm_hour > INT_MAX || itm->tm_hour < INT_MIN)
                 return DTERR_FIELD_OVERFLOW;
             itm->tm_sec = itm->tm_min;
             itm->tm_min = (int) itm->tm_hour;
             itm->tm_hour = 0;
         }
     }
     else if (*cp == '.')
     {
         /* always assume mm:ss.sss is MINUTE TO SECOND */
         dterr = ParseFractionalSecond(cp, &fsec);
         if (dterr)
             return dterr;
         if (itm->tm_hour > INT_MAX || itm->tm_hour < INT_MIN)
             return DTERR_FIELD_OVERFLOW;
         itm->tm_sec = itm->tm_min;
         itm->tm_min = (int) itm->tm_hour;
         itm->tm_hour = 0;
     }
     else if (*cp == ':')
     {
         errno = 0;
         itm->tm_sec = strtoint(cp + 1, &cp, 10);
         if (errno == ERANGE)
             return DTERR_FIELD_OVERFLOW;
         if (*cp == '.')
         {
             dterr = ParseFractionalSecond(cp, &fsec);
             if (dterr)
                 return dterr;
         }
         else if (*cp != '\0')
             return DTERR_BAD_FORMAT;
     }
     else
         return DTERR_BAD_FORMAT;
  
     /* do a sanity check; but caller must check the range of tm_hour */
     if (itm->tm_hour < 0 ||
         itm->tm_min < 0 || itm->tm_min > MINS_PER_HOUR - 1 ||
         itm->tm_sec < 0 || itm->tm_sec > SECS_PER_MINUTE ||
         fsec < 0 || fsec > USECS_PER_SEC)
         return DTERR_FIELD_OVERFLOW;
  
     itm->tm_usec = (int) fsec;
  
     return 0;
 }

References DTERR_BAD_FORMAT, DTERR_FIELD_OVERFLOW, DTK_TIME_M, INTERVAL_MASK, MINS_PER_HOUR, MINUTE, ParseFractionalSecond(), range(), SECOND, SECS_PER_MINUTE, generate_unaccent_rules::str, strtoi64, strtoint(), pg_itm::tm_hour, pg_itm::tm_min, pg_itm::tm_sec, pg_itm::tm_usec, and USECS_PER_SEC.

Referenced by DecodeTime(), and DecodeTimeForInterval().

◆ DecodeTimeForInterval()

static int DecodeTimeForInterval	(	char *	str,
		int	fmask,
		int	range,
		int *	tmask,
		struct pg_itm_in *	itm_in
	)

static

Definition at line 2692 of file datetime.c.

 {
     struct pg_itm itm;
     int         dterr;
  
     dterr = DecodeTimeCommon(str, fmask, range,
                              tmask, &itm);
     if (dterr)
         return dterr;
  
     itm_in->tm_usec = itm.tm_usec;
     if (!int64_multiply_add(itm.tm_hour, USECS_PER_HOUR, &itm_in->tm_usec) ||
         !int64_multiply_add(itm.tm_min, USECS_PER_MINUTE, &itm_in->tm_usec) ||
         !int64_multiply_add(itm.tm_sec, USECS_PER_SEC, &itm_in->tm_usec))
         return DTERR_FIELD_OVERFLOW;
  
     return 0;
 }

References DecodeTimeCommon(), DTERR_FIELD_OVERFLOW, int64_multiply_add(), range(), generate_unaccent_rules::str, pg_itm::tm_hour, pg_itm::tm_min, pg_itm::tm_sec, pg_itm::tm_usec, pg_itm_in::tm_usec, USECS_PER_HOUR, USECS_PER_MINUTE, and USECS_PER_SEC.

Referenced by DecodeInterval().

◆ DecodeTimeOnly()

int DecodeTimeOnly	(	char **	field,
		int *	ftype,
		int	nf,
		int *	dtype,
		struct pg_tm *	tm,
		fsec_t *	fsec,
		int *	tzp,
		DateTimeErrorExtra *	extra
	)

Definition at line 1866 of file datetime.c.

 {
     int         fmask = 0,
                 tmask,
                 type;
     int         ptype = 0;      /* "prefix type" for ISO and Julian formats */
     int         i;
     int         val;
     int         dterr;
     bool        isjulian = false;
     bool        is2digits = false;
     bool        bc = false;
     int         mer = HR24;
     pg_tz      *namedTz = NULL;
     pg_tz      *abbrevTz = NULL;
     char       *abbrev = NULL;
     pg_tz      *valtz;
  
     *dtype = DTK_TIME;
     tm->tm_hour = 0;
     tm->tm_min = 0;
     tm->tm_sec = 0;
     *fsec = 0;
     /* don't know daylight savings time status apriori */
     tm->tm_isdst = -1;
  
     if (tzp != NULL)
         *tzp = 0;
  
     for (i = 0; i < nf; i++)
     {
         switch (ftype[i])
         {
             case DTK_DATE:
  
                 /*
                  * Time zone not allowed? Then should not accept dates or time
                  * zones no matter what else!
                  */
                 if (tzp == NULL)
                     return DTERR_BAD_FORMAT;
  
                 /* Under limited circumstances, we will accept a date... */
                 if (i == 0 && nf >= 2 &&
                     (ftype[nf - 1] == DTK_DATE || ftype[1] == DTK_TIME))
                 {
                     dterr = DecodeDate(field[i], fmask,
                                        &tmask, &is2digits, tm);
                     if (dterr)
                         return dterr;
                 }
                 /* otherwise, this is a time and/or time zone */
                 else
                 {
                     if (isdigit((unsigned char) *field[i]))
                     {
                         char       *cp;
  
                         /*
                          * Starts with a digit but we already have a time
                          * field? Then we are in trouble with time already...
                          */
                         if ((fmask & DTK_TIME_M) == DTK_TIME_M)
                             return DTERR_BAD_FORMAT;
  
                         /*
                          * Should not get here and fail. Sanity check only...
                          */
                         if ((cp = strchr(field[i], '-')) == NULL)
                             return DTERR_BAD_FORMAT;
  
                         /* Get the time zone from the end of the string */
                         dterr = DecodeTimezone(cp, tzp);
                         if (dterr)
                             return dterr;
                         *cp = '\0';
  
                         /*
                          * Then read the rest of the field as a concatenated
                          * time
                          */
                         dterr = DecodeNumberField(strlen(field[i]), field[i],
                                                   (fmask | DTK_DATE_M),
                                                   &tmask, tm,
                                                   fsec, &is2digits);
                         if (dterr < 0)
                             return dterr;
                         ftype[i] = dterr;
  
                         tmask |= DTK_M(TZ);
                     }
                     else
                     {
                         namedTz = pg_tzset(field[i]);
                         if (!namedTz)
                         {
                             extra->dtee_timezone = field[i];
                             return DTERR_BAD_TIMEZONE;
                         }
                         /* we'll apply the zone setting below */
                         ftype[i] = DTK_TZ;
                         tmask = DTK_M(TZ);
                     }
                 }
                 break;
  
             case DTK_TIME:
                 dterr = DecodeTime(field[i], (fmask | DTK_DATE_M),
                                    INTERVAL_FULL_RANGE,
                                    &tmask, tm, fsec);
                 if (dterr)
                     return dterr;
                 break;
  
             case DTK_TZ:
                 {
                     int         tz;
  
                     if (tzp == NULL)
                         return DTERR_BAD_FORMAT;
  
                     dterr = DecodeTimezone(field[i], &tz);
                     if (dterr)
                         return dterr;
                     *tzp = tz;
                     tmask = DTK_M(TZ);
                 }
                 break;
  
             case DTK_NUMBER:
  
                 /*
                  * Deal with cases where previous field labeled this one
                  */
                 if (ptype != 0)
                 {
                     char       *cp;
                     int         value;
  
                     errno = 0;
                     value = strtoint(field[i], &cp, 10);
                     if (errno == ERANGE)
                         return DTERR_FIELD_OVERFLOW;
                     if (*cp != '.' && *cp != '\0')
                         return DTERR_BAD_FORMAT;
  
                     switch (ptype)
                     {
                         case DTK_JULIAN:
                             /* previous field was a label for "julian date" */
                             if (tzp == NULL)
                                 return DTERR_BAD_FORMAT;
                             if (value < 0)
                                 return DTERR_FIELD_OVERFLOW;
                             tmask = DTK_DATE_M;
                             j2date(value, &tm->tm_year, &tm->tm_mon, &tm->tm_mday);
                             isjulian = true;
  
                             if (*cp == '.')
                             {
                                 double      time;
  
                                 dterr = ParseFraction(cp, &time);
                                 if (dterr)
                                     return dterr;
                                 time *= USECS_PER_DAY;
                                 dt2time(time,
                                         &tm->tm_hour, &tm->tm_min,
                                         &tm->tm_sec, fsec);
                                 tmask |= DTK_TIME_M;
                             }
                             break;
  
                         case DTK_TIME:
                             /* previous field was "t" for ISO time */
                             dterr = DecodeNumberField(strlen(field[i]), field[i],
                                                       (fmask | DTK_DATE_M),
                                                       &tmask, tm,
                                                       fsec, &is2digits);
                             if (dterr < 0)
                                 return dterr;
                             ftype[i] = dterr;
  
                             if (tmask != DTK_TIME_M)
                                 return DTERR_BAD_FORMAT;
                             break;
  
                         default:
                             return DTERR_BAD_FORMAT;
                             break;
                     }
  
                     ptype = 0;
                     *dtype = DTK_DATE;
                 }
                 else
                 {
                     char       *cp;
                     int         flen;
  
                     flen = strlen(field[i]);
                     cp = strchr(field[i], '.');
  
                     /* Embedded decimal? */
                     if (cp != NULL)
                     {
                         /*
                          * Under limited circumstances, we will accept a
                          * date...
                          */
                         if (i == 0 && nf >= 2 && ftype[nf - 1] == DTK_DATE)
                         {
                             dterr = DecodeDate(field[i], fmask,
                                                &tmask, &is2digits, tm);
                             if (dterr)
                                 return dterr;
                         }
                         /* embedded decimal and several digits before? */
                         else if (flen - strlen(cp) > 2)
                         {
                             /*
                              * Interpret as a concatenated date or time Set
                              * the type field to allow decoding other fields
                              * later. Example: 20011223 or 040506
                              */
                             dterr = DecodeNumberField(flen, field[i],
                                                       (fmask | DTK_DATE_M),
                                                       &tmask, tm,
                                                       fsec, &is2digits);
                             if (dterr < 0)
                                 return dterr;
                             ftype[i] = dterr;
                         }
                         else
                             return DTERR_BAD_FORMAT;
                     }
                     else if (flen > 4)
                     {
                         dterr = DecodeNumberField(flen, field[i],
                                                   (fmask | DTK_DATE_M),
                                                   &tmask, tm,
                                                   fsec, &is2digits);
                         if (dterr < 0)
                             return dterr;
                         ftype[i] = dterr;
                     }
                     /* otherwise it is a single date/time field... */
                     else
                     {
                         dterr = DecodeNumber(flen, field[i],
                                              false,
                                              (fmask | DTK_DATE_M),
                                              &tmask, tm,
                                              fsec, &is2digits);
                         if (dterr)
                             return dterr;
                     }
                 }
                 break;
  
             case DTK_STRING:
             case DTK_SPECIAL:
                 /* timezone abbrevs take precedence over built-in tokens */
                 dterr = DecodeTimezoneAbbrev(i, field[i],
                                              &type, &val, &valtz, extra);
                 if (dterr)
                     return dterr;
                 if (type == UNKNOWN_FIELD)
                     type = DecodeSpecial(i, field[i], &val);
                 if (type == IGNORE_DTF)
                     continue;
  
                 tmask = DTK_M(type);
                 switch (type)
                 {
                     case RESERV:
                         switch (val)
                         {
                             case DTK_NOW:
                                 tmask = DTK_TIME_M;
                                 *dtype = DTK_TIME;
                                 GetCurrentTimeUsec(tm, fsec, NULL);
                                 break;
  
                             case DTK_ZULU:
                                 tmask = (DTK_TIME_M | DTK_M(TZ));
                                 *dtype = DTK_TIME;
                                 tm->tm_hour = 0;
                                 tm->tm_min = 0;
                                 tm->tm_sec = 0;
                                 tm->tm_isdst = 0;
                                 break;
  
                             default:
                                 return DTERR_BAD_FORMAT;
                         }
  
                         break;
  
                     case DTZMOD:
  
                         /*
                          * daylight savings time modifier (solves "MET DST"
                          * syntax)
                          */
                         tmask |= DTK_M(DTZ);
                         tm->tm_isdst = 1;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp -= val;
                         break;
  
                     case DTZ:
  
                         /*
                          * set mask for TZ here _or_ check for DTZ later when
                          * getting default timezone
                          */
                         tmask |= DTK_M(TZ);
                         tm->tm_isdst = 1;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp = -val;
                         ftype[i] = DTK_TZ;
                         break;
  
                     case TZ:
                         tm->tm_isdst = 0;
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         *tzp = -val;
                         ftype[i] = DTK_TZ;
                         break;
  
                     case DYNTZ:
                         tmask |= DTK_M(TZ);
                         if (tzp == NULL)
                             return DTERR_BAD_FORMAT;
                         /* we'll determine the actual offset later */
                         abbrevTz = valtz;
                         abbrev = field[i];
                         ftype[i] = DTK_TZ;
                         break;
  
                     case AMPM:
                         mer = val;
                         break;
  
                     case ADBC:
                         bc = (val == BC);
                         break;
  
                     case UNITS:
                         tmask = 0;
                         /* reject consecutive unhandled units */
                         if (ptype != 0)
                             return DTERR_BAD_FORMAT;
                         ptype = val;
                         break;
  
                     case ISOTIME:
                         tmask = 0;
                         /* reject consecutive unhandled units */
                         if (ptype != 0)
                             return DTERR_BAD_FORMAT;
                         ptype = val;
                         break;
  
                     case UNKNOWN_FIELD:
  
                         /*
                          * Before giving up and declaring error, check to see
                          * if it is an all-alpha timezone name.
                          */
                         namedTz = pg_tzset(field[i]);
                         if (!namedTz)
                             return DTERR_BAD_FORMAT;
                         /* we'll apply the zone setting below */
                         tmask = DTK_M(TZ);
                         break;
  
                     default:
                         return DTERR_BAD_FORMAT;
                 }
                 break;
  
             default:
                 return DTERR_BAD_FORMAT;
         }
  
         if (tmask & fmask)
             return DTERR_BAD_FORMAT;
         fmask |= tmask;
     }                           /* end loop over fields */
  
     /* reject if prefix type appeared and was never handled */
     if (ptype != 0)
         return DTERR_BAD_FORMAT;
  
     /* do final checking/adjustment of Y/M/D fields */
     dterr = ValidateDate(fmask, isjulian, is2digits, bc, tm);
     if (dterr)
         return dterr;
  
     /* handle AM/PM */
     if (mer != HR24 && tm->tm_hour > HOURS_PER_DAY / 2)
         return DTERR_FIELD_OVERFLOW;
     if (mer == AM && tm->tm_hour == HOURS_PER_DAY / 2)
         tm->tm_hour = 0;
     else if (mer == PM && tm->tm_hour != HOURS_PER_DAY / 2)
         tm->tm_hour += HOURS_PER_DAY / 2;
  
     /* check for time overflow */
     if (time_overflows(tm->tm_hour, tm->tm_min, tm->tm_sec, *fsec))
         return DTERR_FIELD_OVERFLOW;
  
     if ((fmask & DTK_TIME_M) != DTK_TIME_M)
         return DTERR_BAD_FORMAT;
  
     /*
      * If we had a full timezone spec, compute the offset (we could not do it
      * before, because we may need the date to resolve DST status).
      */
     if (namedTz != NULL)
     {
         long int    gmtoff;
  
         /* daylight savings time modifier disallowed with full TZ */
         if (fmask & DTK_M(DTZMOD))
             return DTERR_BAD_FORMAT;
  
         /* if non-DST zone, we do not need to know the date */
         if (pg_get_timezone_offset(namedTz, &gmtoff))
         {
             *tzp = -(int) gmtoff;
         }
         else
         {
             /* a date has to be specified */
             if ((fmask & DTK_DATE_M) != DTK_DATE_M)
                 return DTERR_BAD_FORMAT;
             *tzp = DetermineTimeZoneOffset(tm, namedTz);
         }
     }
  
     /*
      * Likewise, if we had a dynamic timezone abbreviation, resolve it now.
      */
     if (abbrevTz != NULL)
     {
         struct pg_tm tt,
                    *tmp = &tt;
  
         /*
          * daylight savings time modifier but no standard timezone? then error
          */
         if (fmask & DTK_M(DTZMOD))
             return DTERR_BAD_FORMAT;
  
         if ((fmask & DTK_DATE_M) == 0)
             GetCurrentDateTime(tmp);
         else
         {
             /* a date has to be specified */
             if ((fmask & DTK_DATE_M) != DTK_DATE_M)
                 return DTERR_BAD_FORMAT;
             tmp->tm_year = tm->tm_year;
             tmp->tm_mon = tm->tm_mon;
             tmp->tm_mday = tm->tm_mday;
         }
         tmp->tm_hour = tm->tm_hour;
         tmp->tm_min = tm->tm_min;
         tmp->tm_sec = tm->tm_sec;
         *tzp = DetermineTimeZoneAbbrevOffset(tmp, abbrev, abbrevTz);
         tm->tm_isdst = tmp->tm_isdst;
     }
  
     /* timezone not specified? then use session timezone */
     if (tzp != NULL && !(fmask & DTK_M(TZ)))
     {
         struct pg_tm tt,
                    *tmp = &tt;
  
         /*
          * daylight savings time modifier but no standard timezone? then error
          */
         if (fmask & DTK_M(DTZMOD))
             return DTERR_BAD_FORMAT;
  
         if ((fmask & DTK_DATE_M) == 0)
             GetCurrentDateTime(tmp);
         else
         {
             /* a date has to be specified */
             if ((fmask & DTK_DATE_M) != DTK_DATE_M)
                 return DTERR_BAD_FORMAT;
             tmp->tm_year = tm->tm_year;
             tmp->tm_mon = tm->tm_mon;
             tmp->tm_mday = tm->tm_mday;
         }
         tmp->tm_hour = tm->tm_hour;
         tmp->tm_min = tm->tm_min;
         tmp->tm_sec = tm->tm_sec;
         *tzp = DetermineTimeZoneOffset(tmp, session_timezone);
         tm->tm_isdst = tmp->tm_isdst;
     }
  
     return 0;
 }

References ADBC, AM, AMPM, BC, DecodeDate(), DecodeNumber(), DecodeNumberField(), DecodeSpecial(), DecodeTime(), DecodeTimezone(), DecodeTimezoneAbbrev(), DetermineTimeZoneAbbrevOffset(), DetermineTimeZoneOffset(), dt2time(), DateTimeErrorExtra::dtee_timezone, DTERR_BAD_FORMAT, DTERR_BAD_TIMEZONE, DTERR_FIELD_OVERFLOW, DTK_DATE, DTK_DATE_M, DTK_JULIAN, DTK_M, DTK_NOW, DTK_NUMBER, DTK_SPECIAL, DTK_STRING, DTK_TIME, DTK_TIME_M, DTK_TZ, DTK_ZULU, DTZ, DTZMOD, DYNTZ, GetCurrentDateTime(), GetCurrentTimeUsec(), HOURS_PER_DAY, HR24, i, IGNORE_DTF, INTERVAL_FULL_RANGE, ISOTIME, j2date(), ParseFraction(), pg_get_timezone_offset(), pg_tzset(), PM, RESERV, session_timezone, strtoint(), time_overflows(), tm, pg_tm::tm_hour, pg_tm::tm_isdst, pg_tm::tm_mday, pg_tm::tm_min, pg_tm::tm_mon, pg_tm::tm_sec, pg_tm::tm_year, type, TZ, UNITS, UNKNOWN_FIELD, USECS_PER_DAY, val, ValidateDate(), and value.

Referenced by time_in(), and timetz_in().

◆ DecodeTimezone()

int DecodeTimezone	(	const char *	str,
		int *	tzp
	)

Definition at line 2998 of file datetime.c.

 {
     int         tz;
     int         hr,
                 min,
                 sec = 0;
     char       *cp;
  
     /* leading character must be "+" or "-" */
     if (*str != '+' && *str != '-')
         return DTERR_BAD_FORMAT;
  
     errno = 0;
     hr = strtoint(str + 1, &cp, 10);
     if (errno == ERANGE)
         return DTERR_TZDISP_OVERFLOW;
  
     /* explicit delimiter? */
     if (*cp == ':')
     {
         errno = 0;
         min = strtoint(cp + 1, &cp, 10);
         if (errno == ERANGE)
             return DTERR_TZDISP_OVERFLOW;
         if (*cp == ':')
         {
             errno = 0;
             sec = strtoint(cp + 1, &cp, 10);
             if (errno == ERANGE)
                 return DTERR_TZDISP_OVERFLOW;
         }
     }
     /* otherwise, might have run things together... */
     else if (*cp == '\0' && strlen(str) > 3)
     {
         min = hr % 100;
         hr = hr / 100;
         /* we could, but don't, support a run-together hhmmss format */
     }
     else
         min = 0;
  
     /* Range-check the values; see notes in datatype/timestamp.h */
     if (hr < 0 || hr > MAX_TZDISP_HOUR)
         return DTERR_TZDISP_OVERFLOW;
     if (min < 0 || min >= MINS_PER_HOUR)
         return DTERR_TZDISP_OVERFLOW;
     if (sec < 0 || sec >= SECS_PER_MINUTE)
         return DTERR_TZDISP_OVERFLOW;
  
     tz = (hr * MINS_PER_HOUR + min) * SECS_PER_MINUTE + sec;
     if (*str == '-')
         tz = -tz;
  
     *tzp = -tz;
  
     if (*cp != '\0')
         return DTERR_BAD_FORMAT;
  
     return 0;
 }

References DTERR_BAD_FORMAT, DTERR_TZDISP_OVERFLOW, MAX_TZDISP_HOUR, MINS_PER_HOUR, SECS_PER_MINUTE, generate_unaccent_rules::str, and strtoint().

Referenced by DecodeDateTime(), DecodeTimeOnly(), parse_datetime(), parse_sane_timezone(), and to_timestamp().

◆ DecodeTimezoneAbbrev()

int DecodeTimezoneAbbrev	(	int	field,
		const char *	lowtoken,
		int *	ftype,
		int *	offset,
		pg_tz **	tz,
		DateTimeErrorExtra *	extra
	)

Definition at line 3082 of file datetime.c.

 {
     const datetkn *tp;
  
     tp = abbrevcache[field];
     /* use strncmp so that we match truncated tokens */
     if (tp == NULL || strncmp(lowtoken, tp->token, TOKMAXLEN) != 0)
     {
         if (zoneabbrevtbl)
             tp = datebsearch(lowtoken, zoneabbrevtbl->abbrevs,
                              zoneabbrevtbl->numabbrevs);
         else
             tp = NULL;
     }
     if (tp == NULL)
     {
         *ftype = UNKNOWN_FIELD;
         *offset = 0;
         *tz = NULL;
     }
     else
     {
         abbrevcache[field] = tp;
         *ftype = tp->type;
         if (tp->type == DYNTZ)
         {
             *offset = 0;
             *tz = FetchDynamicTimeZone(zoneabbrevtbl, tp, extra);
             if (*tz == NULL)
                 return DTERR_BAD_ZONE_ABBREV;
         }
         else
         {
             *offset = tp->value;
             *tz = NULL;
         }
     }
  
     return 0;
 }

References abbrevcache, TimeZoneAbbrevTable::abbrevs, datebsearch(), DTERR_BAD_ZONE_ABBREV, DYNTZ, FetchDynamicTimeZone(), TimeZoneAbbrevTable::numabbrevs, datetkn::token, TOKMAXLEN, datetkn::type, UNKNOWN_FIELD, datetkn::value, and zoneabbrevtbl.

Referenced by DecodeDateTime(), DecodeTimeOnly(), and DecodeTimezoneName().

◆ DecodeTimezoneName()

int DecodeTimezoneName	(	const char *	tzname,
		int *	offset,
		pg_tz **	tz
	)

Definition at line 3181 of file datetime.c.

 {
     char       *lowzone;
     int         dterr,
                 type;
     DateTimeErrorExtra extra;
  
     /*
      * First we look in the timezone abbreviation table (to handle cases like
      * "EST"), and if that fails, we look in the timezone database (to handle
      * cases like "America/New_York").  This matches the order in which
      * timestamp input checks the cases; it's important because the timezone
      * database unwisely uses a few zone names that are identical to offset
      * abbreviations.
      */
  
     /* DecodeTimezoneAbbrev requires lowercase input */
     lowzone = downcase_truncate_identifier(tzname,
                                            strlen(tzname),
                                            false);
  
     dterr = DecodeTimezoneAbbrev(0, lowzone, &type, offset, tz, &extra);
     if (dterr)
         DateTimeParseError(dterr, &extra, NULL, NULL, NULL);
  
     if (type == TZ || type == DTZ)
     {
         /* fixed-offset abbreviation, return the offset */
         return TZNAME_FIXED_OFFSET;
     }
     else if (type == DYNTZ)
     {
         /* dynamic-offset abbreviation, return its referenced timezone */
         return TZNAME_DYNTZ;
     }
     else
     {
         /* try it as a full zone name */
         *tz = pg_tzset(tzname);
         if (*tz == NULL)
             ereport(ERROR,
                     (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
                      errmsg("time zone \"%s\" not recognized", tzname)));
         return TZNAME_ZONE;
     }
 }

References DateTimeParseError(), DecodeTimezoneAbbrev(), downcase_truncate_identifier(), DTZ, DYNTZ, ereport, errcode(), errmsg(), ERROR, pg_tzset(), type, TZ, TZNAME_DYNTZ, TZNAME_FIXED_OFFSET, and TZNAME_ZONE.

Referenced by DecodeTimezoneNameToTz(), parse_sane_timezone(), timestamp_zone(), timestamptz_zone(), and timetz_zone().

◆ DecodeTimezoneNameToTz()

pg_tz* DecodeTimezoneNameToTz ( const char * tzname )

Definition at line 3236 of file datetime.c.

 {
     pg_tz      *result;
     int         offset;
  
     if (DecodeTimezoneName(tzname, &offset, &result) == TZNAME_FIXED_OFFSET)
     {
         /* fixed-offset abbreviation, get a pg_tz descriptor for that */
         result = pg_tzset_offset(-offset);  /* flip to POSIX sign convention */
     }
     return result;
 }

References DecodeTimezoneName(), pg_tzset_offset(), and TZNAME_FIXED_OFFSET.

Referenced by lookup_timezone().

◆ DecodeUnits()

int DecodeUnits	(	int	field,
		const char *	lowtoken,
		int *	val
	)

Definition at line 3936 of file datetime.c.

 {
     int         type;
     const datetkn *tp;
  
     tp = deltacache[field];
     /* use strncmp so that we match truncated tokens */
     if (tp == NULL || strncmp(lowtoken, tp->token, TOKMAXLEN) != 0)
     {
         tp = datebsearch(lowtoken, deltatktbl, szdeltatktbl);
     }
     if (tp == NULL)
     {
         type = UNKNOWN_FIELD;
         *val = 0;
     }
     else
     {
         deltacache[field] = tp;
         type = tp->type;
         *val = tp->value;
     }
  
     return type;
 }                               /* DecodeUnits() */

References datebsearch(), deltacache, deltatktbl, szdeltatktbl, datetkn::token, TOKMAXLEN, type, datetkn::type, UNKNOWN_FIELD, val, and datetkn::value.

Referenced by DecodeInterval(), extract_date(), interval_part_common(), interval_trunc(), time_part_common(), timestamp_part_common(), timestamp_trunc(), timestamptz_part_common(), timestamptz_trunc_internal(), and timetz_part_common().

◆ DetermineTimeZoneAbbrevOffset()

int DetermineTimeZoneAbbrevOffset	(	struct pg_tm *	tm,
		const char *	abbr,
		pg_tz *	tzp
	)

Definition at line 1748 of file datetime.c.

 {
     pg_time_t   t;
     int         zone_offset;
     int         abbr_offset;
     int         abbr_isdst;
  
     /*
      * Compute the UTC time we want to probe at.  (In event of overflow, we'll
      * probe at the epoch, which is a bit random but probably doesn't matter.)
      */
     zone_offset = DetermineTimeZoneOffsetInternal(tm, tzp, &t);
  
     /*
      * Try to match the abbreviation to something in the zone definition.
      */
     if (DetermineTimeZoneAbbrevOffsetInternal(t, abbr, tzp,
                                               &abbr_offset, &abbr_isdst))
     {
         /* Success, so use the abbrev-specific answers. */
         tm->tm_isdst = abbr_isdst;
         return abbr_offset;
     }
  
     /*
      * No match, so use the answers we already got from
      * DetermineTimeZoneOffsetInternal.
      */
     return zone_offset;
 }

References DetermineTimeZoneAbbrevOffsetInternal(), DetermineTimeZoneOffsetInternal(), tm, and pg_tm::tm_isdst.

Referenced by DecodeDateTime(), DecodeTimeOnly(), parse_sane_timezone(), and timestamp_zone().

◆ DetermineTimeZoneAbbrevOffsetInternal()

static bool DetermineTimeZoneAbbrevOffsetInternal	(	pg_time_t	t,
		const char *	abbr,
		pg_tz *	tzp,
		int *	offset,
		int *	isdst
	)

static

Definition at line 1823 of file datetime.c.

 {
     char        upabbr[TZ_STRLEN_MAX + 1];
     unsigned char *p;
     long int    gmtoff;
  
     /* We need to force the abbrev to upper case */
     strlcpy(upabbr, abbr, sizeof(upabbr));
     for (p = (unsigned char *) upabbr; *p; p++)
         *p = pg_toupper(*p);
  
     /* Look up the abbrev's meaning at this time in this zone */
     if (pg_interpret_timezone_abbrev(upabbr,
                                      &t,
                                      &gmtoff,
                                      isdst,
                                      tzp))
     {
         /* Change sign to agree with DetermineTimeZoneOffset() */
         *offset = (int) -gmtoff;
         return true;
     }
     return false;
 }

References pg_interpret_timezone_abbrev(), pg_toupper(), strlcpy(), and TZ_STRLEN_MAX.

Referenced by DetermineTimeZoneAbbrevOffset(), and DetermineTimeZoneAbbrevOffsetTS().

◆ DetermineTimeZoneAbbrevOffsetTS()

int DetermineTimeZoneAbbrevOffsetTS	(	TimestampTz	ts,
		const char *	abbr,
		pg_tz *	tzp,
		int *	isdst
	)

Definition at line 1786 of file datetime.c.

 {
     pg_time_t   t = timestamptz_to_time_t(ts);
     int         zone_offset;
     int         abbr_offset;
     int         tz;
     struct pg_tm tm;
     fsec_t      fsec;
  
     /*
      * If the abbrev matches anything in the zone data, this is pretty easy.
      */
     if (DetermineTimeZoneAbbrevOffsetInternal(t, abbr, tzp,
                                               &abbr_offset, isdst))
         return abbr_offset;
  
     /*
      * Else, break down the timestamp so we can use DetermineTimeZoneOffset.
      */
     if (timestamp2tm(ts, &tz, &tm, &fsec, NULL, tzp) != 0)
         ereport(ERROR,
                 (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
                  errmsg("timestamp out of range")));
  
     zone_offset = DetermineTimeZoneOffset(&tm, tzp);
     *isdst = tm.tm_isdst;
     return zone_offset;
 }

References DetermineTimeZoneAbbrevOffsetInternal(), DetermineTimeZoneOffset(), ereport, errcode(), errmsg(), ERROR, timestamp2tm(), timestamptz_to_time_t(), tm, and pg_tm::tm_isdst.

Referenced by pg_timezone_abbrevs(), timestamptz_zone(), and timetz_zone().

◆ DetermineTimeZoneOffset()

int DetermineTimeZoneOffset	(	struct pg_tm *	tm,
		pg_tz *	tzp
	)

Definition at line 1587 of file datetime.c.

 {
     pg_time_t   t;
  
     return DetermineTimeZoneOffsetInternal(tm, tzp, &t);
 }

References DetermineTimeZoneOffsetInternal(), and tm.

Referenced by date2timestamptz_opt_overflow(), DecodeDateTime(), DecodeTimeOnly(), DetermineTimeZoneAbbrevOffsetTS(), parse_sane_timezone(), time_timetz(), timestamp2timestamptz_opt_overflow(), timestamp_zone(), timestamptz_pl_interval_internal(), timestamptz_trunc_internal(), and to_timestamp().

◆ DetermineTimeZoneOffsetInternal()

static int DetermineTimeZoneOffsetInternal	(	struct pg_tm *	tm,
		pg_tz *	tzp,
		pg_time_t *	tp
	)

static

Definition at line 1609 of file datetime.c.

 {
     int         date,
                 sec;
     pg_time_t   day,
                 mytime,
                 prevtime,
                 boundary,
                 beforetime,
                 aftertime;
     long int    before_gmtoff,
                 after_gmtoff;
     int         before_isdst,
                 after_isdst;
     int         res;
  
     /*
      * First, generate the pg_time_t value corresponding to the given
      * y/m/d/h/m/s taken as GMT time.  If this overflows, punt and decide the
      * timezone is GMT.  (For a valid Julian date, integer overflow should be
      * impossible with 64-bit pg_time_t, but let's check for safety.)
      */
     if (!IS_VALID_JULIAN(tm->tm_year, tm->tm_mon, tm->tm_mday))
         goto overflow;
     date = date2j(tm->tm_year, tm->tm_mon, tm->tm_mday) - UNIX_EPOCH_JDATE;
  
     day = ((pg_time_t) date) * SECS_PER_DAY;
     if (day / SECS_PER_DAY != date)
         goto overflow;
     sec = tm->tm_sec + (tm->tm_min + tm->tm_hour * MINS_PER_HOUR) * SECS_PER_MINUTE;
     mytime = day + sec;
     /* since sec >= 0, overflow could only be from +day to -mytime */
     if (mytime < 0 && day > 0)
         goto overflow;
  
     /*
      * Find the DST time boundary just before or following the target time. We
      * assume that all zones have GMT offsets less than 24 hours, and that DST
      * boundaries can't be closer together than 48 hours, so backing up 24
      * hours and finding the "next" boundary will work.
      */
     prevtime = mytime - SECS_PER_DAY;
     if (mytime < 0 && prevtime > 0)
         goto overflow;
  
     res = pg_next_dst_boundary(&prevtime,
                                &before_gmtoff, &before_isdst,
                                &boundary,
                                &after_gmtoff, &after_isdst,
                                tzp);
     if (res < 0)
         goto overflow;          /* failure? */
  
     if (res == 0)
     {
         /* Non-DST zone, life is simple */
         tm->tm_isdst = before_isdst;
         *tp = mytime - before_gmtoff;
         return -(int) before_gmtoff;
     }
  
     /*
      * Form the candidate pg_time_t values with local-time adjustment
      */
     beforetime = mytime - before_gmtoff;
     if ((before_gmtoff > 0 &&
          mytime < 0 && beforetime > 0) ||
         (before_gmtoff <= 0 &&
          mytime > 0 && beforetime < 0))
         goto overflow;
     aftertime = mytime - after_gmtoff;
     if ((after_gmtoff > 0 &&
          mytime < 0 && aftertime > 0) ||
         (after_gmtoff <= 0 &&
          mytime > 0 && aftertime < 0))
         goto overflow;
  
     /*
      * If both before or both after the boundary time, we know what to do. The
      * boundary time itself is considered to be after the transition, which
      * means we can accept aftertime == boundary in the second case.
      */
     if (beforetime < boundary && aftertime < boundary)
     {
         tm->tm_isdst = before_isdst;
         *tp = beforetime;
         return -(int) before_gmtoff;
     }
     if (beforetime > boundary && aftertime >= boundary)
     {
         tm->tm_isdst = after_isdst;
         *tp = aftertime;
         return -(int) after_gmtoff;
     }
  
     /*
      * It's an invalid or ambiguous time due to timezone transition.  In a
      * spring-forward transition, prefer the "before" interpretation; in a
      * fall-back transition, prefer "after".  (We used to define and implement
      * this test as "prefer the standard-time interpretation", but that rule
      * does not help to resolve the behavior when both times are reported as
      * standard time; which does happen, eg Europe/Moscow in Oct 2014.  Also,
      * in some zones such as Europe/Dublin, there is widespread confusion
      * about which time offset is "standard" time, so it's fortunate that our
      * behavior doesn't depend on that.)
      */
     if (beforetime > aftertime)
     {
         tm->tm_isdst = before_isdst;
         *tp = beforetime;
         return -(int) before_gmtoff;
     }
     tm->tm_isdst = after_isdst;
     *tp = aftertime;
     return -(int) after_gmtoff;
  
 overflow:
     /* Given date is out of range, so assume UTC */
     tm->tm_isdst = 0;
     *tp = 0;
     return 0;
 }

References date2j(), IS_VALID_JULIAN, MINS_PER_HOUR, pg_next_dst_boundary(), res, SECS_PER_DAY, SECS_PER_MINUTE, tm, pg_tm::tm_hour, pg_tm::tm_isdst, pg_tm::tm_mday, pg_tm::tm_min, pg_tm::tm_mon, pg_tm::tm_sec, pg_tm::tm_year, and UNIX_EPOCH_JDATE.

Referenced by DetermineTimeZoneAbbrevOffset(), and DetermineTimeZoneOffset().

◆ EncodeDateOnly()

void EncodeDateOnly	(	struct pg_tm *	tm,
		int	style,
		char *	str
	)

Definition at line 4116 of file datetime.c.

 {
     Assert(tm->tm_mon >= 1 && tm->tm_mon <= MONTHS_PER_YEAR);
  
     switch (style)
     {
         case USE_ISO_DATES:
         case USE_XSD_DATES:
             /* compatible with ISO date formats */
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             *str++ = '-';
             str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             *str++ = '-';
             str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             break;
  
         case USE_SQL_DATES:
             /* compatible with Oracle/Ingres date formats */
             if (DateOrder == DATEORDER_DMY)
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
                 *str++ = '/';
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             }
             else
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
                 *str++ = '/';
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             }
             *str++ = '/';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             break;
  
         case USE_GERMAN_DATES:
             /* German-style date format */
             str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             *str++ = '.';
             str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             *str++ = '.';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             break;
  
         case USE_POSTGRES_DATES:
         default:
             /* traditional date-only style for Postgres */
             if (DateOrder == DATEORDER_DMY)
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
                 *str++ = '-';
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             }
             else
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
                 *str++ = '-';
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             }
             *str++ = '-';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             break;
     }
  
     if (tm->tm_year <= 0)
     {
         memcpy(str, " BC", 3);  /* Don't copy NUL */
         str += 3;
     }
     *str = '\0';
 }

References Assert(), DateOrder, DATEORDER_DMY, MONTHS_PER_YEAR, pg_ultostr_zeropad(), generate_unaccent_rules::str, tm, pg_tm::tm_mday, pg_tm::tm_mon, pg_tm::tm_year, USE_GERMAN_DATES, USE_ISO_DATES, USE_POSTGRES_DATES, USE_SQL_DATES, and USE_XSD_DATES.

Referenced by date_out(), JsonEncodeDateTime(), map_sql_value_to_xml_value(), and PGTYPESdate_to_asc().

◆ EncodeDateTime()

void EncodeDateTime	(	struct pg_tm *	tm,
		fsec_t	fsec,
		bool	print_tz,
		int	tz,
		const char *	tzn,
		int	style,
		char *	str
	)

Definition at line 4231 of file datetime.c.

 {
     int         day;
  
     Assert(tm->tm_mon >= 1 && tm->tm_mon <= MONTHS_PER_YEAR);
  
     /*
      * Negative tm_isdst means we have no valid time zone translation.
      */
     if (tm->tm_isdst < 0)
         print_tz = false;
  
     switch (style)
     {
         case USE_ISO_DATES:
         case USE_XSD_DATES:
             /* Compatible with ISO-8601 date formats */
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             *str++ = '-';
             str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             *str++ = '-';
             str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             *str++ = (style == USE_ISO_DATES) ? ' ' : 'T';
             str = pg_ultostr_zeropad(str, tm->tm_hour, 2);
             *str++ = ':';
             str = pg_ultostr_zeropad(str, tm->tm_min, 2);
             *str++ = ':';
             str = AppendTimestampSeconds(str, tm, fsec);
             if (print_tz)
                 str = EncodeTimezone(str, tz, style);
             break;
  
         case USE_SQL_DATES:
             /* Compatible with Oracle/Ingres date formats */
             if (DateOrder == DATEORDER_DMY)
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
                 *str++ = '/';
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             }
             else
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
                 *str++ = '/';
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             }
             *str++ = '/';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             *str++ = ' ';
             str = pg_ultostr_zeropad(str, tm->tm_hour, 2);
             *str++ = ':';
             str = pg_ultostr_zeropad(str, tm->tm_min, 2);
             *str++ = ':';
             str = AppendTimestampSeconds(str, tm, fsec);
  
             /*
              * Note: the uses of %.*s in this function would be risky if the
              * timezone names ever contain non-ASCII characters, since we are
              * not being careful to do encoding-aware clipping.  However, all
              * TZ abbreviations in the IANA database are plain ASCII.
              */
             if (print_tz)
             {
                 if (tzn)
                 {
                     sprintf(str, " %.*s", MAXTZLEN, tzn);
                     str += strlen(str);
                 }
                 else
                     str = EncodeTimezone(str, tz, style);
             }
             break;
  
         case USE_GERMAN_DATES:
             /* German variant on European style */
             str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             *str++ = '.';
             str = pg_ultostr_zeropad(str, tm->tm_mon, 2);
             *str++ = '.';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
             *str++ = ' ';
             str = pg_ultostr_zeropad(str, tm->tm_hour, 2);
             *str++ = ':';
             str = pg_ultostr_zeropad(str, tm->tm_min, 2);
             *str++ = ':';
             str = AppendTimestampSeconds(str, tm, fsec);
  
             if (print_tz)
             {
                 if (tzn)
                 {
                     sprintf(str, " %.*s", MAXTZLEN, tzn);
                     str += strlen(str);
                 }
                 else
                     str = EncodeTimezone(str, tz, style);
             }
             break;
  
         case USE_POSTGRES_DATES:
         default:
             /* Backward-compatible with traditional Postgres abstime dates */
             day = date2j(tm->tm_year, tm->tm_mon, tm->tm_mday);
             tm->tm_wday = j2day(day);
             memcpy(str, days[tm->tm_wday], 3);
             str += 3;
             *str++ = ' ';
             if (DateOrder == DATEORDER_DMY)
             {
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
                 *str++ = ' ';
                 memcpy(str, months[tm->tm_mon - 1], 3);
                 str += 3;
             }
             else
             {
                 memcpy(str, months[tm->tm_mon - 1], 3);
                 str += 3;
                 *str++ = ' ';
                 str = pg_ultostr_zeropad(str, tm->tm_mday, 2);
             }
             *str++ = ' ';
             str = pg_ultostr_zeropad(str, tm->tm_hour, 2);
             *str++ = ':';
             str = pg_ultostr_zeropad(str, tm->tm_min, 2);
             *str++ = ':';
             str = AppendTimestampSeconds(str, tm, fsec);
             *str++ = ' ';
             str = pg_ultostr_zeropad(str,
                                      (tm->tm_year > 0) ? tm->tm_year : -(tm->tm_year - 1), 4);
  
             if (print_tz)
             {
                 if (tzn)
                 {
                     sprintf(str, " %.*s", MAXTZLEN, tzn);
                     str += strlen(str);
                 }
                 else
                 {
                     /*
                      * We have a time zone, but no string version. Use the
                      * numeric form, but be sure to include a leading space to
                      * avoid formatting something which would be rejected by
                      * the date/time parser later. - thomas 2001-10-19
                      */
                     *str++ = ' ';
                     str = EncodeTimezone(str, tz, style);
                 }
             }
             break;
     }
  
     if (tm->tm_year <= 0)
     {
         memcpy(str, " BC", 3);  /* Don't copy NUL */
         str += 3;
     }
     *str = '\0';
 }

References AppendTimestampSeconds(), Assert(), date2j(), DateOrder, DATEORDER_DMY, days, EncodeTimezone(), j2day(), MAXTZLEN, months, MONTHS_PER_YEAR, pg_ultostr_zeropad(), sprintf, generate_unaccent_rules::str, tm, pg_tm::tm_hour, pg_tm::tm_isdst, pg_tm::tm_mday, pg_tm::tm_min, pg_tm::tm_mon, pg_tm::tm_wday, pg_tm::tm_year, USE_GERMAN_DATES, USE_ISO_DATES, USE_POSTGRES_DATES, USE_SQL_DATES, and USE_XSD_DATES.

Referenced by JsonEncodeDateTime(), map_sql_value_to_xml_value(), PGTYPEStimestamp_to_asc(), timestamp_out(), timestamptz_out(), and timestamptz_to_str().

◆ EncodeInterval()

void EncodeInterval	(	struct pg_itm *	itm,
		int	style,
		char *	str
	)

Definition at line 4474 of file datetime.c.

 {
     char       *cp = str;
     int         year = itm->tm_year;
     int         mon = itm->tm_mon;
     int64       mday = itm->tm_mday;    /* tm_mday could be INT_MIN */
     int64       hour = itm->tm_hour;
     int         min = itm->tm_min;
     int         sec = itm->tm_sec;
     int         fsec = itm->tm_usec;
     bool        is_before = false;
     bool        is_zero = true;
  
     /*
      * The sign of year and month are guaranteed to match, since they are
      * stored internally as "month". But we'll need to check for is_before and
      * is_zero when determining the signs of day and hour/minute/seconds
      * fields.
      */
     switch (style)
     {
             /* SQL Standard interval format */
         case INTSTYLE_SQL_STANDARD:
             {
                 bool        has_negative = year < 0 || mon < 0 ||
                     mday < 0 || hour < 0 ||
                     min < 0 || sec < 0 || fsec < 0;
                 bool        has_positive = year > 0 || mon > 0 ||
                     mday > 0 || hour > 0 ||
                     min > 0 || sec > 0 || fsec > 0;
                 bool        has_year_month = year != 0 || mon != 0;
                 bool        has_day_time = mday != 0 || hour != 0 ||
                     min != 0 || sec != 0 || fsec != 0;
                 bool        has_day = mday != 0;
                 bool        sql_standard_value = !(has_negative && has_positive) &&
                     !(has_year_month && has_day_time);
  
                 /*
                  * SQL Standard wants only 1 "<sign>" preceding the whole
                  * interval ... but can't do that if mixed signs.
                  */
                 if (has_negative && sql_standard_value)
                 {
                     *cp++ = '-';
                     year = -year;
                     mon = -mon;
                     mday = -mday;
                     hour = -hour;
                     min = -min;
                     sec = -sec;
                     fsec = -fsec;
                 }
  
                 if (!has_negative && !has_positive)
                 {
                     sprintf(cp, "0");
                 }
                 else if (!sql_standard_value)
                 {
                     /*
                      * For non sql-standard interval values, force outputting
                      * the signs to avoid ambiguities with intervals with
                      * mixed sign components.
                      */
                     char        year_sign = (year < 0 || mon < 0) ? '-' : '+';
                     char        day_sign = (mday < 0) ? '-' : '+';
                     char        sec_sign = (hour < 0 || min < 0 ||
                                             sec < 0 || fsec < 0) ? '-' : '+';
  
                     sprintf(cp, "%c%d-%d %c%lld %c%lld:%02d:",
                             year_sign, abs(year), abs(mon),
                             day_sign, (long long) i64abs(mday),
                             sec_sign, (long long) i64abs(hour), abs(min));
                     cp += strlen(cp);
                     cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, true);
                     *cp = '\0';
                 }
                 else if (has_year_month)
                 {
                     sprintf(cp, "%d-%d", year, mon);
                 }
                 else if (has_day)
                 {
                     sprintf(cp, "%lld %lld:%02d:",
                             (long long) mday, (long long) hour, min);
                     cp += strlen(cp);
                     cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, true);
                     *cp = '\0';
                 }
                 else
                 {
                     sprintf(cp, "%lld:%02d:", (long long) hour, min);
                     cp += strlen(cp);
                     cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, true);
                     *cp = '\0';
                 }
             }
             break;
  
             /* ISO 8601 "time-intervals by duration only" */
         case INTSTYLE_ISO_8601:
             /* special-case zero to avoid printing nothing */
             if (year == 0 && mon == 0 && mday == 0 &&
                 hour == 0 && min == 0 && sec == 0 && fsec == 0)
             {
                 sprintf(cp, "PT0S");
                 break;
             }
             *cp++ = 'P';
             cp = AddISO8601IntPart(cp, year, 'Y');
             cp = AddISO8601IntPart(cp, mon, 'M');
             cp = AddISO8601IntPart(cp, mday, 'D');
             if (hour != 0 || min != 0 || sec != 0 || fsec != 0)
                 *cp++ = 'T';
             cp = AddISO8601IntPart(cp, hour, 'H');
             cp = AddISO8601IntPart(cp, min, 'M');
             if (sec != 0 || fsec != 0)
             {
                 if (sec < 0 || fsec < 0)
                     *cp++ = '-';
                 cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, false);
                 *cp++ = 'S';
                 *cp++ = '\0';
             }
             break;
  
             /* Compatible with postgresql < 8.4 when DateStyle = 'iso' */
         case INTSTYLE_POSTGRES:
             cp = AddPostgresIntPart(cp, year, "year", &is_zero, &is_before);
  
             /*
              * Ideally we should spell out "month" like we do for "year" and
              * "day".  However, for backward compatibility, we can't easily
              * fix this.  bjm 2011-05-24
              */
             cp = AddPostgresIntPart(cp, mon, "mon", &is_zero, &is_before);
             cp = AddPostgresIntPart(cp, mday, "day", &is_zero, &is_before);
             if (is_zero || hour != 0 || min != 0 || sec != 0 || fsec != 0)
             {
                 bool        minus = (hour < 0 || min < 0 || sec < 0 || fsec < 0);
  
                 sprintf(cp, "%s%s%02lld:%02d:",
                         is_zero ? "" : " ",
                         (minus ? "-" : (is_before ? "+" : "")),
                         (long long) i64abs(hour), abs(min));
                 cp += strlen(cp);
                 cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, true);
                 *cp = '\0';
             }
             break;
  
             /* Compatible with postgresql < 8.4 when DateStyle != 'iso' */
         case INTSTYLE_POSTGRES_VERBOSE:
         default:
             strcpy(cp, "@");
             cp++;
             cp = AddVerboseIntPart(cp, year, "year", &is_zero, &is_before);
             cp = AddVerboseIntPart(cp, mon, "mon", &is_zero, &is_before);
             cp = AddVerboseIntPart(cp, mday, "day", &is_zero, &is_before);
             cp = AddVerboseIntPart(cp, hour, "hour", &is_zero, &is_before);
             cp = AddVerboseIntPart(cp, min, "min", &is_zero, &is_before);
             if (sec != 0 || fsec != 0)
             {
                 *cp++ = ' ';
                 if (sec < 0 || (sec == 0 && fsec < 0))
                 {
                     if (is_zero)
                         is_before = true;
                     else if (!is_before)
                         *cp++ = '-';
                 }
                 else if (is_before)
                     *cp++ = '-';
                 cp = AppendSeconds(cp, sec, fsec, MAX_INTERVAL_PRECISION, false);
                 /* We output "ago", not negatives, so use abs(). */
                 sprintf(cp, " sec%s",
                         (abs(sec) != 1 || fsec != 0) ? "s" : "");
                 is_zero = false;
             }
             /* identically zero? then put in a unitless zero... */
             if (is_zero)
                 strcat(cp, " 0");
             if (is_before)
                 strcat(cp, " ago");
             break;
     }
 }

References AddISO8601IntPart(), AddPostgresIntPart(), AddVerboseIntPart(), AppendSeconds(), i64abs, INTSTYLE_ISO_8601, INTSTYLE_POSTGRES, INTSTYLE_POSTGRES_VERBOSE, INTSTYLE_SQL_STANDARD, MAX_INTERVAL_PRECISION, sprintf, generate_unaccent_rules::str, pg_itm::tm_hour, pg_itm::tm_mday, pg_itm::tm_min, pg_itm::tm_mon, pg_itm::tm_sec, pg_itm::tm_usec, and pg_itm::tm_year.

Referenced by interval_out().

◆ EncodeTimeOnly()

void EncodeTimeOnly	(	struct pg_tm *	tm,
		fsec_t	fsec,
		bool	print_tz,
		int	tz,
		int	style,
		char *	str
	)

Definition at line 4201 of file datetime.c.

 {
     str = pg_ultostr_zeropad(str, tm->tm_hour, 2);
     *str++ = ':';
     str = pg_ultostr_zeropad(str, tm->tm_min, 2);
     *str++ = ':';
     str = AppendSeconds(str, tm->tm_sec, fsec, MAX_TIME_PRECISION, true);
     if (print_tz)
         str = EncodeTimezone(str, tz, style);
     *str = '\0';
 }

References AppendSeconds(), EncodeTimezone(), MAX_TIME_PRECISION, pg_ultostr_zeropad(), generate_unaccent_rules::str, tm, pg_tm::tm_hour, pg_tm::tm_min, and pg_tm::tm_sec.

Referenced by JsonEncodeDateTime(), time_out(), and timetz_out().

◆ EncodeTimezone()

static char* EncodeTimezone	(	char *	str,
		int	tz,
		int	style
	)

static

Definition at line 4078 of file datetime.c.

 {
     int         hour,
                 min,
                 sec;
  
     sec = abs(tz);
     min = sec / SECS_PER_MINUTE;
     sec -= min * SECS_PER_MINUTE;
     hour = min / MINS_PER_HOUR;
     min -= hour * MINS_PER_HOUR;
  
     /* TZ is negated compared to sign we wish to display ... */
     *str++ = (tz <= 0 ? '+' : '-');
  
     if (sec != 0)
     {
         str = pg_ultostr_zeropad(str, hour, 2);
         *str++ = ':';
         str = pg_ultostr_zeropad(str, min, 2);
         *str++ = ':';
         str = pg_ultostr_zeropad(str, sec, 2);
     }
     else if (min != 0 || style == USE_XSD_DATES)
     {
         str = pg_ultostr_zeropad(str, hour, 2);
         *str++ = ':';
         str = pg_ultostr_zeropad(str, min, 2);
     }
     else
         str = pg_ultostr_zeropad(str, hour, 2);
     return str;
 }

References MINS_PER_HOUR, pg_ultostr_zeropad(), SECS_PER_MINUTE, generate_unaccent_rules::str, and USE_XSD_DATES.

Referenced by EncodeDateTime(), and EncodeTimeOnly().

◆ FetchDynamicTimeZone()

static pg_tz * FetchDynamicTimeZone	(	TimeZoneAbbrevTable *	tbl,
		const datetkn *	tp,
		DateTimeErrorExtra *	extra
	)

static

Definition at line 4859 of file datetime.c.

 {
     DynamicZoneAbbrev *dtza;
  
     /* Just some sanity checks to prevent indexing off into nowhere */
     Assert(tp->type == DYNTZ);
     Assert(tp->value > 0 && tp->value < tbl->tblsize);
  
     dtza = (DynamicZoneAbbrev *) ((char *) tbl + tp->value);
  
     /* Look up the underlying zone if we haven't already */
     if (dtza->tz == NULL)
     {
         dtza->tz = pg_tzset(dtza->zone);
         if (dtza->tz == NULL)
         {
             /* Ooops, bogus zone name in config file entry */
             extra->dtee_timezone = dtza->zone;
             extra->dtee_abbrev = tp->token;
         }
     }
     return dtza->tz;
 }

References Assert(), DateTimeErrorExtra::dtee_abbrev, DateTimeErrorExtra::dtee_timezone, DYNTZ, pg_tzset(), TimeZoneAbbrevTable::tblsize, datetkn::token, datetkn::type, DynamicZoneAbbrev::tz, datetkn::value, and DynamicZoneAbbrev::zone.

Referenced by DecodeTimezoneAbbrev(), and pg_timezone_abbrevs().

◆ GetCurrentDateTime()

void GetCurrentDateTime ( struct pg_tm * tm )

Definition at line 368 of file datetime.c.

 {
     fsec_t      fsec;
  
     GetCurrentTimeUsec(tm, &fsec, NULL);
 }

References GetCurrentTimeUsec(), and tm.

Referenced by DecodeDateTime(), DecodeTimeOnly(), GetSQLCurrentDate(), PGTYPESdate_today(), PGTYPEStimestamp_current(), and time_timetz().

◆ GetCurrentTimeUsec()

void GetCurrentTimeUsec	(	struct pg_tm *	tm,
		fsec_t *	fsec,
		int *	tzp
	)

Definition at line 389 of file datetime.c.

 {
     TimestampTz cur_ts = GetCurrentTransactionStartTimestamp();
  
     /*
      * The cache key must include both current time and current timezone.  By
      * representing the timezone by just a pointer, we're assuming that
      * distinct timezone settings could never have the same pointer value.
      * This is true by virtue of the hashtable used inside pg_tzset();
      * however, it might need another look if we ever allow entries in that
      * hash to be recycled.
      */
     static TimestampTz cache_ts = 0;
     static pg_tz *cache_timezone = NULL;
     static struct pg_tm cache_tm;
     static fsec_t cache_fsec;
     static int  cache_tz;
  
     if (cur_ts != cache_ts || session_timezone != cache_timezone)
     {
         /*
          * Make sure cache is marked invalid in case of error after partial
          * update within timestamp2tm.
          */
         cache_timezone = NULL;
  
         /*
          * Perform the computation, storing results into cache.  We do not
          * really expect any error here, since current time surely ought to be
          * within range, but check just for sanity's sake.
          */
         if (timestamp2tm(cur_ts, &cache_tz, &cache_tm, &cache_fsec,
                          NULL, session_timezone) != 0)
             ereport(ERROR,
                     (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
                      errmsg("timestamp out of range")));
  
         /* OK, so mark the cache valid. */
         cache_ts = cur_ts;
         cache_timezone = session_timezone;
     }
  
     *tm = cache_tm;
     *fsec = cache_fsec;
     if (tzp != NULL)
         *tzp = cache_tz;
 }

References ereport, errcode(), errmsg(), ERROR, GetCurrentTransactionStartTimestamp(), session_timezone, timestamp2tm(), and tm.

Referenced by DecodeDateTime(), DecodeTimeOnly(), GetCurrentDateTime(), GetSQLCurrentTime(), and GetSQLLocalTime().

◆ InstallTimeZoneAbbrevs()

void InstallTimeZoneAbbrevs ( TimeZoneAbbrevTable * tbl )

Definition at line 4846 of file datetime.c.

 {
     zoneabbrevtbl = tbl;
     /* reset abbrevcache, which may contain pointers into old table */
     memset(abbrevcache, 0, sizeof(abbrevcache));
 }

References abbrevcache, and zoneabbrevtbl.

Referenced by assign_timezone_abbreviations().

◆ int64_multiply_add()

static bool int64_multiply_add	(	int64	val,
		int64	multiplier,
		int64 *	sum
	)

static

Definition at line 524 of file datetime.c.

 {
     int64       product;
  
     if (pg_mul_s64_overflow(val, multiplier, &product) ||
         pg_add_s64_overflow(*sum, product, sum))
         return false;
     return true;
 }

References pg_add_s64_overflow(), pg_mul_s64_overflow(), and val.

Referenced by AdjustMicroseconds(), and DecodeTimeForInterval().

◆ ISO8601IntegerWidth()

static int ISO8601IntegerWidth ( char * fieldstart )

static

Definition at line 3691 of file datetime.c.

 {
     /* We might have had a leading '-' */
     if (*fieldstart == '-')
         fieldstart++;
     return strspn(fieldstart, "0123456789");
 }

Referenced by DecodeISO8601Interval().

◆ j2date()

void j2date	(	int	jd,
		int *	year,
		int *	month,
		int *	day
	)

Definition at line 313 of file datetime.c.

 {
     unsigned int julian;
     unsigned int quad;
     unsigned int extra;
     int         y;
  
     julian = jd;
     julian += 32044;
     quad = julian / 146097;
     extra = (julian - quad * 146097) * 4 + 3;
     julian += 60 + quad * 3 + extra / 146097;
     quad = julian / 1461;
     julian -= quad * 1461;
     y = julian * 4 / 1461;
     julian = ((y != 0) ? ((julian + 305) % 365) : ((julian + 306) % 366))
         + 123;
     y += quad * 4;
     *year = y - 4800;
     quad = julian * 2141 / 65536;
     *day = julian - 7834 * quad / 256;
     *month = (quad + 10) % MONTHS_PER_YEAR + 1;
 }                               /* j2date() */

References MONTHS_PER_YEAR, and y.

Referenced by date2timestamptz_opt_overflow(), date_out(), DecodeDateTime(), DecodeTimeOnly(), do_to_timestamp(), extract_date(), isoweek2date(), isoweekdate2date(), JsonEncodeDateTime(), map_sql_value_to_xml_value(), PGTYPESdate_fmt_asc(), PGTYPESdate_julmdy(), PGTYPESdate_to_asc(), timestamp2tm(), timestamp_pl_interval(), timestamptz_pl_interval_internal(), and ValidateDate().

◆ j2day()

int j2day ( int date )

Definition at line 346 of file datetime.c.

 {
     date += 1;
     date %= 7;
     /* Cope if division truncates towards zero, as it probably does */
     if (date < 0)
         date += 7;
  
     return date;
 }                               /* j2day() */

Referenced by date2isoweek(), date2isoyear(), EncodeDateTime(), extract_date(), isoweek2j(), timestamp_part_common(), and timestamptz_part_common().

◆ ParseDateTime()

int ParseDateTime	(	const char *	timestr,
		char *	workbuf,
		size_t	buflen,
		char **	field,
		int *	ftype,
		int	maxfields,
		int *	numfields
	)

Definition at line 756 of file datetime.c.

 {
     int         nf = 0;
     const char *cp = timestr;
     char       *bufp = workbuf;
     const char *bufend = workbuf + buflen;
  
     /*
      * Set the character pointed-to by "bufptr" to "newchar", and increment
      * "bufptr". "end" gives the end of the buffer -- we return an error if
      * there is no space left to append a character to the buffer. Note that
      * "bufptr" is evaluated twice.
      */
 #define APPEND_CHAR(bufptr, end, newchar)       \
     do                                          \
     {                                           \
         if (((bufptr) + 1) >= (end))            \
             return DTERR_BAD_FORMAT;            \
         *(bufptr)++ = newchar;                  \
     } while (0)
  
     /* outer loop through fields */
     while (*cp != '\0')
     {
         /* Ignore spaces between fields */
         if (isspace((unsigned char) *cp))
         {
             cp++;
             continue;
         }
  
         /* Record start of current field */
         if (nf >= maxfields)
             return DTERR_BAD_FORMAT;
         field[nf] = bufp;
  
         /* leading digit? then date or time */
         if (isdigit((unsigned char) *cp))
         {
             APPEND_CHAR(bufp, bufend, *cp++);
             while (isdigit((unsigned char) *cp))
                 APPEND_CHAR(bufp, bufend, *cp++);
  
             /* time field? */
             if (*cp == ':')
             {
                 ftype[nf] = DTK_TIME;
                 APPEND_CHAR(bufp, bufend, *cp++);
                 while (isdigit((unsigned char) *cp) ||
                        (*cp == ':') || (*cp == '.'))
                     APPEND_CHAR(bufp, bufend, *cp++);
             }
             /* date field? allow embedded text month */
             else if (*cp == '-' || *cp == '/' || *cp == '.')
             {
                 /* save delimiting character to use later */
                 char        delim = *cp;
  
                 APPEND_CHAR(bufp, bufend, *cp++);
                 /* second field is all digits? then no embedded text month */
                 if (isdigit((unsigned char) *cp))
                 {
                     ftype[nf] = ((delim == '.') ? DTK_NUMBER : DTK_DATE);
                     while (isdigit((unsigned char) *cp))
                         APPEND_CHAR(bufp, bufend, *cp++);
  
                     /*
                      * insist that the delimiters match to get a three-field
                      * date.
                      */
                     if (*cp == delim)
                     {
                         ftype[nf] = DTK_DATE;
                         APPEND_CHAR(bufp, bufend, *cp++);
                         while (isdigit((unsigned char) *cp) || *cp == delim)
                             APPEND_CHAR(bufp, bufend, *cp++);
                     }
                 }
                 else
                 {
                     ftype[nf] = DTK_DATE;
                     while (isalnum((unsigned char) *cp) || *cp == delim)
                         APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
                 }
             }
  
             /*
              * otherwise, number only and will determine year, month, day, or
              * concatenated fields later...
              */
             else
                 ftype[nf] = DTK_NUMBER;
         }
         /* Leading decimal point? Then fractional seconds... */
         else if (*cp == '.')
         {
             APPEND_CHAR(bufp, bufend, *cp++);
             while (isdigit((unsigned char) *cp))
                 APPEND_CHAR(bufp, bufend, *cp++);
  
             ftype[nf] = DTK_NUMBER;
         }
  
         /*
          * text? then date string, month, day of week, special, or timezone
          */
         else if (isalpha((unsigned char) *cp))
         {
             bool        is_date;
  
             ftype[nf] = DTK_STRING;
             APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
             while (isalpha((unsigned char) *cp))
                 APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
  
             /*
              * Dates can have embedded '-', '/', or '.' separators.  It could
              * also be a timezone name containing embedded '/', '+', '-', '_',
              * or ':' (but '_' or ':' can't be the first punctuation). If the
              * next character is a digit or '+', we need to check whether what
              * we have so far is a recognized non-timezone keyword --- if so,
              * don't believe that this is the start of a timezone.
              */
             is_date = false;
             if (*cp == '-' || *cp == '/' || *cp == '.')
                 is_date = true;
             else if (*cp == '+' || isdigit((unsigned char) *cp))
             {
                 *bufp = '\0';   /* null-terminate current field value */
                 /* we need search only the core token table, not TZ names */
                 if (datebsearch(field[nf], datetktbl, szdatetktbl) == NULL)
                     is_date = true;
             }
             if (is_date)
             {
                 ftype[nf] = DTK_DATE;
                 do
                 {
                     APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
                 } while (*cp == '+' || *cp == '-' ||
                          *cp == '/' || *cp == '_' ||
                          *cp == '.' || *cp == ':' ||
                          isalnum((unsigned char) *cp));
             }
         }
         /* sign? then special or numeric timezone */
         else if (*cp == '+' || *cp == '-')
         {
             APPEND_CHAR(bufp, bufend, *cp++);
             /* soak up leading whitespace */
             while (isspace((unsigned char) *cp))
                 cp++;
             /* numeric timezone? */
             /* note that "DTK_TZ" could also be a signed float or yyyy-mm */
             if (isdigit((unsigned char) *cp))
             {
                 ftype[nf] = DTK_TZ;
                 APPEND_CHAR(bufp, bufend, *cp++);
                 while (isdigit((unsigned char) *cp) ||
                        *cp == ':' || *cp == '.' || *cp == '-')
                     APPEND_CHAR(bufp, bufend, *cp++);
             }
             /* special? */
             else if (isalpha((unsigned char) *cp))
             {
                 ftype[nf] = DTK_SPECIAL;
                 APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
                 while (isalpha((unsigned char) *cp))
                     APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
             }
             /* otherwise something wrong... */
             else
                 return DTERR_BAD_FORMAT;
         }
         /* ignore other punctuation but use as delimiter */
         else if (ispunct((unsigned char) *cp))
         {
             cp++;
             continue;
         }
         /* otherwise, something is not right... */
         else
             return DTERR_BAD_FORMAT;
  
         /* force in a delimiter after each field */
         *bufp++ = '\0';
         nf++;
     }
  
     *numfields = nf;
  
     return 0;
 }

References APPEND_CHAR, datebsearch(), datetktbl, DTERR_BAD_FORMAT, DTK_DATE, DTK_NUMBER, DTK_SPECIAL, DTK_STRING, DTK_TIME, DTK_TZ, pg_tolower(), and szdatetktbl.

Referenced by check_recovery_target_time(), date_in(), interval_in(), pg_logdir_ls_internal(), PGTYPESdate_from_asc(), PGTYPESinterval_from_asc(), PGTYPEStimestamp_from_asc(), time_in(), timestamp_in(), timestamptz_in(), and timetz_in().

◆ ParseFraction()

static int ParseFraction	(	char *	cp,
		double *	frac
	)

static

Definition at line 682 of file datetime.c.

 {
     /* Caller should always pass the start of the fraction part */
     Assert(*cp == '.');
  
     /*
      * We want to allow just "." with no digits, but some versions of strtod
      * will report EINVAL for that, so special-case it.
      */
     if (cp[1] == '\0')
     {
         *frac = 0;
     }
     else
     {
         errno = 0;
         *frac = strtod(cp, &cp);
         /* check for parse failure */
         if (*cp != '\0' || errno != 0)
             return DTERR_BAD_FORMAT;
     }
     return 0;
 }

References Assert(), and DTERR_BAD_FORMAT.

Referenced by DecodeDateTime(), DecodeInterval(), DecodeTimeOnly(), and ParseFractionalSecond().

◆ ParseFractionalSecond()

static int ParseFractionalSecond	(	char *	cp,
		fsec_t *	fsec
	)

static

Definition at line 711 of file datetime.c.

 {
     double      frac;
     int         dterr;
  
     dterr = ParseFraction(cp, &frac);
     if (dterr)
         return dterr;
     *fsec = rint(frac * 1000000);
     return 0;
 }

References ParseFraction().

Referenced by DecodeNumber(), and DecodeTimeCommon().

◆ ParseISO8601Number()

static int ParseISO8601Number	(	char *	str,
		char **	endptr,
		int64 *	ipart,
		double *	fpart
	)

static

Definition at line 3649 of file datetime.c.

 {
     double      val;
  
     /*
      * Historically this has accepted anything that strtod() would take,
      * notably including "e" notation, so continue doing that.  This is
      * slightly annoying because the precision of double is less than that of
      * int64, so we would lose accuracy for inputs larger than 2^53 or so.
      * However, historically we rejected inputs outside the int32 range,
      * making that concern moot.  What we do now is reject abs(val) above
      * 1.0e15 (a round number a bit less than 2^50), so that any accepted
      * value will have an exact integer part, and thereby a fraction part with
      * abs(*fpart) less than 1.  In the absence of field complaints it doesn't
      * seem worth working harder.
      */
     if (!(isdigit((unsigned char) *str) || *str == '-' || *str == '.'))
         return DTERR_BAD_FORMAT;
     errno = 0;
     val = strtod(str, endptr);
     /* did we not see anything that looks like a double? */

Macros

Functions

Variables

Macro Definition Documentation

◆ APPEND_CHAR

Function Documentation

◆ AddISO8601IntPart()

◆ AddPostgresIntPart()

◆ AddVerboseIntPart()

◆ AdjustDays()

◆ AdjustFractDays()

◆ AdjustFractMicroseconds()

◆ AdjustFractYears()

◆ AdjustMicroseconds()

◆ AdjustMonths()

◆ AdjustYears()

◆ AppendSeconds()

◆ AppendTimestampSeconds()

◆ CheckDateTokenTable()

◆ CheckDateTokenTables()

◆ ClearPgItmIn()

◆ ConvertTimeZoneAbbrevs()

◆ date2j()

◆ datebsearch()

◆ DateTimeParseError()

◆ DecodeDate()

◆ DecodeDateTime()

◆ DecodeInterval()

◆ DecodeISO8601Interval()

◆ DecodeNumber()

◆ DecodeNumberField()

◆ DecodeSpecial()

◆ DecodeTime()

◆ DecodeTimeCommon()

◆ DecodeTimeForInterval()

◆ DecodeTimeOnly()

◆ DecodeTimezone()

◆ DecodeTimezoneAbbrev()

◆ DecodeTimezoneName()

◆ DecodeTimezoneNameToTz()

◆ DecodeUnits()

◆ DetermineTimeZoneAbbrevOffset()

◆ DetermineTimeZoneAbbrevOffsetInternal()

◆ DetermineTimeZoneAbbrevOffsetTS()

◆ DetermineTimeZoneOffset()

◆ DetermineTimeZoneOffsetInternal()

◆ EncodeDateOnly()

◆ EncodeDateTime()

◆ EncodeInterval()

◆ EncodeTimeOnly()

◆ EncodeTimezone()

◆ FetchDynamicTimeZone()

◆ GetCurrentDateTime()

◆ GetCurrentTimeUsec()

◆ InstallTimeZoneAbbrevs()

◆ int64_multiply_add()

◆ ISO8601IntegerWidth()

◆ j2date()

◆ j2day()

◆ ParseDateTime()

◆ ParseFraction()

◆ ParseFractionalSecond()

◆ ParseISO8601Number()