postgresql/doc/src/sgml/datetime.sgml

<!-- doc/src/sgml/datetime.sgml -->

 <appendix id="datetime-appendix">
  <title>Date/Time Support</title>

  <para>
   <productname>PostgreSQL</productname> uses an internal heuristic
   parser for all date/time input support. Dates and times are input as
   strings, and are broken up into distinct fields with a preliminary
   determination of what kind of information can be in the
   field. Each field is interpreted and either assigned a numeric
   value, ignored, or rejected.
   The parser contains internal lookup tables for all textual fields,
   including months, days of the week, and time zones.
  </para>

  <para>
   This appendix includes information on the content of these
   lookup tables and describes the steps used by the parser to decode
   dates and times.
  </para>

  <sect1 id="datetime-input-rules">
   <title>Date/Time Input Interpretation</title>

   <para>
    The date/time type inputs are all decoded using the following procedure.
   </para>

   <procedure>
    <step>
     <para>
      Break the input string into tokens and categorize each token as
      a string, time, time zone, or number.
     </para>

     <substeps>
      <step>
       <para>
        If the numeric token contains a colon (<literal>:</>), this is
        a time string. Include all subsequent digits and colons.
       </para>
      </step>

      <step>
       <para>
        If the numeric token contains a dash (<literal>-</>), slash
        (<literal>/</>), or two or more dots (<literal>.</>), this is
        a date string which might have a text month.  If a date token has
        already been seen, it is instead interpreted as a time zone
        name (e.g., <literal>America/New_York</>).
       </para>
      </step>

      <step>
       <para>
        If the token is numeric only, then it is either a single field
        or an ISO 8601 concatenated date (e.g.,
        <literal>19990113</literal> for January 13, 1999) or time
        (e.g., <literal>141516</literal> for 14:15:16).
       </para>
      </step>

      <step>
       <para>
        If the token starts with a plus (<literal>+</>) or minus
        (<literal>-</>), then it is either a numeric time zone or a special
        field.
       </para>
      </step>
     </substeps>
    </step>

    <step>
     <para>
      If the token is a text string, match up with possible strings:
     </para>
     
     <substeps>
      <step>
       <para>
        Do a binary-search table lookup for the token as a time zone
        abbreviation.
       </para>
      </step>
      
      <step>
       <para>
        If not found, do a similar binary-search table lookup to match
        the token as either a special string (e.g., <literal>today</literal>),
        day (e.g., <literal>Thursday</literal>),
        month (e.g., <literal>January</literal>),
        or noise word (e.g., <literal>at</literal>, <literal>on</literal>).
       </para>
      </step>

      <step>
       <para>
        If still not found, throw an error.
       </para>
      </step>
     </substeps>
    </step>
    
    <step>
     <para>
      When the token is a number or number field:
     </para>

     <substeps>
      <step>
       <para>
        If there are eight or six digits,
        and if no other date fields have been previously read, then interpret 
        as a <quote>concatenated date</quote> (e.g.,
        <literal>19990118</literal> or <literal>990118</literal>).
        The interpretation is <literal>YYYYMMDD</> or <literal>YYMMDD</>.
       </para>
      </step>

      <step>
       <para>
        If the token is three digits
        and a year has already been read, then interpret as day of year.
       </para>
      </step>
      
      <step>
       <para>
        If four or six digits and a year has already been read, then
        interpret as a time (<literal>HHMM</> or <literal>HHMMSS</>).
       </para>
      </step>

      <step>
       <para>
        If three or more digits and no date fields have yet been found,
        interpret as a year (this forces yy-mm-dd ordering of the remaining
        date fields).
       </para>
      </step>

      <step>
       <para>
        Otherwise the date field ordering is assumed to follow the
        <varname>DateStyle</> setting: mm-dd-yy, dd-mm-yy, or yy-mm-dd.
        Throw an error if a month or day field is found to be out of range.
       </para>
      </step>
     </substeps>
    </step>

    <step>
     <para>
      If BC has been specified, negate the year and add one for
      internal storage.  (There is no year zero in the Gregorian
      calendar, so numerically 1 BC becomes year zero.)
     </para>
    </step>

    <step>
     <para>
      If BC was not specified, and if the year field was two digits in length,
      then adjust the year to four digits. If the field is less than 70, then
      add 2000, otherwise add 1900.

      <tip>
       <para>
        Gregorian years AD 1-99 can be entered by using 4 digits with leading
        zeros (e.g., <literal>0099</> is AD 99).
       </para>
      </tip>
     </para>
    </step>
   </procedure>
  </sect1>


  <sect1 id="datetime-keywords">
   <title>Date/Time Key Words</title>

   <para>
    <xref linkend="datetime-month-table"> shows the tokens that are
    recognized as names of months.
   </para>

    <table id="datetime-month-table">
     <title>Month Names</title>
     <tgroup cols="2">
      <thead>
       <row>
        <entry>Month</entry>
        <entry>Abbreviations</entry>
       </row>
      </thead>
      <tbody>
       <row>
        <entry>January</entry>
        <entry>Jan</entry>
       </row>
       <row>
        <entry>February</entry>
        <entry>Feb</entry>
       </row>
       <row>
        <entry>March</entry>
        <entry>Mar</entry>
       </row>
       <row>
        <entry>April</entry>
        <entry>Apr</entry>
       </row>
       <row>
        <entry>May</entry>
        <entry></entry>
       </row>
       <row>
        <entry>June</entry>
        <entry>Jun</entry>
       </row>
       <row>
        <entry>July</entry>
        <entry>Jul</entry>
       </row>
       <row>
        <entry>August</entry>
        <entry>Aug</entry>
       </row>
       <row>
        <entry>September</entry>
        <entry>Sep, Sept</entry>
       </row>
       <row>
        <entry>October</entry>
        <entry>Oct</entry>
       </row>
       <row>
        <entry>November</entry>
        <entry>Nov</entry>
       </row>
       <row>
        <entry>December</entry>
        <entry>Dec</entry>
       </row>
      </tbody>
     </tgroup>
    </table>

    <para>
     <xref linkend="datetime-dow-table"> shows the tokens that are
     recognized as names of days of the week.
    </para>

     <table id="datetime-dow-table">
      <title>Day of the Week Names</title>
      <tgroup cols="2">
       <thead>
        <row>
         <entry>Day</entry>
         <entry>Abbreviations</entry>
        </row>
       </thead>
       <tbody>
        <row>
         <entry>Sunday</entry>
         <entry>Sun</entry>
        </row>
        <row>
         <entry>Monday</entry>
         <entry>Mon</entry>
        </row>
        <row>
         <entry>Tuesday</entry>
         <entry>Tue, Tues</entry>
        </row>
        <row>
         <entry>Wednesday</entry>
         <entry>Wed, Weds</entry>
        </row>
        <row>
         <entry>Thursday</entry>
         <entry>Thu, Thur, Thurs</entry>
        </row>
        <row>
         <entry>Friday</entry>
         <entry>Fri</entry>
        </row>
        <row>
         <entry>Saturday</entry>
         <entry>Sat</entry>
        </row>
       </tbody>
      </tgroup>
     </table>

   <para>
    <xref linkend="datetime-mod-table"> shows the tokens that serve
    various modifier purposes.
   </para>

    <table id="datetime-mod-table">
     <title>Date/Time Field Modifiers</title>
     <tgroup cols="2">
      <thead>
       <row>
        <entry>Identifier</entry>
        <entry>Description</entry>
       </row>
      </thead>
      <tbody>
       <row>
        <entry><literal>AM</literal></entry>
        <entry>Time is before 12:00</entry>
       </row>
       <row>
        <entry><literal>AT</literal></entry>
        <entry>Ignored</entry>
       </row>
       <row>
        <entry><literal>JULIAN</>, <literal>JD</>, <literal>J</></entry>
        <entry>Next field is Julian Day</entry>
       </row>
       <row>
        <entry><literal>ON</literal></entry>
        <entry>Ignored</entry>
       </row>
       <row>
        <entry><literal>PM</literal></entry>
        <entry>Time is on or after 12:00</entry>
       </row>
       <row>
        <entry><literal>T</literal></entry>
        <entry>Next field is time</entry>
       </row>
      </tbody>
     </tgroup>
    </table>
  </sect1>

  <sect1 id="datetime-config-files">
  <title>Date/Time Configuration Files</title>

   <indexterm>
    <primary>time zone</primary>
    <secondary>input abbreviations</secondary>
   </indexterm>

   <para>
    Since timezone abbreviations are not well standardized,
    <productname>PostgreSQL</productname> provides a means to customize
    the set of abbreviations accepted by the server.  The
    <xref linkend="guc-timezone-abbreviations"> run-time parameter
    determines the active set of abbreviations.  While this parameter
    can be altered by any database user, the possible values for it
    are under the control of the database administrator &mdash; they
    are in fact names of configuration files stored in
    <filename>.../share/timezonesets/</> of the installation directory.
    By adding or altering files in that directory, the administrator
    can set local policy for timezone abbreviations.
   </para>

   <para>
    <literal>timezone_abbreviations</> can be set to any file name
    found in <filename>.../share/timezonesets/</>, if the file's name
    is entirely alphabetic.  (The prohibition against non-alphabetic
    characters in <literal>timezone_abbreviations</> prevents reading
    files outside the intended directory, as well as reading editor
    backup files and other extraneous files.)
   </para>

   <para>
    A timezone abbreviation file can contain blank lines and comments
    beginning with <literal>#</>.  Non-comment lines must have one of
    these formats:

<synopsis>
<replaceable>time_zone_name</replaceable> <replaceable>offset</replaceable>
<replaceable>time_zone_name</replaceable> <replaceable>offset</replaceable> D
@INCLUDE <replaceable>file_name</replaceable>
@OVERRIDE
</synopsis>
   </para>

   <para>
    A <replaceable>time_zone_name</replaceable> is just the abbreviation
    being defined.  The <replaceable>offset</replaceable> is the zone's
    offset in seconds from UTC, positive being east from Greenwich and
    negative being west.  For example, -18000 would be five hours west
    of Greenwich, or North American east coast standard time.  <literal>D</>
    indicates that the zone name represents local daylight-savings time
    rather than standard time. Since all known time zone offsets are on
    15 minute boundaries, the number of seconds has to be a multiple of 900.
   </para>

   <para>
    The <literal>@INCLUDE</> syntax allows inclusion of another file in the
    <filename>.../share/timezonesets/</> directory.  Inclusion can be nested,
    to a limited depth.
   </para>

   <para>
    The <literal>@OVERRIDE</> syntax indicates that subsequent entries in the
    file can override previous entries (i.e., entries obtained from included
    files).  Without this, conflicting definitions of the same timezone
    abbreviation are considered an error.
   </para>

   <para>
    In an unmodified installation, the file <filename>Default</> contains
    all the non-conflicting time zone abbreviations for most of the world.
    Additional files <filename>Australia</> and <filename>India</> are
    provided for those regions: these files first include the
    <literal>Default</> file and then add or modify timezones as needed.
   </para>

   <para>
    For reference purposes, a standard installation also contains files
    <filename>Africa.txt</>, <filename>America.txt</>, etc, containing
    information about every time zone abbreviation known to be in use
    according to the <literal>zoneinfo</> timezone database.  The zone name
    definitions found in these files can be copied and pasted into a custom
    configuration file as needed.  Note that these files cannot be directly
    referenced as <literal>timezone_abbreviations</> settings, because of
    the dot embedded in their names.
   </para>

   <note>
    <para>
     If an error occurs while reading the time zone data sets, no new value is
     applied but the old set is kept. If the error occurs while starting the
     database, startup fails.
    </para>
   </note>

   <caution>
    <para>
     Time zone abbreviations defined in the configuration file override
     non-timezone meanings built into <productname>PostgreSQL</productname>.
     For example, the <filename>Australia</> configuration file defines
     <literal>SAT</> (for South Australian Standard Time).  When this
     file is active, <literal>SAT</> will not be recognized as an abbreviation
     for Saturday.
    </para>
   </caution>

   <caution>
    <para>
     If you modify files in <filename>.../share/timezonesets/</>,
     it is up to you to make backups &mdash; a normal database dump
     will not include this directory.
    </para>
   </caution>

  </sect1>

  <sect1 id="datetime-units-history">
  <title>History of Units</title>

  <para>
   The Julian calendar was introduced by Julius Caesar in 45 BC.
   It was in common use in the Western world
   until the year 1582, when countries started changing to the Gregorian
   calendar.  In the Julian calendar, the tropical year is
   approximated as 365 1/4 days = 365.25 days. This gives an error of
   about 1 day in 128 years.
  </para>

  <para>   
   The accumulating calendar error prompted
   Pope Gregory XIII to reform the calendar in accordance with
   instructions from the Council of Trent.
   In the Gregorian calendar, the tropical year is approximated as
   365 + 97 / 400 days = 365.2425 days. Thus it takes approximately 3300
   years for the tropical year to shift one day with respect to the
   Gregorian calendar.
  </para>

  <para>
   The approximation 365+97/400 is achieved by having 97 leap years
   every 400 years, using the following rules:

   <simplelist>
    <member>
     Every year divisible by 4 is a leap year.
    </member>
    <member>
     However, every year divisible by 100 is not a leap year.
    </member>
    <member>
     However, every year divisible by 400 is a leap year after all.
    </member>
   </simplelist>

   So, 1700, 1800, 1900, 2100, and 2200 are not leap years. But 1600,
   2000, and 2400 are leap years.

   By contrast, in the older Julian calendar all years divisible by 4 are leap
   years.
  </para>

  <para>
   The papal bull of February 1582 decreed that 10 days should be dropped
   from October 1582 so that 15 October should follow immediately after
   4 October.
   This was observed in Italy, Poland, Portugal, and Spain. Other Catholic
   countries followed shortly after, but Protestant countries were
   reluctant to change, and the Greek Orthodox countries didn't change
   until the start of the 20th century.

   The reform was observed by Great Britain and Dominions (including what is
   now the USA) in 1752.
   Thus 2 September 1752 was followed by 14 September 1752.

   This is why Unix systems have the <command>cal</command> program
   produce the following:

<screen>
$ <userinput>cal 9 1752</userinput>
   September 1752
 S  M Tu  W Th  F  S
       1  2 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
</screen>
  </para>

  <para>
   The SQL standard states that <quote>Within the definition of a
   <quote>datetime literal</quote>, the <quote>datetime
   value</quote>s are constrained by the natural rules for dates and
   times according to the Gregorian calendar</quote>.  Dates between
   1582-10-05 and 1582-10-14, although eliminated in some countries
   by Papal fiat, conform to <quote>natural rules</quote> and are
   hence valid dates.  <productname>PostgreSQL</> follows the SQL
   standard's lead by counting dates exclusively in the Gregorian
   calendar, even for years before that calendar was in use.
  </para>

  <para>
   Different calendars have been developed in various parts of the
   world, many predating the Gregorian system.

   For example,
   the beginnings of the Chinese calendar can be traced back to the 14th
   century BC. Legend has it that the Emperor Huangdi invented that
   calendar in 2637 BC.
   
   The People's Republic of China uses the Gregorian calendar
   for civil purposes. The Chinese calendar is used for determining
   festivals.
  </para>

  <para>
   The <quote>Julian Date</quote> is unrelated to the <quote>Julian
   calendar</quote>.  
   The Julian Date system was invented by the French scholar
   Joseph Justus Scaliger (1540-1609)
   and probably takes its name from Scaliger's father,
   the Italian scholar Julius Caesar Scaliger (1484-1558).
   In the Julian Date system, each day has a sequential number, starting
   from JD 0 (which is sometimes called <emphasis>the</> Julian Date).
   JD 0 corresponds to 1 January 4713 BC in the Julian calendar, or
   24 November 4714 BC in the Gregorian calendar.  Julian Date counting
   is most often used by astronomers for labeling their nightly observations,
   and therefore a date runs from noon UTC to the next noon UTC, rather than
   from midnight to midnight: JD 0 designates the 24 hours from noon UTC on
   1 January 4713 BC to noon UTC on 2 January 4713 BC.
  </para>

  <para>
   Although <productname>PostgreSQL</> supports Julian Date notation for
   input and output of dates (and also uses them for some internal datetime
   calculations), it does not observe the nicety of having dates run from
   noon to noon.  <productname>PostgreSQL</> treats a Julian Date as running
   from midnight to midnight.
  </para>

 </sect1>
</appendix>