database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-31 22:04:40 +03:00
Code Activity

Improve behavior of to_timestamp()/to_date() functions

Browse Source 
to_timestamp()/to_date() functions were introduced mainly for Oracle
compatibility, and became very popular among PostgreSQL users.  However, some
behavior of to_timestamp()/to_date() functions are both incompatible with Oracle
and confusing for our users.  This behavior is related to handling of spaces and
separators in non FX (fixed format) mode.  This commit reworks this behavior
making less confusing, better documented and more compatible with Oracle.

Nevertheless, there are still following incompatibilities with Oracle.
1) We don't insist that there are no format string patterns unmatched to
   input string.
2) In FX mode we don't insist space and separators in format string to exactly
   match input string.
3) When format string patterns are divided by mix of spaces and separators, we
   don't distinguish them, while Oracle takes into account only last group of
   spaces/separators.

Discussion: https://postgr.es/m/1873520224.1784572.1465833145330.JavaMail.yahoo%40mail.yahoo.com
Author: Artur Zakirov, Alexander Korotkov, Liudmila Mantrova
Review: Amul Sul, Robert Haas, Tom Lane, Dmitry Dolgov, David G. Johnston
This commit is contained in:
Alexander Korotkov
2018-09-09 21:19:51 +03:00
parent 5f08accdad
commit cf98467242
4 changed files with 280 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
doc/src/sgml/func.sgml
View File

@ -6262,16 +6262,57 @@ SELECT regexp_match('abc01234xyz', '(?:(.*?)(\d+)(.*)){1,1}');
<listitem>
<para>
<function>to_timestamp</function> and <function>to_date</function>
skip multiple blank spaces in the input string unless the
<literal>FX</literal> option is used. For example,
<literal>to_timestamp('2000&nbsp;&nbsp;&nbsp;&nbsp;JUN', 'YYYY MON')</literal> works, but
skip multiple blank spaces at the beginning of the input string and
around date and time values unless the <literal>FX</literal> option is used. For example,
<literal>to_timestamp('&nbsp;2000&nbsp;&nbsp;&nbsp;&nbsp;JUN', 'YYYY MON')</literal> and
<literal>to_timestamp('2000 - JUN', 'YYYY-MON')</literal> work, but
<literal>to_timestamp('2000&nbsp;&nbsp;&nbsp;&nbsp;JUN', 'FXYYYY MON')</literal> returns an error
because <function>to_timestamp</function> expects one space only.
because <function>to_timestamp</function> expects a single space only.
<literal>FX</literal> must be specified as the first item in
the template.
</para>
</listitem>
<listitem>
<para>
A separator (a space or a non-letter/non-digit character) in the template string of
<function>to_timestamp</function> and <function>to_date</function>
matches any single separator in the input string or is skipped,
unless the <literal>FX</literal> option is used.
For example, <literal>to_timestamp('2000JUN', 'YYYY///MON')</literal> and
<literal>to_timestamp('2000/JUN', 'YYYY MON')</literal> work, but
<literal>to_timestamp('2000//JUN', 'YYYY/MON')</literal>
returns an error because the number of separators in the input string
exceeds the number of separators in the template.
</para>
<para>
If <literal>FX</literal> is specified, separator in template string
matches to exactly one character in input string. Notice we don't insist
input string character to be the same as template string separator.
For example, <literal>to_timestamp('2000/JUN', 'FXYYYY MON')</literal>
works, but <literal>to_timestamp('2000/JUN', 'FXYYYY&nbsp;&nbsp;MON')</literal>
returns an error because a space second template string space consumed
letter <literal>J</literal> from the input string.
</para>
</listitem>
<listitem>
<para>
<literal>TZH</literal> template pattern can match a signed number.
Without the <literal>FX</literal> option, it may lead to ambiguity in
interpretation of the minus sign, which can also be interpreted as a separator.
This ambiguity is resolved as follows. If the number of separators before
<literal>TZH</literal> in the template string is less than the number of
separators before the minus sign in the input string, the minus sign
is interpreted as part of <literal>TZH</literal>.
Otherwise, the minus sign is considered to be a separator between values.
For example, <literal>to_timestamp('2000 -10', 'YYYY TZH')</literal> matches
<literal>-10</literal> to <literal>TZH</literal>, but
<literal>to_timestamp('2000 -10', 'YYYY&nbsp;&nbsp;TZH')</literal>
matches <literal>10</literal> to <literal>TZH</literal>.
</para>
</listitem>
<listitem>
<para>
Ordinary text is allowed in <function>to_char</function>
@ -6287,6 +6328,19 @@ SELECT regexp_match('abc01234xyz', '(?:(.*?)(\d+)(.*)){1,1}');
string; for example <literal>"XX"</literal> skips two input characters
(whether or not they are <literal>XX</literal>).
</para>
<tip>
<para>
Prior to <productname>PostgreSQL</productname> 12, it was possible to
skip arbitrary text in the input string using non-letter or non-digit
characters. For example,
<literal>to_timestamp('2000y6m1d', 'yyyy-MM-DD')</literal> used to
work. Now you can only use letter characters for this purpose. For example,
<literal>to_timestamp('2000y6m1d', 'yyyytMMtDDt')</literal> and
<literal>to_timestamp('2000y6m1d', 'yyyy"y"MM"m"DD"d"')</literal>
skip <literal>y</literal>, <literal>m</literal>, and
<literal>d</literal>.
</para>
</tip>
</listitem>
<listitem>