From: Arthur Zakirov <a.zakirov@postgrespro.ru>
To: Dmitry Dolgov <9erthalion6@gmail.com>
Cc: Robert Haas <robertmhaas@gmail.com>
Cc: Thomas Munro <thomas.munro@enterprisedb.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>
Cc: amul sul <sulamul@gmail.com>
Cc: Pavel Stehule <pavel.stehule@gmail.com>
Cc: David G. Johnston <david.g.johnston@gmail.com>
Cc: Alex Ignatov <a.ignatov@postgrespro.ru>
Cc: Bruce Momjian <bruce@momjian.us>
Cc: amul sul <sul_amul@yahoo.co.in>
Cc: PostgreSQL-development <pgsql-hackers@postgresql.org>
Subject: Re: [HACKERS] Bug in to_timestamp().
Date: Sat, 17 Feb 2018 12:02:40 +0300
Message-ID: <20180217090239.GA14632@arthur.localdomain> (raw)
In-Reply-To: <CA+q6zcUS3fQGLoeLcuJLtz-gRD6vHqpn1fBe0cnORdx93QtO4w@mail.gmail.com>
References: <CAEepm=1Y7z1VSisBKxa6x3E-jP-+=rWfw25q_PH2nGjqVcX-rw@mail.gmail.com>
<20180112125848.GA32559@zakirov.localdomain>
<CA+q6zcVQBu9E8-RjRPc_sh1XUWhcug4V5bCSjN=O4w==Lpfp8A@mail.gmail.com>
<CA+TgmoY2KAFz9H=kK0rh0ZYqVXXTOHACLbCHJV-eSwcNNn03Rg@mail.gmail.com>
<CA+q6zcXAXS2srjDq5nNHReYo21MurOrt=c88yryUUixe5vPGag@mail.gmail.com>
<20180206091648.GA8507@zakirov.localdomain>
<CA+q6zcVOmcmbDZnrJdxighp=95xdkGYOucabtay+CpWU8UDqkA@mail.gmail.com>
<CA+q6zcX6Svkn1Cf0GDvSqCEV-SFC6ET2bxjXQpWaoZYWAordUA@mail.gmail.com>
<20180212114943.GA6629@zakirov.localdomain>
<CA+q6zcUS3fQGLoeLcuJLtz-gRD6vHqpn1fBe0cnORdx93QtO4w@mail.gmail.com>
On Wed, Feb 14, 2018 at 05:23:53PM +0100, Dmitry Dolgov wrote:
> * On line 52 there is a removed empty line, without any other changes around
Agree, it is unnecessary change. There was the macro there before.
> * On line 177 there is a new commented out line of code. I assume it's not
> an
> explanation or something and we don't need it, am I right?
It explains a node type we deal with. But maybe it is not very useful,
so I removed it.
> And Oracle complains about this:
>
> SELECT to_timestamp('2000 + JUN', ' /') FROM dual
> ORA-01830: date format picture ends before converting entire input string
>
> SELECT to_timestamp('2000 + JUN', ' ') FROM dual
> ORA-01830: date format picture ends before converting entire input string
>
> It's sort of corner case, but anyway maybe you would be interested to handle
> it.
I think it could be fixed by another patch. But I'm not sure that it
will be accepted as well as this patch :).
--
Arthur Zakirov
Postgres Professional: http://www.postgrespro.com
Russian Postgres Company
Attachments:0001-to-timestamp-format-checking-v12.patch (text/plain, 12.3K) downloaddiff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgmlindex 1e535cf215..1cc8b078f7 100644--- a/doc/src/sgml/func.sgml+++ b/doc/src/sgml/func.sgml@@ -6174,7 +6174,8 @@ SELECT regexp_match('abc01234xyz', '(?:(.*?)(\d+)(.*)){1,1}');
<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 JUN', 'YYYY MON')</literal> works, but+ <literal>to_timestamp('2000 JUN', 'YYYY MON')</literal> and+ <literal>to_timestamp('2000 JUN', 'YYYY MON')</literal> work, but
<literal>to_timestamp('2000 JUN', 'FXYYYY MON')</literal> returns an error
because <function>to_timestamp</function> expects one space only.
<literal>FX</literal> must be specified as the first item in
@@ -6182,6 +6183,19 @@ SELECT regexp_match('abc01234xyz', '(?:(.*?)(\d+)(.*)){1,1}');
</para>
</listitem>
+ <listitem>+ <para>+ <function>to_timestamp</function> and <function>to_date</function> don't+ skip multiple printable non letter and non digit characters in the input+ string, but skip them in the formatting string. For example,+ <literal>to_timestamp('2000-JUN', '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 count of the "/" character in the input string+ doesn't match count of it in the formatting string.+ </para>+ </listitem>+
<listitem>
<para>
Ordinary text is allowed in <function>to_char</function>
diff --git a/src/backend/utils/adt/formatting.c b/src/backend/utils/adt/formatting.cindex b8bd4caa3e..a23122d3de 100644--- a/src/backend/utils/adt/formatting.c+++ b/src/backend/utils/adt/formatting.c@@ -171,6 +171,8 @@ typedef struct
#define NODE_TYPE_END 1
#define NODE_TYPE_ACTION 2
#define NODE_TYPE_CHAR 3
+#define NODE_TYPE_SEPARATOR 4+#define NODE_TYPE_SPACE 5
#define SUFFTYPE_PREFIX 1
#define SUFFTYPE_POSTFIX 2
@@ -961,6 +963,7 @@ typedef struct NUMProc
static const KeyWord *index_seq_search(const char *str, const KeyWord *kw,
const int *index);
static const KeySuffix *suff_search(const char *str, const KeySuffix *suf, int type);
+static bool is_separator_char(const char *str);
static void NUMDesc_prepare(NUMDesc *num, FormatNode *n);
static void parse_format(FormatNode *node, const char *str, const KeyWord *kw,
const KeySuffix *suf, const int *index, int ver, NUMDesc *Num);
@@ -1050,6 +1053,16 @@ suff_search(const char *str, const KeySuffix *suf, int type)
return NULL;
}
+static bool+is_separator_char(const char *str)+{+ /* ASCII printable character, but not letter or digit */+ return (*str > 0x20 && *str < 0x7F &&+ !(*str >= 'A' && *str <= 'Z') &&+ !(*str >= 'a' && *str <= 'z') &&+ !(*str >= '0' && *str <= '9'));+}+
/* ----------
* Prepare NUMDesc (number description struct) via FormatNode struct
* ----------
@@ -1325,7 +1338,14 @@ parse_format(FormatNode *node, const char *str, const KeyWord *kw,
if (*str == '\\' && *(str + 1) == '"')
str++;
chlen = pg_mblen(str);
- n->type = NODE_TYPE_CHAR;++ if (ver == DCH_TYPE && is_separator_char(str))+ n->type = NODE_TYPE_SEPARATOR;+ else if (isspace((unsigned char) *str))+ n->type = NODE_TYPE_SPACE;+ else+ n->type = NODE_TYPE_CHAR;+
memcpy(n->character, str, chlen);
n->character[chlen] = '\0';
n->key = NULL;
@@ -2996,12 +3016,73 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out)
for (n = node, s = in; n->type != NODE_TYPE_END && *s != '\0'; n++)
{
- if (n->type != NODE_TYPE_ACTION)+ if (n->type == NODE_TYPE_SPACE)+ {+ /*+ * In non FX (fixed format) mode we don't insist that the consumed+ * character matches the format's character.+ */+ if (!fx_mode)+ {+ if (isspace((unsigned char) *s) || is_separator_char(s))+ s++;++ continue;+ }+ /*+ * In FX mode we insist that whitespace from the format string+ * matches whitespace from the input string.+ */+ else if (!isspace((unsigned char) *s))+ ereport(ERROR,+ (errcode(ERRCODE_INVALID_DATETIME_FORMAT),+ errmsg("unexpected character \"%.*s\", expected space character \"%s\"",+ pg_mblen(s), s, n->character),+ errhint("In FX mode, punctuation in the input string "+ "must exactly match the format string.")));+ }+ else if (n->type == NODE_TYPE_SEPARATOR)+ {+ if (!fx_mode)+ {+ /*+ * Ignore spaces before separator character when not in FX+ * mode.+ */+ while (*s != '\0' && isspace((unsigned char) *s))+ s++;++ /*+ * In non FX mode we don't insist that the consumed character+ * matches the format's character.+ */+ if (is_separator_char(s))+ s++;++ continue;+ }+ /*+ * In FX mode we insist that separator character from the format+ * string matches separator character from the input string.+ */+ else if (*n->character != *s)+ ereport(ERROR,+ (errcode(ERRCODE_INVALID_DATETIME_FORMAT),+ errmsg("unexpected character \"%.*s\", expected separator character \"%s\"",+ pg_mblen(s), s, n->character),+ errhint("In FX mode, punctuation in the input string "+ "must exactly match the format string.")));++ s++;+ continue;+ }+ else if (n->type != NODE_TYPE_ACTION)
{
/*
- * Separator, so consume one character from input string. Notice- * we don't insist that the consumed character match the format's- * character.+ * Text character, so consume one character from input string.+ * Notice we don't insist that the consumed character match the+ * format's character.+ * Text field ignores FX mode.
*/
s += pg_mblen(s);
continue;
diff --git a/src/test/regress/expected/horology.out b/src/test/regress/expected/horology.outindex 63e39198e6..fb595c7b17 100644--- a/src/test/regress/expected/horology.out+++ b/src/test/regress/expected/horology.out@@ -2769,14 +2769,32 @@ SELECT to_timestamp('97/2/16 8:14:30', 'FMYYYY/FMMM/FMDD FMHH:FMMI:FMSS');
Sat Feb 16 08:14:30 0097 PST
(1 row)
+SELECT to_timestamp('2011$03!18 23_38_15', 'YYYY-MM-DD HH24:MI:SS');+ to_timestamp +------------------------------+ Fri Mar 18 23:38:15 2011 PDT+(1 row)+
SELECT to_timestamp('1985 January 12', 'YYYY FMMonth DD');
to_timestamp
------------------------------
Sat Jan 12 00:00:00 1985 PST
(1 row)
+SELECT to_timestamp('1985 FMMonth 12', 'YYYY "FMMonth" DD');+ to_timestamp +------------------------------+ Sat Jan 12 00:00:00 1985 PST+(1 row)++SELECT to_timestamp('1985 \ 12', 'YYYY \\ DD');+ to_timestamp +------------------------------+ Sat Jan 12 00:00:00 1985 PST+(1 row)+
SELECT to_timestamp('My birthday-> Year: 1976, Month: May, Day: 16',
- '"My birthday-> Year" YYYY, "Month:" FMMonth, "Day:" DD');+ '"My birthday-> Year:" YYYY, "Month:" FMMonth, "Day:" DD');
to_timestamp
------------------------------
Sun May 16 00:00:00 1976 PDT
@@ -2789,7 +2807,7 @@ SELECT to_timestamp('1,582nd VIII 21', 'Y,YYYth FMRM DD');
(1 row)
SELECT to_timestamp('15 "text between quote marks" 98 54 45',
- E'HH24 "\\text between quote marks\\"" YY MI SS');+ E'HH24 "\\"text between quote marks\\"" YY MI SS');
to_timestamp
------------------------------
Thu Jan 01 15:54:45 1998 PST
@@ -2810,6 +2828,21 @@ SELECT to_timestamp('2000January09Sunday', 'YYYYFMMonthDDFMDay');
SELECT to_timestamp('97/Feb/16', 'YYMonDD');
ERROR: invalid value "/Fe" for "Mon"
DETAIL: The given value did not match any of the allowed values for this field.
+SELECT to_timestamp('97/Feb/16', 'YY:Mon:DD');+ to_timestamp +------------------------------+ Sun Feb 16 00:00:00 1997 PST+(1 row)++SELECT to_timestamp('97/Feb/16', 'FXYY:Mon:DD');+ERROR: unexpected character "/", expected separator character ":"+HINT: In FX mode, punctuation in the input string must exactly match the format string.+SELECT to_timestamp('97/Feb/16', 'FXYY/Mon/DD');+ to_timestamp +------------------------------+ Sun Feb 16 00:00:00 1997 PST+(1 row)+
SELECT to_timestamp('19971116', 'YYYYMMDD');
to_timestamp
------------------------------
@@ -2966,7 +2999,7 @@ SELECT to_timestamp('2011-12-18 11:38 20', 'YYYY-MM-DD HH12:MI TZM');
SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
to_timestamp
------------------------------
- Sun Dec 18 03:38:15 2011 PST+ Sun Dec 18 23:38:15 2011 PST
(1 row)
SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
@@ -2996,7 +3029,31 @@ SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
to_timestamp
------------------------------
- Sun Dec 18 03:38:15 2011 PST+ Sun Dec 18 23:38:15 2011 PST+(1 row)++SELECT to_timestamp('2000+ JUN', 'YYYY/MON');+ to_timestamp +------------------------------+ Thu Jun 01 00:00:00 2000 PDT+(1 row)++SELECT to_timestamp('2000 +JUN', 'YYYY/MON');+ to_timestamp +------------------------------+ Thu Jun 01 00:00:00 2000 PDT+(1 row)++SELECT to_timestamp('2000 +JUN', 'YYYY//MON');+ to_timestamp +------------------------------+ Thu Jun 01 00:00:00 2000 PDT+(1 row)++SELECT to_timestamp('2000 +JUN', 'YYYY//MON');+ to_timestamp +------------------------------+ Thu Jun 01 00:00:00 2000 PDT
(1 row)
SELECT to_date('2011 12 18', 'YYYY MM DD');
@@ -3014,13 +3071,13 @@ SELECT to_date('2011 12 18', 'YYYY MM DD');
SELECT to_date('2011 12 18', 'YYYY MM DD');
to_date
------------
- 12-08-2011+ 12-18-2011
(1 row)
SELECT to_date('2011 12 18', 'YYYY MM DD');
to_date
------------
- 02-18-2011+ 12-18-2011
(1 row)
SELECT to_date('2011 12 18', 'YYYY MM DD');
diff --git a/src/test/regress/sql/horology.sql b/src/test/regress/sql/horology.sqlindex ebb196a1cf..e9accdf30c 100644--- a/src/test/regress/sql/horology.sql+++ b/src/test/regress/sql/horology.sql@@ -392,15 +392,21 @@ SELECT to_timestamp('0097/Feb/16 --> 08:14:30', 'YYYY/Mon/DD --> HH:MI:SS');
SELECT to_timestamp('97/2/16 8:14:30', 'FMYYYY/FMMM/FMDD FMHH:FMMI:FMSS');
+SELECT to_timestamp('2011$03!18 23_38_15', 'YYYY-MM-DD HH24:MI:SS');+
SELECT to_timestamp('1985 January 12', 'YYYY FMMonth DD');
+SELECT to_timestamp('1985 FMMonth 12', 'YYYY "FMMonth" DD');++SELECT to_timestamp('1985 \ 12', 'YYYY \\ DD');+
SELECT to_timestamp('My birthday-> Year: 1976, Month: May, Day: 16',
- '"My birthday-> Year" YYYY, "Month:" FMMonth, "Day:" DD');+ '"My birthday-> Year:" YYYY, "Month:" FMMonth, "Day:" DD');
SELECT to_timestamp('1,582nd VIII 21', 'Y,YYYth FMRM DD');
SELECT to_timestamp('15 "text between quote marks" 98 54 45',
- E'HH24 "\\text between quote marks\\"" YY MI SS');+ E'HH24 "\\"text between quote marks\\"" YY MI SS');
SELECT to_timestamp('05121445482000', 'MMDDHH24MISSYYYY');
@@ -408,6 +414,12 @@ SELECT to_timestamp('2000January09Sunday', 'YYYYFMMonthDDFMDay');
SELECT to_timestamp('97/Feb/16', 'YYMonDD');
+SELECT to_timestamp('97/Feb/16', 'YY:Mon:DD');++SELECT to_timestamp('97/Feb/16', 'FXYY:Mon:DD');++SELECT to_timestamp('97/Feb/16', 'FXYY/Mon/DD');+
SELECT to_timestamp('19971116', 'YYYYMMDD');
SELECT to_timestamp('20000-1116', 'YYYY-MMDD');
@@ -464,6 +476,11 @@ SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
SELECT to_timestamp('2011-12-18 23:38:15', 'YYYY-MM-DD HH24:MI:SS');
+SELECT to_timestamp('2000+ JUN', 'YYYY/MON');+SELECT to_timestamp('2000 +JUN', 'YYYY/MON');+SELECT to_timestamp('2000 +JUN', 'YYYY//MON');+SELECT to_timestamp('2000 +JUN', 'YYYY//MON');+
SELECT to_date('2011 12 18', 'YYYY MM DD');
SELECT to_date('2011 12 18', 'YYYY MM DD');
SELECT to_date('2011 12 18', 'YYYY MM DD');
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-hackers@postgresql.org
Cc: a.zakirov@postgrespro.ru, 9erthalion6@gmail.com, robertmhaas@gmail.com, thomas.munro@enterprisedb.com, tgl@sss.pgh.pa.us, sulamul@gmail.com, pavel.stehule@gmail.com, david.g.johnston@gmail.com, a.ignatov@postgrespro.ru, bruce@momjian.us, sul_amul@yahoo.co.in
Subject: Re: [HACKERS] Bug in to_timestamp().
In-Reply-To: <20180217090239.GA14632@arthur.localdomain>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox