Postgres Pro
Downloads
Home   >   mailing lists

Re: [GENERAL] ERROR: functions in index expression must be marked IMMUTABLE - Mailing list pgsql-general

From David G. Johnston
Subject Re: [GENERAL] ERROR: functions in index expression must be marked IMMUTABLE
Date
Msg-id CAKFQuwZ4DHu9zcdDkrPx-v2_=JCD0vbD-NbvRyORikQEH5VV9w@mail.gmail.com
Whole thread Raw
In response to Re: [GENERAL] ERROR: functions in index expression must be marked IMMUTABLE  (Geoff Winkless <pgsqladmin@geoff.dj>)
Responses Re: [GENERAL] ERROR: functions in index expression must be markedIMMUTABLE  ("Sven R. Kunze" <srkunze@mail.de>)
List pgsql-general
Tree view
On Tue, Feb 28, 2017 at 9:35 AM, Geoff Winkless <pgsqladmin@geoff.dj> wrote:
On 28 February 2017 at 15:59, Adrian Klaver <adrian.klaver@aklaver.com> wrote:
On 02/28/2017 07:30 AM, Sven R. Kunze wrote:
On 28.02.2017 15:40, Adrian Klaver wrote:
[explanation of why date casting and to_datetime don't work]

Why is to_date not immutable?

 Not sure, but if I where to hazard a guess, from the source code in formatting.c:

https://git.postgresql.org/gitweb/?p=postgresql.git;a=blob;f=src/backend/utils/adt/formatting.c;h=e552c8d20b61a082049068d2f8d776e35fef1179;hb=HEAD

​Would the fact that you can have month names in to_date strings make it dependent on current locale?


​That would seem to be it.

cache_locale_time() at the top of DCH_to_char which is in the call stack of the shared parsing code for both to_date and to_timestamp.

​https://git.postgresql.org/gitweb/?p=postgresql.git;a=blob;f=src/backend/utils/adt/formatting.c;h=e552c8d20b61a082049068d2f8d776e35fef1179;hb=HEAD#l2363

Supposedly one could provide a version of to_date that accepts a locale in which to interpret names in the input data - or extend the format string with some kind of "{locale=en_US}" syntax to avoid changing the function signature.

David J.

pgsql-general by date:

Previous
From: Geoff Winkless
Date:
Subject: Re: [GENERAL] ERROR: functions in index expression must be marked IMMUTABLE
Next
From: Ivan Voras
Date:
Subject: [GENERAL] Making a unique constraint deferrable?