The following bug has been logged online:

Bug reference: 2632
Logged by: Massimo Fidanza
Email address: max (AT) massimofidanza (DOT) it
PostgreSQL version: 8.1.4
Operating system: Ubuntu 6.04
Description: createuser language poblem
Details:

Hi, I discover a very annoying bug in the create user command. I have the
environment variable LANG=it_IT.UTF-8 and when I try to crate a new user wit
the create user command, at he prompt I have the three questions
Shall the new role be a superuser? (y/n)
Shall the new role be allowed to create databases? (y/n)
Shall the new role be allowed to create more new roles? (y/n)
So I ask y only to the second, but when I try to create a new database with
the new just created user it claims that I don't have permissions. So I drop
the user and recreated it several times. After a lot of tentative I discover
the arcane I must ask s=sì the italian affermative answer during
createuser, but the question is in english. This is a very stupid bug. I
think the right is to accept the italian and the english answer as oher
commands do like apt. Translate the question to is not a bad idea.

---------------------------(end of broadcast)---------------------------
TIP 1: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo (AT) postgresql (DOT) org so that your
message can get through to the mailing list cleanly

"Massimo Fidanza" <max (AT) massimofidanza (DOT) it> writes:
The reason those aren't getting translated is that the Italian
translations have not been updated for 8.1 --- the entries in it.po
still look like the 8.0 wording:

#: createuser.c:170
msgid "Shall the new user be allowed to create databases? (y/n) "
msgstr "Il nuovo utente può creare database? (s/n)"

#: createuser.c:181
msgid "Shall the new user be allowed to create more new users? (y/n) "
msgstr "Il nuovo utente può creare altri utenti? (s/n)"

If this annoys you, please join the PG translation project and help to
fix it:
http://pgtranslation.projects.postgresql.org/

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

Tom Lane wrote:
It might be possible with little effort to rework this so that the
prompts display the letters that are actually expected as answers.
I'll look into it.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

This message is in MIME format.

--=_5x5pbtclm0co
Content-Type: text/plain;
charset=ISO-8859-1;
DelSp="Yes";
format="flowed"
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Quoting Tom Lane <tgl (AT) sss (DOT) pgh.pa.us>:

Yes, I can join the translation team and translate the missing strigs,=20=
=20
but I think this is not the right answer. I think the best solution is=20=
=20
to accept the english answer too.
I made a patch to common.c (src/bin/scripts/common.c) that compare the=20=
=20
user answer with the translated and the english affermative (y) and=20=20
negative (n) answer. I add a case insensitive comparison too.
I know that this can generate conflicts between languages, but looking=20=
=20
at the other languages in wich Postgresql is already translated there=20=20
are no conflict between english and non english languages. I look for=20=20
other languages too and I find a possible conflict with african=20=20
Swahili language, where the answers are (ndiyo=3D=3Dyes/kataa=3D=3Dno), but=
=20=20
they can for example specify (y/n) as (nd/k).

Bye Max!

--=_5x5pbtclm0co
Content-Type: text/plain;
charset=UTF-8;
name="common.c.diff"
Content-Disposition: attachment;
filename="common.c.diff"
Content-Transfer-Encoding: 7bit

--- common.c 2006-09-19 22:41:35.000000000 +0200
+++ common.c 2006-09-19 22:45:29.000000000 +0200
@@ -25,2 +25,5 @@

+#ifdef WIN32
+#define strcasecmp(s1,s2) lstrcmpiA((s1), (s2))
+#endif

@@ -208,5 +211,5 @@
{
- if (strcmp(string, _(PG_YESLETTER)) == 0)
+ if ((strcasecmp(string, _(PG_YESLETTER)) == 0) || (strcasecmp(string, "y") == 0))
return 1;
- else if (strcmp(string, _(PG_NOLETTER)) == 0)
+ else if ((strcasecmp(string, _(PG_NOLETTER)) == 0) || (strcasecmp(string, "n") == 0))
return 0;

--=_5x5pbtclm0co
Content-Type: text/plain
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0


---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend

--=_5x5pbtclm0co--

Massimo Fidanza <max (AT) massimofidanza (DOT) it> writes:
I liked Peter's idea better (ie, not have the y/n in the
string-to-be-translated but generate it dynamically using the same
strings that will actually be matched to the input). Your approach
fails if there are any languages that use y or n in a way conflicting
with the English usage ... and whether or not we have any such
translations today, who knows what will be out there tomorrow?

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 1: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo (AT) postgresql (DOT) org so that your
message can get through to the mailing list cleanly