increase message string buffer size of watch command of psql

On Sun, Jun 12, 2016 at 10:55 AM, Ioseph Kim <pgsql-kr@postgresql.kr> wrote:

Hello.
In po.ko (korean message) at psql
#: command.c:2971
#, c-format
msgid "Watch every %lds\t%s"
msgstr "%ld초 간격으로 지켜보기\t%s"

this message string is a cut string, because buffer size is small.
At line 2946 in src/bin/psql/command.c
char title[50];

size of message string for korean is over 50 bytes.
(at least 80 columns terminal for common)

Increase size of this title, please.
50 bytes is so small for multi language.

And. I suggest that date string might be local language,
or current_timestamp string.

This was already changed in commit dea2b5960.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Robert Haas <robertmhaas@gmail.com> writes:

On Sun, Jun 12, 2016 at 10:55 AM, Ioseph Kim <pgsql-kr@postgresql.kr> wrote:

Increase size of this title, please.
50 bytes is so small for multi language.

And. I suggest that date string might be local language,
or current_timestamp string.

This was already changed in commit dea2b5960.

Well, we did part of that, but it's still using asctime(). Should we
change that to strftime(..."%c"...) to be less English-centric?
(The result seems to be the same in C locale. pg_controldata has done
it that way for a long time, with few complaints.) If we want to do so,
now would be the time, since 9.6 already whacked around the format
of \watch output.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

I wrote:

Robert Haas <robertmhaas@gmail.com> writes:

On Sun, Jun 12, 2016 at 10:55 AM, Ioseph Kim <pgsql-kr@postgresql.kr> wrote:

Increase size of this title, please.
50 bytes is so small for multi language.
And. I suggest that date string might be local language,
or current_timestamp string.

This was already changed in commit dea2b5960.

Well, we did part of that, but it's still using asctime(). Should we
change that to strftime(..."%c"...) to be less English-centric?
(The result seems to be the same in C locale. pg_controldata has done
it that way for a long time, with few complaints.) If we want to do so,
now would be the time, since 9.6 already whacked around the format
of \watch output.

I take it from the vast silence that nobody particularly cares one way
or the other. On reflection I think that this would be a good change
to make, so I'll go do so unless I hear complaints soon.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

2016년 06월 15일 01:56에 Tom Lane 이(가) 쓴 글:

I take it from the vast silence that nobody particularly cares one way
or the other. On reflection I think that this would be a good change
to make, so I'll go do so unless I hear complaints soon. regards, tom
lane

I propose to change from asctime() to sql current_timestamp value,
then users will change date format with set command DateStyle.

Regards, Ioseph.

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Ioseph Kim <pgsql-kr@postgresql.kr> writes:

2016년 06월 15일 01:56에 Tom Lane 이(가) 쓴 글:

I take it from the vast silence that nobody particularly cares one way
or the other. On reflection I think that this would be a good change
to make, so I'll go do so unless I hear complaints soon. regards, tom
lane

I propose to change from asctime() to sql current_timestamp value,
then users will change date format with set command DateStyle.

That would require an additional SQL query each time through the loop,
which seems like undue expense. It's also not terribly friendly to the
goal of localization, I should think, given the limited number of
datestyle options and the fact that none of them actually change day or
month names to non-English choices. And it would imply changing the
timestamps from psql's timezone to the backend's. While that might have
been a good way to do it in a green field, it's not the way \watch has
worked in the past, and given the lack of complaints I'm disinclined
to change that.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Tom Lane wrote:

I wrote:

Well, we did part of that, but it's still using asctime(). Should we
change that to strftime(..."%c"...) to be less English-centric?
(The result seems to be the same in C locale. pg_controldata has done
it that way for a long time, with few complaints.) If we want to do so,
now would be the time, since 9.6 already whacked around the format
of \watch output.

I take it from the vast silence that nobody particularly cares one way
or the other. On reflection I think that this would be a good change
to make, so I'll go do so unless I hear complaints soon.

+1 to strftime("%c"). If we wanted to provide additional flexibility we
could have a \pset option to change the strftime format string to
something other than %c, but I don't think there's enough demand to
justify it.

--
�lvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Alvaro Herrera <alvherre@2ndquadrant.com> writes:

+1 to strftime("%c"). If we wanted to provide additional flexibility we
could have a \pset option to change the strftime format string to
something other than %c, but I don't think there's enough demand to
justify it.

Agreed, that seems like something for later (or never). Pushed that way.

I also widened the buffer a bit in the back branches, to address the
original complaint.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Thanks, I agree that strftime() is better then asctime().

regards, Ioseph

2016占쏙옙 06占쏙옙 16占쏙옙 08:37占쏙옙 Tom Lane 占쏙옙(占쏙옙) 占쏙옙 占쏙옙:

Alvaro Herrera <alvherre@2ndquadrant.com> writes:

+1 to strftime("%c"). If we wanted to provide additional flexibility we
could have a \pset option to change the strftime format string to
something other than %c, but I don't think there's enough demand to
justify it.

Agreed, that seems like something for later (or never). Pushed that way.

I also widened the buffer a bit in the back branches, to address the
original complaint.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers