Re: Unicode UTF-8 table formatting for psql text output

On Sunday 27 September 2009 19:03:33 Robert Haas wrote:
> On Sun, Sep 27, 2009 at 9:24 PM, Selena Deckelmann
>
> <selenamarie(at)gmail(dot)com> wrote:
> > Hi!
> >
> > On Wed, Sep 23, 2009 at 2:16 AM, Roger Leigh <rleigh(at)codelibre(dot)net> wrote:
> >> On Fri, Sep 18, 2009 at 11:30:05AM -0700, Selena Deckelmann wrote:
> >>> Brad says:
> >>>
> >>>        The patched code compiles without any additional warnings.
> >>> Lint gripes about a trailing ',' in 'typedef enum printTextRule' in
> >>> print.h.  Other additional lint seem to be false positives.  The
> >>> regression tests pass against the new patch.
> >>
> >> I've attached a new patch which tidies up those extra commas, plus
> >> a patch showing the changes from the previous patch.
> >
> > Great!  Thank you.
> >
> > Brad -- can you review this patch? Is it ready for a committer?
>
> Brad already marked it that way on the CommitFest application, so I
> think that probably means yes. :-)
>
> ...Robert

I looked at the new (psql-utf8-table-4.patch) patch. I think the style issues are fixed.

During this review I found that `gmake check` will fail when LANG=en_US.UTF-8 in the environment. In this case
the patched psql produces UTF8 line art and the tests expect ASCII line art.

pg_regress clears LC_ALL by default, but does not clear LANG by default. Please find attached a patch that
causes pg_regress to also clear LANG by default.

Thanks,

--bts