Re: pgadmin3 on FreeBSD4.8

Hi Dave.

From: "Dave Page" <dpage(at)vale-housing(dot)co(dot)uk>
> > From: Hiroshi Saito [mailto:saito(at)inetrt(dot)skcapi(dot)co(dot)jp]
(snip)
> > I am the environment that FreeBSD4.8 is important.
(snip)
> :-(. Is that just because of the older version of FreeBSD?

patch-kit of the ports collection and the package hadn't satisfied it yet
with 5.x in Japan.
It seems that it was very much gathered at present.
However, the user who still hopes for the stability is using 4.x.

> Do you have a test database you can email me that I can play with
> please?

pgdump-sql that it Andreas play here, is put.
http://cre-ent.skcapi.co.jp/~saito/pgadmin/lib/pgadmin3_unicode_debug6.sql

> > BTW.
> > Attached patch is necessary with this environment.
> > But, Is it likely to be unnecessary with FreeBSD5.1?
>
> I don't think so - it compiles and runs OK on 5.1 with tweaks to
> configure. Is there anything defined that we could use easily to detect
> FreeBSD 4 or 5?

It doesn't seem to be a problem with 5.1.

But, make with UNICODE is difficult with 4.8.

c++ -c -I./lib/wx/include/gtk2ud-2.5 -I../include -I../src/zlib -pthread -
I/usr/local/include/gtk-2.0 -I/usr/local/lib/gtk-2.0/include -I/usr/local/in
clude/atk-1.0 -I/usr/local/include/pango-1.0 -I/usr/X11R6/include -I/usr/loc
al/include/glib-2.0 -I/usr/local/lib/glib-2.0/include -I/usr/local/include -
D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -D__WXGTK__ -D__WXDEBUG__ -g -O0 -MM
D -pthread -D_THREAD_SAFE -Wall -o accel.o ../src/generic/accel.cpp
In file included from ../include/wx/memory.h:20,
from ../include/wx/object.h:25,
from ../include/wx/event.h:20,
from ../src/generic/accel.cpp:34:
../include/wx/string.h: In method `bool wxString::IsSameAs(__wchar_t, bool =
1) const':
../include/wx/string.h:653: implicit declaration of function `int
towupper(...)'
../src/generic/accel.cpp: In method
`wxAcceleratorTable::wxAcceleratorTable(int, const wxAcceleratorEntry *)':
../src/generic/accel.cpp:98: implicit declaration of function `int
iswlower(...)'
gmake: *** [accel.o] error 1

Because, it isn't defined fundamentally.
/usr/include/wctype.h
#if 0
/* XXX: not implemented */
.....
wint_t towlower __P((wint_t));
.....
#endif

By one, this is the case that it is not UNICODE, and it is about the
attached patch.
atoll() is glibc2.0.2 higher.
It doesn't go into 4.8.

But, these are not problems if 4.8 is the policy which isn't made a support
object.
I can move by the thing which is not UNICODE at least. (single-byte)
I may have to have 5.1. But, I want to solve this.
Until it can be used in Japanese, ...

Regards,
Hiroshi-Saito