Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Peter Geoghegan
Hmmmseem to be getting a weird linker error when I build OGL (from the independent package) against 2.9: /usr/bin/ld: ogl_dll_composit.o: relocation R_X86_64_32 against `wxDivisionControlPoint::ms_classInfo' can not be used when making a shared object; recompile with -fPIC ogl_dll_composit.o:

Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Guillaume Lelarge
Le 17/01/2011 23:20, Dave Page a écrit : > On Mon, Jan 17, 2011 at 10:13 PM, Guillaume Lelarge > wrote: >> Actually, I don't see us maintaining OGL. We don't have the manpower to >> do that. >> >> I'm wondering if we really need OGL. What do we use it for? because if >> it's now out of wxWidgets a

Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Dave Page
On Mon, Jan 17, 2011 at 10:13 PM, Guillaume Lelarge wrote: > Actually, I don't see us maintaining OGL. We don't have the manpower to > do that. > > I'm wondering if we really need OGL. What do we use it for? because if > it's now out of wxWidgets and if it's an important component for us, we > are

Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Dave Page
On Mon, Jan 17, 2011 at 3:30 PM, Peter Geoghegan wrote: >>> in pgsDictionaryGen.cpp, why do we do this?: >>> >>>                        wxString line; >>>                        while ((line = text.ReadLine()) && !input.Eof()) >>>                        { >>>                                ++resul

Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Guillaume Lelarge
Le 17/01/2011 16:30, Peter Geoghegan a écrit : > On 16 January 2011 22:39, Dave Page wrote: >> On Sun, Jan 16, 2011 at 6:47 PM, Peter Geoghegan > [...] >> How did the OGL port go? I looked at that briefly, and had a rough >> build in 10 minutes or so iirc. Oh, and in answer to your previous >> com

Re: [pgadmin-hackers] wxWidgets 2.9 build

2011-01-17 Thread Peter Geoghegan
On 16 January 2011 22:39, Dave Page wrote: > On Sun, Jan 16, 2011 at 6:47 PM, Peter Geoghegan > From what I can see in a quick glance at the docs, the only conversion > operator is to const char* in 2.8, so I guess that is a legacy thing > as you suggest. There are good reasons for the different c