1) Compile pgAdmin3 in non-unicode mode. In which case, pgAdmin3 port would be unable to display non-european languages and query PostgreSQL Unicode databases.I like this order as well (with one caveat), although I think that #5 is unnecessary. Whatever method we use for compiling pgAdmin3, distribution will almost certainly be done as in binary form, so there will be no need to involve fink, which AIUI, is largely for compiling ported packages for Mac OS X.
2) Use the "iconv" library linked against wxWindows in order to bring a Unicode abstraction layer to wxWindows. Julian, I had this idea in my sleep, is this a workable solution?
3) Wait for MacOsX 10.3 which should be Unicode compliant. We already discussed this opportunity, which does not fit our ideas because it would require MacOsX users to upgrade.
4) Use donated proprietary compilers in order to compile pgAdmin3 under MacOsX. My opinion is not to use proprietary software, but other members in the team are not agains this idea. This needs to be dicussed.
5) Compile pgAdmin as a fink package (http://www.fink.org). Maybe too complicated for MacOsX users.
Could we discuss these opportunities? My opinion would be to go for 2), 1) and 4) and 5) in order of preference. What is the opinion of other members and contributors?
My caveat to the above order is that item #3, assuming we have no problem with the compiler, should be #1, since it offers us the oppurtunity to have a Unicode pgAdmin3 on Mac OS X. If there are philosophical problems with #3, then let's compile a non-unicode version, at least to get something to beta test. There's a pretty good chance that we're going to have bugs in the Mac version, since the last time I had pgAdmin3 running on the Mac, it crashed on startup.
As soon as we decide which direction we want to go, I can contribute however much time is neccessary to get the port to work.
ahp
---------------------------(end of broadcast)--------------------------- TIP 8: explain analyze is your friend
