On Mon, Aug 20, 2007 at 09:23:25PM -0400, Tom Lane wrote:
> I've applied version 0.58 of the patch with a lot of further
> editorializing. I feel fairly confident now in the code that interfaces
> between tsearch and the rest of the system, but a lot of the
> lowest-level "guts" of tsearch (mainly in src/backend/tsearch/*.c and
> src/backend/utils/adt/ts*.c) left my eyes glazing over. Perhaps someone
> else can make an extra review pass over that stuff.
> I am quite confident that this commit broke the MSVC build, which seems
> to need to know individually about each shared library ... Magnus,
> can you do something about that? We'll see what other portability
> problems emerge from the buildfarm.
Looking at that now.
I get awarning from the following line in bison generated parse.h:
#define TEXT 577
Because TEXT is a macro that's used in the windows headers. (A redefinition
Any chance to change that, or is it coming out of the syntax itself? (bison
newbie here, as you know :-P)
> Also, we need to decide what to do with contrib/tsearch2, which is
> currently DOA because of conflicts with the new core code. We could
> either rip it out entirely, or try to convert it into a compatibility
> package. In view of the renamings of functions we agreed to do, I
> think there is some scope for a compatibility package, but I have no
> time to work on that.
OTOH, if we do it as a compat package, we need to set a firm end-date on
it, so we don't have to maintain it forever. Given the issues always at
hand for doing such an upgrade, my vote is actually for ripping it out
completely and take the migration pain once and then be done with it.
> This is, by a wide margin, the largest single patch ever to hit the
> Postgres CVS tree. Congratulations to Oleg and Teodor on seeing
> it through!
Yes, congratulations indeed. Great to see this in!
---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?