It never fails (bsddb retirement in 2.3)

[Nick Vargish]

"Martin v. Löwis" <martin at v.loewis.de> writes:

> No, I envision no such change, and yes, the solution is not perfect
> for him. He has to actively port his code to Python 2.3.

Well, it's not my code. I write it, sure, but it's really the
government's code. Unless I can show that maintaining this code
doesn't require jumping through hoops, getting Python accepted here is
going to be more of an uphill battle. Given how things work here, and
at other large institutions I've worked at, one person doesn't always
have the opportunity to maintain the code they originally wrote.

> I don't think we should attempt to continue to transparently support
> the bsddb 1.85 library. People will need to go through some effort,
> and they may consider whether their effort is better spend elsewhere
> (e.g. in obtaining a copy of Sleepycat, or invoking the database
> upgrade utility). It should be a reasonable effort, where apparently
> changing the source in a few places is reasonable, and convincing
> powers-that-be is not.

I'm trying to get what I can working, between your suggestion to
modify Modules/Setup to enable bsd185 and Skip's recent patch to
anydbm (as found on sf.net under the bug I filed).

I am actively looking into adding "install the Sleepycat bsddb
libraries", but the documentation is not completely clear. Where
should I install these libraries so that they (1) do not interfere
with /lib/libdb.a and (2) are automatically found by Python's
configure script?

Thanks again for all your help,

Nick

-- 
#include<stdio.h> /* SigMask 0.3 (sig.c) 19990429 PUBLIC DOMAIN "Compile Me" */
int main(c,v)char *v;{return !c?putchar(*v-1)&&main(0,v+ /* Tweaks welcomed. */
1):main(0,"Ojdl!Wbshjti!=obwAqbusjpu/ofu?\v\1");}  /* build: cc -o sig sig.c */