The MUA Mutt uses the options framework, with the option "mutt-hcache"
default enabled. It needs some database too, and there are some options
where you can select something different from the default, which is db4:
###
### Header cache
###
.if !empty(PKG_OPTIONS:Mmutt-hcache)
. if !empty(PKG_OPTIONS:Mtokyocabinet)
. include "../../databases/tokyocabinet/buildlink3.mk"
CONFIGURE_ARGS+= --enable-hcache
CONFIGURE_ARGS+= --enable-tokyocabinet
CONFIGURE_ARGS+= --without-gdbm
CONFIGURE_ARGS+= --without-bdb
. else
BDB_ACCEPTED= db4 db5
BUILDLINK_TRANSFORM+= l:db:${BDB_TYPE}
. include "../../mk/bdb.buildlink3.mk"
CONFIGURE_ARGS+= --enable-hcache
CONFIGURE_ARGS+= --without-gdbm
# BDB_INCLUDE_DIR_ and BDB_LIB_DIR don't have to be particularly accurate
# since the real -I and -L flags are added by buildlink already.
CONFIGURE_ENV+= BDB_INCLUDE_DIR=${BDBBASE}/include
CONFIGURE_ENV+= BDB_LIB_DIR=${BDBBASE}/lib
CONFIGURE_ENV+= BDB_LIB=${BDB_LIBS:S/^-l//:M*:Q}
. endif
.else
CONFIGURE_ARGS+= --disable-hcache
.endif
However when building, configure complained it could not find a suitable
database. Indeed, some debugging showed that db1 was selected somehow,
even though "BDB_ACCEPTED=db4 db5".
It turned out in the end that "../../mk/bdb.buildlink3.mk" was
prematuredly included, before BDB_ACCEPTED was set. Indeed it happened
as early as somewhere during the
.include "../../mk/bsd.options.mk"
near the top of the file.
When I moved the BDB_ACCEPTED line before that, all was ok, but that
looks ugly and is probably not expected behaviour.
Any thoughts, other than just moving the line?
In any case a fix should probably be pulled up to the branch.
-Olaf.
--
Olaf 'Rhialto' Seibert -- rhialto at falu dot nl
___ Anyone who is capable of getting themselves made President should on
\X/ no account be allowed to do the job. --Douglas Adams, "THGTTG"
Attachment:
signature.asc
Description: PGP signature