autoreconf errored out with:
src/lib/datasrc/Makefile.am:34: `pkglibexecdir' is not a legitimate directory for `LTLIBRARIES'
and consecutive errors such as:
src/lib/datasrc/Makefile.am:43: variable `memory_ds_la_SOURCES' is defined but no program or
src/lib/datasrc/Makefile.am:43: library has `memory_ds_la' as canonical name (possible typo)
src/lib/datasrc/Makefile.am:36: variable `sqlite3_ds_la_SOURCES' is defined but no program or
src/lib/datasrc/Makefile.am:36: library has `sqlite3_ds_la' as canonical name (possible typo)
src/lib/datasrc/Makefile.am:45: variable `memory_ds_la_LIBADD' is defined but no program or
src/lib/datasrc/Makefile.am:45: library has `memory_ds_la' as canonical name (possible typo)
src/lib/datasrc/Makefile.am:39: variable `sqlite3_ds_la_LIBADD' is defined but no program or
src/lib/datasrc/Makefile.am:39: library has `sqlite3_ds_la' as canonical name (possible typo)
src/lib/datasrc/Makefile.am:44: variable `memory_ds_la_LDFLAGS' is defined but no program or
src/lib/datasrc/Makefile.am:44: library has `memory_ds_la' as canonical name (possible typo)
src/lib/datasrc/Makefile.am:37: variable `sqlite3_ds_la_LDFLAGS' is defined but no program or
src/lib/datasrc/Makefile.am:37: library has `sqlite3_ds_la' as canonical name (possible typo)
According to automake "bug#7647: Diagnose invalid primary/directory combinations"
as in http://www.mail-archive.com/bug-automake@gnu.org/msg01927.html
we nowadays have to use pkglibdir for LTLIBRARIES, not pkglibexecdir.
+2010-12-17 Stefano Lattarini <stefano.lattar...@gmail.com>
+
+ Warnings about primary/prefix mismatch fixed and extended.
+ * automake.in (%standard_prefix): Add `doc' and `locale'.
+ Rename `pkgdatadir' to `pkgdata'. Similarly for`pkglibdir',
+ `pkgincludedir' and `pkglibexecdir'.
+ (handle_programs): List `pkglibexec', not `pkglib', among the
+ prefixes valid for the `PROGRAMS' primary.
[]
+for primary in $primaries; do
+ prefixes_ok=''
+ case $primary in
+ LIBRARIES|LTLIBRARIES)
+ prefixes_ok='lib pkglib'
+ ;;
+ PROGRAMS)
+ prefixes_ok='bin sbin libexec pkglibexec'
+ ;;
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
If no botan-config was found we would end up doing:
checking for botan-config... no
/scratch/src/bind10/configure: line 16704: --libs: command not found
/scratch/src/bind10/configure: line 16705: --cflags: command not found
Instead, do not run botan-config if there was no botan-config found
as we will bail out gracefully later on if the botan headers could not
be found.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
JINMEI Tatuya [Thu, 16 Feb 2012 08:19:11 +0000 (00:19 -0800)]
[1580] a suggested cleanup: rename addNSCOMAINProof to addNSCOMAINProofByNSEC
so it's clearer that it's NSEC specific (now that we have an NSEC3 version
of addNXDOMAINProof). For consistency rename the NSEC3 version to
addNSCOMAINProofByNSEC3. also a bit reorganized the code calling these methods
(mostly a matter of taste though), and adjusted tests to make them pass.
JINMEI Tatuya [Thu, 16 Feb 2012 07:38:54 +0000 (23:38 -0800)]
[1580] updated addNSEC3NXDOMAINProof() so we don't check duplicate NSEC3 for
now, as we discussed in #1583. Adjusted the nxdomainWithNSEC3Proof test
accordingly.
JINMEI Tatuya [Thu, 16 Feb 2012 07:20:36 +0000 (23:20 -0800)]
[1580] simplified the cases with unexpected NULL results of findNSEC3().
adjusted the expected exception for nxdomainWithBadNextNSEC3Proof accordingly.
JINMEI Tatuya [Thu, 16 Feb 2012 07:07:45 +0000 (23:07 -0800)]
[1580] corrected the comment about how to construct the corresponding
wildcard name (the previous one was seemingly copied from the NSEC
counterpart, which is not applicable here).
Or, everybody who accepts a shutdown command with pid now. Stats needed
to be updated because it did call functions based on the name and spec
file somehow.
JINMEI Tatuya [Tue, 14 Feb 2012 02:06:51 +0000 (18:06 -0800)]
[1583] one more cleanup: removed the wildcardEmptyWithNSEC3 test. This case
is no different from the non-empty wildcard NXRRRSET in terms of query
processing.
JINMEI Tatuya [Tue, 14 Feb 2012 01:44:19 +0000 (17:44 -0800)]
[1583] added a test case of bogus findNSEC3 result for the wildcard NXRRSET case.
also made a non-related cleanup: removed resetting of nsec3_fake_ in the
fake findNSEC3() method. This reset wasn't really necessary, and unless/until
we need that it'd be better to limit the place of tweaking such mutable
variables.
JINMEI Tatuya [Tue, 14 Feb 2012 00:52:06 +0000 (16:52 -0800)]
[1583] supported the normal case of wildcard no data response with NSEC3.
the older version of wildcardNxrrsetWithNSEC3 was removed. also extended
some existing tests so it's easier to detect the cause when subsequent
updates introduce a regression.