No c+p of warnings please! Use gnupg-devel for such things.

No c+p of warnings please! Use gnupg-devel for such things.

Changed status to 'unread'. I'm not chatting.

Updated cert.c which initializes the variable algo.

Updated status to 'unread'. I'm not chatting.

Updated to include line numbers.

54400==ERROR: LeakSanitizer: detected memory leaks

Direct leak of 240 byte(s) in 12 object(s) allocated from:

#0 0x49f45b in __interceptor_malloc

/home/gpg-user/Clang-3.5/llvm/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:40:3

#1 0x5462d0 in _ksba_oid_to_str

/home/gpg-user/gcrypt-2.0-sanitize/libksba-1.3.2/src/oid.c:75:16

#2 0x4bcce9 in main

/home/gpg-user/gcrypt-2.0-sanitize/libksba-1.3.2/tests/t-oid.c:164:7

#3 0x2b8edcf1aec4 in __libc_start_main

/build/buildd/eglibc-2.19/csu/libc-start.c:287

Direct leak of 58 byte(s) in 2 object(s) allocated from:

#0 0x49f45b in __interceptor_malloc

/home/gpg-user/Clang-3.5/llvm/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:40:3

#1 0x5570e4 in _ksba_strdup

/home/gpg-user/gcrypt-2.0-sanitize/libksba-1.3.2/src/util.c:162:13

#2 0x4bcce9 in main

/home/gpg-user/gcrypt-2.0-sanitize/libksba-1.3.2/tests/t-oid.c:164:7

#3 0x2b8edcf1aec4 in __libc_start_main

/build/buildd/eglibc-2.19/csu/libc-start.c:287

SUMMARY: AddressSanitizer: 298 byte(s) leaked in 14 allocation(s).

FAIL: t-oid

1 of 4 tests failed

Please report to http://bugs.gnupg.org

Attached is the script I am using to acceptance test the suite. It requires
Clang 3.5 (Clang 3.5 recipe was provided with Bug 1872).

Ah yes, I had that before in our build system where RUNPATH is passed via LD_OPTIONS.
If I pass the flags via LDFLAGS it works.

Thanks! -- Dago

That pretty much looks like you are using this new test program with an older
Libksba.

Duplicate of T1467

See the other bug. Will be fixed with the next release.

This happens because AM_PATH_GPG_ERROR uses AC_ARG_WITH
in a bad way. First it parses the official name, which
is --with-libgpg-error-prefix. But then it tries to implement
a fallback to the old undocumented option name
--with-gpg-error-prefix. Unfortunately, that fallback
unconditionally overwrites the result of the first
AC_ARG_WITH.

The enclosed patch fixes this issue.

This patch needs to be installed in both the libksba
and the gnupg repositories.

The same issue was however fixed in the libassuan repo
in 97ce28a430129ce997783c6196ccfe737f5b3007. Applying
that solution in the libksba and gnupg repos would work
just as fine as my patch (and reduce the proliferation
of differing versions).

I think T1526 is a duplicate of this bug.

I think T1561 is the same bug, but in the gnupg
repository.

D166: 424_0001-configure-fix-with-libgpg-error-prefix.patch

Fixed with commit ab3fe5d.

The grammar is quite old and should anyway adjusted to modern standard. There
is a separately mainatained toke table which does not make much sense, given
that that it is possible to re-use the bison generated token table. I see what
I can do.

D184: 409_libksba-1.3.0-undef-true.patch

Good catch. Fixed in SVN. Thanks.

That was just a stupid error, not catched for unknown reasons by gcc. Already
marked as resolved in Gentoo, so we can set it to resolved too. 1.0.3 will be
released RSN.

D46: 146_libksba-1.0.2-segfault.patch

I would love to be able to add my-self to CC of issues...

issue 582 as been reopened. Please see there.

Duplicate of T582

The problem is due to an incosnsistent CA. Running with --debug 1 will give
more clues on this. Solved by direct discussion, thus closing.

Bogus bug report.

Added gnulib stuff. Inserted alloca.h.

Fixed in 0.9.11 released today

Needed to use per-target flags to rename object used by for
the library and the build time helpers.

I can duplicate that problem. Thanks.

We have upgraded to automake 1.9.3 and fixed al tehse issues.