Our tests were not validating the return value in all cases, nor was
it guaranteeing our documented claim that 'res' is unchanged on error.
For that matter, it wasn't as thorough as the existing tests for
qemu_strtoi() and friends for proving that endptr and res are sanely
set. Enhancing the test found one case where we violated our
documentation: namely, when failing with EINVAL when endptr is NULL,
we shouldn't modify res.
Signed-off-by: Eric Blake <eblake@redhat.com>
---
tests/unit/test-cutils.c | 117 +++++++++++++++++++++++++++++++++++++--
util/cutils.c | 4 +-
2 files changed, 114 insertions(+), 7 deletions(-)