Make the data type explicit in transport mock #1914
Conversation
|
This user does not have permission to start the build. Can one of the admins verify this patch and start the build? (admin please type: ok to test) |
|
@kira-syslogng ok to test |
|
Build SUCCESS, the tests were executed on test branch: master and test suite: functions |
mochrul
force-pushed
the
hurd_compile_fix
branch
from
9279ff3 to
aad4eb8
Compare
|
Build SUCCESS, the tests were executed on test branch: master and test suite: functions |
|
Hi @mochrul! Just a question out of curiosity: |
|
If you check the bug ticket you can see that that is just a replica. Also, I do not use Hurd but I'm the Debian maintainer of the package and Debian wants to compile it on Hurd. Most of my bug reports and patches are related to the Debian packaging. |
|
Thanks for the clarification, I missed to see the 2014 date and that it is
the same as #355.
…On Wed, Mar 7, 2018 at 7:39 PM, Attila Szalay ***@***.***> wrote:
If you check the bug ticket you can see that that is just a replica.
The original ticket is dated back to 2014 when 3.6 is not yet released
officially (or just barely released). I just copied all the information
from that ticket into this new, because that ticket is closed and I was not
able to reopen it. And because the bug is more than 3 years old, I think
fixing this in HEAD is enough.
Also, I do not use Hurd but I'm the Debian maintainer of the package and
Debian wants to compile it on Hurd. Most of my bug reports and patches are
related to the Debian packaging.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#1914 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AYbP5UooY_E6WSztYvGyVJIyx0pnpN23ks5tcCaVgaJpZM4SbfsP>
.
|
libtest/mock-transport.c
Outdated
| self->iov_cnt++; | ||
| g_assert(self->value_cnt < sizeof(self->value) / sizeof(self->value[0])); | ||
|
|
||
| if (length == -2) |
There was a problem hiding this comment.
It is kinda hard to connect this -2 value and the one in the LTM_INJECT_ERROR macro.
Would you create a constant for this ?
#define LTM_INJECT_ERROR_LENGTH -2
#define LTM_INJECT_ERROR(err) (GUINT_TO_POINTER(err)), LTM_INJECT_ERROR_LENGTHThere was a problem hiding this comment.
Thank you, one small thing. Your update does not compile because you forget the -2 from the LTM_InJECT_ERROR_LENGTH definition.
|
This user does not have permission to start the build. Can one of the admins verify this patch and start the build? (admin please type: ok to test) |
|
success |
mochrul
force-pushed
the
hurd_compile_fix
branch
from
aad4eb8 to
12c7298
Compare
|
Build FAILURE, the tests were executed on test branch: master and test suite: functions |
|
success |
mochrul
force-pushed
the
hurd_compile_fix
branch
2 times, most recently
from
16dc6f4 to
8e2b8c9
Compare
|
success |
|
Build FAILURE, the tests were executed on test branch: master and test suite: functions |
|
success |
mochrul
force-pushed
the
hurd_compile_fix
branch
from
8e2b8c9 to
9437ff0
Compare
Previously there were a heuristic in that object which tried to detect if the data is a string or an error code. But this heuristic was failing in GNU Hurd. With this patch the heuristic is replaced with a type field in a structure.
mochrul
force-pushed
the
hurd_compile_fix
branch
from
9437ff0 to
970da15
Compare
mochrul
changed the title
|
Build SUCCESS, the tests were executed on test branch: master and test suite: functions |
|
success |
Previously there was a heuristic in that object
which tried to detect if the data is a string or an
error code. But this heuristic was failing in GNU Hurd.
With this patch, the heuristic is replaced with a
type field in a structure.
This patch fixes issue #1912
I welcome every suggestion and, as you can see, the giving name is still not my cup of tea, so if somebody knows a better name for the variables or the new structure, I would be more than happy to rename them.