Skip to content

Releases: haproxytech/opentracing-c-wrapper

v1.1.3: BUILD: fixed rpath for FreeBSD for linking opentracing-cpp (again)

14 Oct 22:37
Compare
Choose a tag to compare
'-Wl' must be specified for each argument separately.

Also, the package version has been changed to 1.1.3.

v1.1.2: BUILD: corrected system_clock duration for clang

09 Oct 13:34
Compare
Choose a tag to compare
In GCC STL system_clock the duration is in nanoseconds, but in clang it
is in microseconds.

Also, the package version has been changed to 1.1.2 and the library
version to 1.1.1.

v1.1.1: BUILD: test: added libgen.h include directive

08 Oct 23:41
Compare
Choose a tag to compare
On FreeBSD UNIX it is necessary to add the libgen.h include directive
in order to be able to use the basename() function declared there.

Also, the package version has been changed to 1.1.1.

v1.1.0: MINOR: separating the library load from running the tracer thread

09 Jun 23:12
Compare
Choose a tag to compare
Two functions have been added, otc_tracer_load() and otc_tracer_start(),
which are actually the otc_tracer_init() function divided into two parts:
the dynamic loading of a tracer library (a tracer plugin) and the launch
of the tracer thread (i.e. initialization).

This is done so that the dynamic loading of the library can be separated
from the running of the tracer thread.  For example, if we use this in a
daemon program, we can load the library before switching to daemon mode
(before the fork()/exec() functions) and then initialize the tracing
process in the child process.

The otc_tracer_init() function has not been removed from the library,
and can still be used in the previous way.

Also, the library version has been changed to 1.1.0.

v1.0.0: MINOR/BUG: fixed calculation of data size in the otc_text_map structure

04 Jun 17:23
Compare
Choose a tag to compare
When increasing the number of data pairs in the otc_text_map structure,
this value was miscalculated for unallocated data (ie those for which
the number of pairs was 0).

The newly allocated memory was cleared, but this was done at the wrong
offset.

Also, there was an error in the macro OT_TEXT_MAP_SIZE() where sizeof
was taken from the wrong data type.