On Wed, Dec 18, 2013 at 4:35 PM, Kyungmin Park <[email protected]> wrote: > Now we’re trying to build buxton at Tizen v2.2 for testing purpose. But it’s > hard to use it. Fix one build issues, it makes another problem and so on.
We're likely not ready for integration with Tizen 2.2 yet as we're developing for Tizen 3.0, so if you do come up with any issues, please let us know (of course, github pull requests and even bug reports will be highly appreciated!) > How do you test it at Tizen platform? How can it test it? Requested tests > are simple. > > 1. Get/set > > 2. Thread safe read/write bt-daemon is single-threaded, so all DB operations are done without threading. > 3. Performance comparison with many properties, more than 100,000? There are 2 test options built into the source code repository: 1) make check This runs an internal unit testing suite that currently tests about 80+% of the code base. It runs as non-root, and bypasses some of the Smack code as it is meant to not run as root. We use `make check` to verify that all commits are functional and don't break the project before accepting a patch. If you enable coverage during configure/building, you will be able to create a coverage report after running make check as well with 'make check'. 2) bxt_timing this test tool runs against the real, installed bt-daemon. run `make install` as root, use systemd to start buxton.socket, and then run `bxt_timing` as non-root (or, any user really). This test tool performs many requests, and we use it to estimate round-trip latency. The times reported by the tool are per request. Several interface methods are tested (set/get/unset, for int/string/large strings) > P.S., Here’s buxton.spec. Thanks! We'll certianly use this. Auke _______________________________________________ Dev mailing list [email protected] https://lists.tizen.org/listinfo/dev
