What did I do in this period?
Here’s a quick overview of the progress in this week,
- t-oidmap: merged into ’next'
- t-hashmap: sent v2 to the list -> merged into ‘seen’
- t-oid-array: sent v1 to the list
- t-oidset: ready in my tree
Other than these there is also t-urlmatch-normalization which I had sent 10 days ago, it is waiting for a review.
Apart from sending these patches, I also participated in a patch series discussion of ‘rs/unit-tests-test-run’, which introduces a new macro called TEST_RUN() for running unit tests. TEST_RUN() negates the need of defining functions to run unit tests, which was the case with TEST(). This macro would be helpful in ’t-oid-array’ as it uses internal functions test__run_*(). Using TEST_RUN() would help us in avoiding the use of these internal functions.
However, in the mailing list, there was also scepticism around using callbacks in setup() functions in the tests which use TEST(). Personally, I had been using that pattern to initialize a data-structure and release them after the callback, which reduces boilerplate. Examples can be found in t-{oidmap, hashmap}. However, the scepticism was that use of this pattern reduces readability and it makes manual verifying harder. However, there are still not concrete guidelines as to the most idiomatic way to write unit tests.
So what is the plan ahead?
The plan is to get feedback on ’t-oidset’ from the mentors and post it to the list in the coming week. And to also iterate on and respond to the reviews on the patches already on the list, but not yet in ’next’. Apart from that, hopefully work on ’lib-repo’.
Thanks and see ya’ next week!