Small optimizations #3
Conversation
11% improvement on iter_attributes bench 19% improvement on try_get_attribute bench
|
Current implementation of attributes parsing have bugs on some corner cases, so I've rewrite it from scratch, so all this changes now irrelevant. So I'll close this PR, but thanks for the work! |
|
@Mingun How about I add the benchmarks to this PR, then merge the optimizations anyway, and rebase changes on top? So that you can properly compare the performance of the two implementations. |
|
There is no point in measuring the performance of incorrect code, because it will still be rewritten (already done: #4). But it makes sense to measure the performance of the new code and try to make the same optimizations. So I think it is better to open a new PR with benchmark changes, then merge it, rebase #4 over it and merge it (with optimizations if that will be valuable) |
Split out from the criteron benchmarks because it will need more manual rework to backport
commentary on tafia/quick-xml#348