Boris Zhguchev
Boris Zhguchev
closes #59
It has some fmt and clippy failures. You can fix it locally just kicking off the appropriate commands and correct the code
There are merge conflicts since the aforementioned in the PR structure obtained a new field (cfg)! I can take a look at the weekend and either resolve the conflicts or...
> You dont need a cache because you just use the existing regex, and this can cache the regex at the caller. I don't grasp fully what you mean tbh,...
The introduced cache handles a specific situation. The compilation of regex is a voracious operation and takes quite some time. Thus, if a user has either multithreading or a stream...
But that is precisely the case. We parse a string that represents regex. The compilation of this string into the particular regular expression is what the cache is summoned to...
It is not bad or good. It does not solve the initial problem. The problem is the following: Given - we have 1K requests for jspath like `$.[?(@.field == '[a-zA-Z]')]`....
I think in general you are correct. > add a cache that espeically only caches regexes and stores them. Yeah, that is how it is implemented now. > just compile...
Very good point. I don't know why but the idea of caching the whole jspath slipped through my fingers. I will think about it in the evening a bit, searching...
I have taken a look to the benches and truly it seems the benefit of having a regex cache is leveled by the alternative of caching jspathinst itself. Thanks for...