Docs: Added section about strict-mode to how-to-use #1265
Conversation
There was a problem hiding this comment.
Looks good!
Side note: I am currently looking through the hundreds of lark grammars available on github and found out that sometimes the current example generation algorithm can sometimes take multiple GBs of memory and take a few dozen seconds to actually find something, even if the intersection test was very fast. I gotta change that...
|
Cool, well done on taking the initiative and testing them. I think it will come down to either optimizing the algorithm, or setting a maximum number of steps, after which it exits without an example. |
|
I will actually do both I think. Stopping after a certain number of steps is easy and reliable no matter how the exact regex(es) look like. Optimizations can only provide so much I think, it will always be possible to create a very slow example I think. |
No description provided.