Performance ot fromTable and toTable is very bad
acmeguy opened this issue · 3 comments
Hello,
The performance of this library leaves a lot to be desired.
I have been optimizing some parts of if (fromTable/toTable) and I'd like to contribute that work.
I've written a toTable function that handles big payloads. The current one does not complete those and takes forever processing fairly-big payloads. (the test that ran here for 5881ms was reduced to 1ms by proper handling)
What is the best way to contribute here?
Regards,
- Stefán
fromTable belongs to the jsonstat-suite, a library of independent solutions for specific problems. fromTable (like fromCSV) is more like a proof of concept than anything else as it will only work with the right ordered input. Performance improvements to fromTable are very welcome but not critical (of course, anyone is free to develop her own fromTable function). jsonstat-suite join can also be very improved in terms of performance. Any contribution to join is very welcome.
toTable is a root method belongs to the jsonstat-toolkit and many methods rely on it. It's critical to improve it but it must be done safely: the current toTable interface must be kept exactly as it is.
I've written a toTable function that handles big payloads.
Does it keep the current interface, with all its features and setup options? If so, a pull request would be the right method but, because toTable is critical, I'd need some time to run extensive tests to make sure that this is so.
If your optimizations only affect parts of toTable, please share them with me here to study them and decide what would be the simplest way to reflect your contribution.
Is there any news about this contribution?