Closed jonashaag closed 1 year ago
Apparently some bottleneck is there. Will be cool to find it and fix.
I did a profiling via:
CPU_PROFILE=cpu.prof ./fx random.json
go tool pprof -http 8080 ./fx cpu.prof
Looks like we can optimize collectSiblings() call to be a lazy loading. It will help with the first load.
Ditto here, fx also lags in navigating json documents, navigating a large json document with jless is immediate and responsive.
Yes, looks like jless got ahead of fx here) I have a few ideas on how to improve fx in the future.
I took a look at profile and spotted a few slow places. I will rewrite them to be faster.
Wow. It took me the whole week (7*12h) to refactor fx from the ground up. Now fx can open JSON files of any size. I mean HUGE files and will be flying across it. I tested on a 3GB JSON file. I'm so tired. 😮💨
It took me forever. But I spent the whole week rewriting fx from scratch! Now fx is a rocket!!
GitHub sponsors are appreciated :)
I'm trying to view a ~ 1 GB JSON file (not JSON-per-line file).
With
jless
it takes around 2 seconds to open the viewer. Withfx
it takes longer than 60 seconds (actually I stoppedfx
after a while).