Fix #12241: Make space computation lazy #12271
Conversation
This corner case optimization is unsafe and no code depends on it.
|
test performance with #exhaustivity please |
|
performance test scheduled: 1 job(s) in queue, 0 running. |
|
Performance test finished successfully: Visit https://dotty-bench2.epfl.ch/12271/ to see the changes. Benchmarks is based on merging with master (4e8ca78) |
|
Performance test finished successfully: Visit https://dotty-bench.epfl.ch/12271/ to see the changes. Benchmarks is based on merging with master (4e8ca78) |
| def dedup(spaces: Seq[Space])(using Context): Seq[Space] = | ||
| if (spaces.lengthCompare(1) <= 0 || spaces.lengthCompare(10) >= 0) spaces | ||
| else { | ||
| val res = spaces.map(sp => (sp, spaces.filter(_ ne sp))).find { |
There was a problem hiding this comment.
I wonder if it would help performance if we do not compute spaces.filter(_ ne sp) eagerly.
There was a problem hiding this comment.
Here spaces has a size < 10, so should not matter much.
|
The newly added test does get json data (from Chrom inspector): [
[12271,"2021-04-29T12:36:56+0200","16a622fd1b",903.4235,846.695],
[12271,"2021-04-29T13:16:47+0200","16a622fd1b",871.691,831.192],
[12271,"2021-04-29T13:51:41+0200","16a622fd1b",955.0745,835.253],
[12271,"2021-04-29T14:26:22+0200","16a622fd1b",950.573,842.552]
]The UI does not show them due to too few points (need to improve the JavaScript). |
Fix #12241: Make space computation lazy