Update colored dep#142
Conversation
Merging this PR will improve performance by 5.57%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ⚡ | WallTime | recursive_memoized[BTreeMap<u64, u64>, 0] |
131 ns | 14 ns | ×9.4 |
| ⚡ | WallTime | hamiltonian_cycle[6] |
1,499 ns | 917 ns | +63.47% |
| ⚡ | WallTime | generate_combinations[5] |
1.5 µs | 1.1 µs | +33.71% |
| ⚡ | WallTime | bench_array2[10] |
11 ns | 9 ns | +22.22% |
| ⚡ | WallTime | find_highest_set_bit[255] |
6 ns | 5 ns | +20% |
| ⚡ | WallTime | bench_array1[42] |
55 ns | 47 ns | +17.02% |
| ⚡ | WallTime | sum_fold |
355 ns | 308 ns | +15.26% |
| ⚡ | Simulation | recursive_memoized[BTreeMap<u64, u64>, 0] |
488.6 ns | 430.3 ns | +13.56% |
| ⚡ | WallTime | bench_array2[42] |
52 ns | 47 ns | +10.64% |
| ⚡ | WallTime | hamiltonian_cycle[5] |
944 ns | 893 ns | +5.71% |
| 👁 | WallTime | system_time |
39 ns | 53 ns | -26.42% |
| 👁 | WallTime | find_highest_set_bit[42] |
5 ns | 7 ns | -28.57% |
| 👁 | WallTime | find_highest_set_bit[0] |
5 ns | 6 ns | -16.67% |
| 👁 | WallTime | permutations[6] |
68.4 µs | 70.8 µs | -3.31% |
| 👁 | WallTime | iter_batched_per_iteration |
47 ns | 56 ns | -16.07% |
| 👁 | WallTime | iter_batched_ref_large_input |
4 ns | 5 ns | -20% |
| 👁 | WallTime | iter_batched_large_input |
7 ns | 8 ns | -12.5% |
| 👁 | WallTime | from_elem_decimal[1024] |
81 ns | 85 ns | -4.71% |
| 👁 | WallTime | fib_10 |
386 ns | 455 ns | -15.16% |
| 👁 | WallTime | init_array[42] |
48 ns | 65 ns | -26.15% |
| ... | ... | ... | ... | ... | ... |
ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.
Tip
Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.
Comparing rukai:update_colored (b30ce7d) with main (73170f9)
|
@rukai tests are failing but no action on your part is required, we have a msrv bump to decide on, and the other one is CI flakiness on this repo that we need to fix. |
|
Hello! As merging this would require us to raise the rust msrv of the core and the criterion compatibilty crate, we've decided not to merge this for now if the benefit is only getting rid of lazy static. Next codspeed-rust major is probably when we'll bump msrv to 1.80 since criterion did so upstream and divan already has an 1.80 msrv. We'll keep this open until then unless bumping is required to fix a security issue or if there's a more compelling argument than just getting rid of lazy static. |
|
Oh, oops I didn't realize colored had bumped it's msrv, totally understandable. Feel free to keep the pr open until you bump your msrv, or feel free to close it for now and just update your deps together once you do bump msrv. |
|
Edit: I’ve reached out to |
|
Similarly, this is adding a separate It seems reasonable to bump the MSRV to 1.80? |
58111d5 to
b1525f8
Compare
Greptile SummaryThis PR upgrades
Confidence Score: 5/5Safe to merge — the change is a straightforward version bump with no API breakage; the public Colorize trait is unchanged in colored 3.x and all call sites compile cleanly. The colored 3.x upgrade removes lazy_static without any breaking API changes to the Colorize trait used in the codebase. The MSRV bump to 1.80 is consistent across all crates and matches colored 3's requirement. The build.rs refactor is correct: inspect_err (stable since 1.76) fits within the new 1.80 floor. Only minor cosmetic issues were found. crates/cargo-codspeed/src/build.rs has a remaining typo and a stale MSRV comment that were missed during the partial cleanup. Important Files Changed
|
b1525f8 to
b30ce7d
Compare
Mostly I just want to remove lazy_static from my dep tree, and colored 3.0.0 has removed its dependency on lazy_static.