Switch to toml-edit for spanned deserialization #2074
Conversation
|
Bencher Report| Branch | 2074/merge |
| Testbed | ubuntu-latest |
⚠️ WARNING: The following Measure does not have a Threshold. Without a Threshold, no Alerts will ever be generated!Click here to create a new Threshold
For more information, see the Threshold documentation.
To only post results if a Threshold exists, set the--ci-only-thresholdsCLI flag.
Click to view all benchmark results
| Benchmark | Latency | nanoseconds (ns) |
|---|---|---|
| fibonacci 10 | 📈 view plot | 492,970.00 |
| foldl arrays 50 | 📈 view plot | 1,718,900.00 |
| foldl arrays 500 | 📈 view plot | 6,657,600.00 |
| foldr strings 50 | 📈 view plot | 7,194,900.00 |
| foldr strings 500 | 📈 view plot | 62,789,000.00 |
| generate normal 250 | 📈 view plot | 45,623,000.00 |
| generate normal 50 | 📈 view plot | 2,095,500.00 |
| generate normal unchecked 1000 | 📈 view plot | 3,479,200.00 |
| generate normal unchecked 200 | 📈 view plot | 767,850.00 |
| pidigits 100 | 📈 view plot | 3,230,800.00 |
| pipe normal 20 | 📈 view plot | 1,500,900.00 |
| pipe normal 200 | 📈 view plot | 10,229,000.00 |
| product 30 | 📈 view plot | 832,330.00 |
| scalar 10 | 📈 view plot | 1,514,000.00 |
| sum 30 | 📈 view plot | 830,770.00 |
jneem
changed the title
There was a problem hiding this comment.
Purely stylistic nitpick: when I start to have 3 or more (xxx_to_yyy), I now don't hesitate to declare a local, private trait (it's a bit more LoC but it feels less icky than having to call different combinations of methods based on the type). It's just a suggestion, it's also fine as it is.
Otherwise, I think a bit more doc would be nice. Not necessarily on conversion methods that are self-explanatory, but at least at the module level: if I'm a new contributor, I wouldn't necessarily understand immediately why we would do toml deserialization differently and using this lower level API while we have the serde machinery.
| @@ -27,7 +27,6 @@ doc = ["dep:comrak"] | |||
| format = ["dep:topiary-core", "dep:topiary-queries", "dep:tree-sitter-nickel"] | |||
| metrics = ["dep:metrics"] | |||
| nix-experimental = [ "dep:cxx", "dep:cxx-build", "dep:pkg-config" ] | |||
| spanned-deser = ["dep:serde-untagged"] | |||
There was a problem hiding this comment.
Ah, nice, we can even get rid of the feature.
Yes, that turned out nicely (and also removed the asymmetry between _to_term and _to_rich_term) |
This little experiment shows that we can use toml-edit for spanned parsing instead of the toml crate. It doesn't actually add any dependencies, since
tomlusestoml_editfor parsing anyway.