Parametrized tests #3669
Parametrized tests #3669
Conversation
franciszekjob
changed the base branch from
master
to
1431-refactor-with-parsed-values-function
…tps://github.com/foundry-rs/starknet-foundry into 1431-parametrized-tests
franciszekjob
changed the base branch from
1431-refactor-with-parsed-values-function
to
1431-refactor-function-names-generated-by-plugin
| ); | ||
| } | ||
|
|
||
| // TODO: Use separate process (with oracle) instead of heavy fib calculation. |
There was a problem hiding this comment.
note: Currently this tests uses same approach as the one in fuzzer - fuzzing_exit_first. However, this is technically non-deterministic and ultimately we should use an external process (oracles) to make one of the tests fail before another.
If you agree with above, I will create and link an issue.
franciszekjob
requested review from
ddoktorski and
MKowalski8
and removed request for
a team
There was a problem hiding this comment.
I think we should also have an e2e contract test that uses name:.
| ); | ||
| } | ||
|
|
||
| // TODO: Use separate process (with oracle) instead of heavy fib calculation. |
| pub fn new(map: &HashMap<SmolStr, Vec<Expr>>) -> Self { | ||
| Self(map.clone()) | ||
| } |
There was a problem hiding this comment.
IMO it's bit weird to hide a clone like this, I'd prefer an explicit clone at caller side.
| use std::sync::LazyLock; | ||
|
|
||
| static RE_SANITIZE: LazyLock<Regex> = | ||
| LazyLock::new(|| Regex::new(r"[^a-zA-Z0-9]+").expect("Failed to create regex")); |
There was a problem hiding this comment.
Can we even have this kind of invalid characters in the test name itself?
| fn sanitize_expr(expr: &Expr, db: &SimpleParserDatabase) -> String { | ||
| let expr_text = &expr.as_syntax_node().get_text(db); | ||
| let expr_sanitized = RE_SANITIZE | ||
| .replace_all(expr_text, "_") |
There was a problem hiding this comment.
How does the replacement work exactly? If I have multiple consecutive invalid characters, will they be replaced with multiple _ or a single one?
|
|
||
| Ok(quote!( | ||
| #[implicit_precedence(core::pedersen::Pedersen, core::RangeCheck, core::integer::Bitwise, core::ec::EcOp, core::poseidon::Poseidon, core::SegmentArena, core::circuit::RangeCheck96, core::circuit::AddMod, core::circuit::MulMod, core::gas::GasBuiltin, System)] | ||
| #[snforge_internal_test_executable] |
There was a problem hiding this comment.
What if #[test] managed to execute already before? Won't that duplicate the executables?
| let param_count = func | ||
| .declaration(db) | ||
| .signature(db) | ||
| .parameters(db) | ||
| .elements(db) | ||
| .len(); |
There was a problem hiding this comment.
This query could be reused, not sure how expensive it is though.
| Ok(()) | ||
| } | ||
|
|
||
| fn get_filtered_func_attributes( |
There was a problem hiding this comment.
If it's filtering out just test attr in case there is no fuzzer then maybe it can be named something more explicit.
There was a problem hiding this comment.
Can we also add some multiple attributes tests cases with at least fuzzer, test and test_case present?
And please explicitly test the expansion of this order:
#[test]
#[fuzzer]
#[test_case(...)]
There was a problem hiding this comment.
Please also add/expand e2e/plugin_diagnostics.rs tests to check for diagnostics with #[test_case] attribute.
Towards #1431
Stack:
with_parsed_values#3670Introduced changes
Introduce parametrized tests by addin
#[test_case]attribute.Checklist
CHANGELOG.md