e5972640d2
Remove 'eval' command.
...
Was introduced as a work-around to get some debugging info out of scripts, but tests do now provide the same capability with a better output and, do so automatically.
2022-12-14 22:14:35 +01:00
978a6c6981
Collect and display evaluation hints on test failures.
2022-12-14 22:00:56 +01:00
7b22b63ad8
move 'EvalInfo' to project::script & define a new 'EvalHint'
2022-12-14 22:00:09 +01:00
921e7abbb6
Move pretty-printing utilities to project::pretty
2022-12-14 21:59:09 +01:00
87546e0abd
Return non-zero exit code on test failure
...
And integrated test results with miette report.
2022-12-14 18:44:31 +01:00
3a9cc668fc
Use bright_black (i.e. grey) for box drawing.
2022-12-14 17:45:16 +01:00
2d6fc8bd54
Group test results by module.
2022-12-14 17:34:24 +01:00
3c664b9651
test 1-9 passing
2022-12-14 09:45:26 +01:00
b71315ba2f
chore: trying to fix test 5
2022-12-14 09:45:26 +01:00
6635a918b5
clean up minor warnings
2022-12-14 09:45:26 +01:00
b6962ba9d3
Add 'eval' command to evaluate target aiken function
...
Pretty useful for debbugging. Though, on second-thoughts, this is
something we may want to review later and maybe have that done by
default for tests.
At the moment, we expects tests to unify to `bool`, and treat `false`
values as failing tests. Yet, on failures, this gives little
information about what's wrong with the test.
It'd be nice to either have better way to assert in tests, or, to
simply accept non-bool tests, and show whatever the test evaluates
to as a debug output.
2022-12-14 09:45:26 +01:00
11c793dd2a
feat(Type): add is_option method
2022-12-14 09:45:25 +01:00
8393d8555c
fill constants to data so now test 006 passes
2022-12-14 09:45:25 +01:00
dc5ae296db
implement anonymous functions, more tests pass now
2022-12-14 09:45:25 +01:00
d78e2c9c6f
feat: finish up generic match cases
2022-12-14 09:45:23 +01:00
e43063d447
overhaul monomorphize to cover all generic cases
...
test b passes
2022-12-14 09:45:23 +01:00
15dc202810
Feat: generic function call tests work
2022-12-14 09:45:23 +01:00
e4d9ca4586
support generics
2022-12-14 09:45:23 +01:00
16fbf5bbcd
feat: fix recursive functions
2022-12-14 09:45:22 +01:00
3d3b3d7e10
checkpoint
2022-12-14 09:45:22 +01:00
95986fed83
Fix lexer for signed integers.
2022-12-13 19:52:30 -05:00
18bf89418a
Fix codegen for binary operator 'or'
...
a && b → if a { b } else { false }
a || b → if a { true } else { b }
2022-12-13 18:52:23 -05:00
431b0cfcf2
Preserve newlines after blocks of comments.
...
This is an example of output from the formatter now:
```
//// Some module documentation
// foo
const a: Int = 42
// Some comment block
// For which newlines are respected.
// Foo
// Another one
/// add_one documentation
pub fn add_one(n: Int) -> Int {
// n + 1
n + 1
}
```
before this commit, comments would all be collapsed into one group
above the function as:
```
// Some comment block
// For which newlines are respected.
// Foo
// Another one
/// add_one documentation
pub fn add_one(n: Int) -> Int {
```
2022-12-13 18:52:23 -05:00
22a526bb13
Enforce unique top-level names for tests too.
...
This prevents the compiler from crashing later on. Test names should be unique and not clash with function names.
2022-12-13 18:52:23 -05:00
4dfb454d8a
refactor: change match to if matches!
2022-12-13 10:34:34 -05:00
1637a0d30e
Add --match-tests to 'check' cmd
...
For running only tests matching a certain pattern. Useful when doing TDD.
2022-12-13 10:34:34 -05:00
69db9696d6
feat: prefix tuples with hash again cause ambguity with call
2022-12-13 10:27:17 -05:00
6dc4738b66
feat: update formatter for new changes
2022-12-12 19:22:11 -05:00
a4f6388eca
feat: remove the need for # in front of tuples
2022-12-12 19:22:11 -05:00
dfc57b347a
feat: switch generic type args to be delimited by chevrons
2022-12-12 19:22:11 -05:00
a18aad7daf
refactor(project): use new Options instead of a spiderweb of bools
2022-12-10 19:14:58 -05:00
0eb3cf221b
Trigger and report on more events in the compilation pipeline.
2022-12-09 15:04:02 +01:00
749d8ecb10
Fix formatter to not prefix tests as 'pub'
2022-12-09 14:35:50 +01:00
d09d38d65b
Add a flag '--skip-tests' to the 'check' cmd.
...
So that tests can be skipped, and the old behavior recovered if necessary.
Tests execution is on by default however.
2022-12-09 14:14:15 +01:00
4ad74bef1f
refactor: move prints to cli via EventListener trait
2022-12-08 19:25:28 -05:00
801ab3989e
feat: display test runner summary alongside test results.
...
Moar sexy.
2022-12-08 19:25:28 -05:00
4cae4a4467
pad left instead of right, so units/thousands are aligned.
2022-12-08 19:25:28 -05:00
2ba712eef6
feat: start adding padding to test output
2022-12-08 19:25:28 -05:00
5770ea2456
Make test runner more sexy.
...
Using colors.
2022-12-08 19:25:28 -05:00
db25ff3817
refactor: run_tests to avoid repetition.
2022-12-08 19:25:28 -05:00
e9d8e1d317
feat: print budget consumed by test
2022-12-08 19:25:24 -05:00
141a9aef30
feat: impl Sub for ExBudget
2022-12-08 19:24:20 -05:00
f250d3df84
feat: eval should take initial budget
2022-12-08 19:24:20 -05:00
4cdb5d8d02
Implement test runner.
...
easy.
2022-12-08 19:24:20 -05:00
384c4daa4a
feat: add test_gen function
2022-12-08 19:24:20 -05:00
bc785673b2
Fix compilation errors for the newly introduce test & add type inference.
...
Tests are basically functions for which the return type should unify with bool. In principle, the type checker could also check that a test function has no arguments but, a test function with arguments wouldn't parse in the first place; feels a bit hacky but it works when considering the pipeline as a whole.
Note that the code generation is still to be done.
2022-12-08 19:24:20 -05:00
ea48747825
Extend parser for 'test' keyword.
2022-12-08 19:24:20 -05:00
a65b4aa471
feat: add test def and test token
2022-12-08 19:24:20 -05:00
80a9b7b36a
feat: add map support and tuple deconstruction in let
2022-12-08 14:28:30 -05:00
26d2a95618
missing a couple forces
2022-12-08 14:28:30 -05:00
KtorZ