Remove type based duplicates #745
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
collapse_bind returned nothing if last non-deterministic branch returned empty result. This commit fixes the behavior by adding a dummy non-empty branch which returns result not included into collapse_bind results.
Interpreter and type-checking code treat calls of the functions without type declared differently. Type checker represents this call as a tuple type. But interpreter calls it thus effectively treating it as %Undefined% type. This commit adds %Undefined% to the list of types for tuples which are not function calls removing the difference.
Before this logic got the types of the function and tried to apply each to the parameter's types. Thus each time effectively evaluating the function result with the given function type. It leads to the duplication of the function results when the same function has few different type signatures (non-deterministic type). This commit changes behavior. New behavior is filtering list of the function type by checking if type is applicable to the parameter's types first. Then get any type from the list and call the function using this type.
|
This also blocks #668 from merging. |
Necr0x0Der
reviewed
Jul 23, 2024
| !(assertEqualToResult | ||
| (f (+ 5 "S")) | ||
| ((Error "S" BadType))) | ||
| ; TODO: This test has different behavior in old_interpreter and minimal interpreter. |
There was a problem hiding this comment.
Yes, I like the new behavior better
Necr0x0Der
reviewed
Jul 23, 2024
| !(assertEqualToResult | ||
| (get-type (fmap (curry-a + 2) (UntypedC (Null) 5))) | ||
| ()) | ||
| ; Two examples below are type-checked successfully because, (UntypedC "5") |
There was a problem hiding this comment.
Yes, I agree. The type of (UntypedC "5") should be reducible to %Undefined% meaning that the whole expression can type-check. I'd add "FIXME" or "TODO", because we may need to revisit these tests and uncomment their updated versions
vsbogd
commented
Jul 23, 2024
vsbogd
commented
Jul 23, 2024
Necr0x0Der
approved these changes
Jul 23, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #235. Type checking behavior is changed (see tests
c1_,d2_andd5_. In particular for the case like:previous behavior was to return the most inner type error, i.e.
(Error "S" BadType), now interpreter returns less inner type error, i.e.(Error (+ 5 "S") BadType). I am not sure whether one should be preferred over another.