199 commits

Author	SHA1	Message	Date
ayazhafiz	d352d2cdf8	Revert "Include annotation type signatures in Expected struct" ... This reverts commit 6e4fd5f06a1ae6138659b0073b4e2b375a499588. This idea didn't work out because cloning the type and storing it on a variable still resulted in the solver trying to uify the variable with the type. When there were errors, which there certainly would be if we tried to unify the variable with a structure that had nested flex/rigid vars, the nested flex/rigid vars would inherit those errors, and the program wouldn't typecheck. Since the motivation here was to expose the signature type to `reporting` so that we could modify it with suggestions, we should instead pass that information along in something analogous to the `Expected` struct.	2021-11-25 13:24:42 -05:00
ayazhafiz	a8e38172ac	Remove redundant refs	2021-11-25 11:22:19 -05:00
ayazhafiz	ee34e79790	Include annotation type signatures in Expected struct ... To provide better error messages and suggestions related to changing type annotations, we now pass annotation type signatures all the way down through the constraint solver. At constraint generation we associate the type signature with a unique variable, and during error reporting, we pull out an `ErrorType` corresponding to the original type signature, by looking up the unique variable. This gives us two nice things: 1. It means we don't have to pass the original, AST-like type annotation, which can be quite large, to everyone who looks at an expectation. 2. It gives us a translation from a `Type` to an `ErrorType` for free using the existing translation procedure in `roc_types::subs`, without having to create a new translation function.	2021-11-25 11:16:17 -05:00
Richard Feldman	618925fe27	Fix typo in comment	2021-11-24 23:17:24 -05:00
Folkert	093b733a2c	add tag name dedup	2021-11-25 00:11:11 +01:00
Folkert	7c2e8cd812	optimize NAT	2021-11-24 23:00:03 +01:00
Folkert	9c551a0276	use insertion sort	2021-11-24 22:55:45 +01:00
Folkert	e0d7c54906	simplify alias insertion	2021-11-24 22:39:42 +01:00
Folkert	4e4f827bb0	simplify function-or-tag-union insertion	2021-11-24 22:11:05 +01:00
Folkert	379abeed26	refactor tag insertion	2021-11-24 21:57:19 +01:00
Folkert	340055fca5	point to empty VariableSubsSlice	2021-11-24 21:57:00 +01:00
Folkert	cb8e9acb92	special-case the closure tags	2021-11-24 21:14:58 +01:00
Folkert	3e640f78bf	cleanup	2021-11-24 20:44:51 +01:00
Folkert	308b91153e	add faster insertion path for tag unions	2021-11-24 20:38:52 +01:00
Folkert	541dd9777a	thread_local experiment	2021-11-24 19:59:53 +01:00
Folkert	f1541d56bd	use Entry in some places	2021-11-24 19:45:30 +01:00
Folkert	5e76e3a97e	more reserving into subs	2021-11-24 19:14:37 +01:00
Folkert	739ae60b1e	clean up usage of reserve_into_subs	2021-11-24 18:52:22 +01:00
Folkert	61f6687dea	fix using the wrong length	2021-11-24 18:50:47 +01:00
Folkert	1fbd2a7e7b	reserve variable slices	2021-11-24 16:02:55 +01:00
Folkert	a0c9270124	keep track of visited variables in deep_copy_var	2021-11-24 15:14:17 +01:00
Folkert de Vries	6944c900c2	Merge pull request #2065 from rtfeldman/alias-remove-lowercases ... Alias remove lowercases	2021-11-23 10:21:05 +01:00
hafiz	682043c16c	Merge pull request #2063 from rtfeldman/type-to-var-prealloc ... preallocate variables to prevent extra vector allocations	2021-11-22 20:05:16 -06:00
Folkert	b2aa33f839	Merge remote-tracking branch 'origin/trunk' into alias-remove-lowercases	2021-11-22 21:19:27 +01:00
Folkert	2c1005fdf0	remove lowercases from Content	2021-11-22 21:15:58 +01:00
Folkert	2506875970	preallocate variables to prevent extra vector allocations	2021-11-22 20:17:03 +01:00
Folkert	8afdf282cb	assert that no tag union or record with no fields/tags makes it into Subs	2021-11-21 22:25:03 +01:00
Folkert	594e754e1e	manually track visited varibles in deep_copy_var_to	2021-11-20 01:39:57 +01:00
Folkert	2bccc2d28d	optimize instantiate_rigids_help	2021-11-20 01:26:17 +01:00
Folkert	497bc2db02	make instantiate_rigids_help use a loop/stack	2021-11-20 00:24:46 +01:00
Folkert	eb9bb12725	bumpalo type_to_var	2021-11-19 19:48:51 +01:00
Folkert	2810c1fc52	clean up instantiate_rigids	2021-11-19 17:51:05 +01:00
Folkert	206c8889df	Use StorageSubs for pending specializations	2021-11-19 13:22:17 +01:00
Folkert	65e0866d0a	add assert; a returned copy must always be in the target subs	2021-11-17 16:32:20 +01:00
Folkert	849510c42c	WIP	2021-11-16 23:55:11 +01:00
Folkert	7c4300a7a1	clarify what pending means	2021-11-15 22:43:32 +01:00
Folkert	db081cd84b	WIP	2021-11-15 20:12:48 +01:00
Folkert	6d833845f9	fix typo	2021-08-14 00:53:18 +02:00
Folkert	3e448c1b74	get rid of inefficient conversion from mutmap to uniontags	2021-08-14 00:32:01 +02:00
Folkert	4cbb1c9b25	optimize initial tag union creation	2021-08-13 23:53:27 +02:00
Folkert	df83bf0d48	Merge remote-tracking branch 'origin/soa-alias' into alias-nominal-equality	2021-08-13 12:11:36 +02:00
Folkert	c9040c012e	make things work	2021-08-13 11:04:31 +02:00
Folkert	42f04fcc77	use variablesubsslice for Apply	2021-08-12 22:49:56 +02:00
Folkert	3bbf1a5506	use UnionTags for recursive tag unions	2021-08-12 20:50:33 +02:00
Folkert	6c0860b6bf	milestone all but one gen tests passing	2021-08-11 21:50:11 +02:00
Folkert	0e66995437	Merge remote-tracking branch 'origin/trunk' into tags-soa-phase-0	2021-08-11 17:25:00 +02:00
Folkert	83ad331bb1	remove chase_record_ext_var	2021-08-07 15:35:57 +02:00
Richard Feldman	49aa0b8ffd	Eliminate a Vec in record solving	2021-08-07 08:35:43 -04:00
Folkert	7a44572ebc	make occurs return Result to use the ? operator	2021-08-07 13:56:16 +02:00
Folkert	ad3d006a11	use a SubsIndex instead of Box	2021-08-07 13:50:53 +02:00