Commit Graph

296 Commits

Author SHA1 Message Date
Alex Knauth
9f3cf01d26 don't throw internal errors for unreachable cast exprs (#386)
This fills the corresponding entries in the cast table with a Dead-Code
type so that when the contract-generation pass calls the contract-def
thunk, it finds that in the table.
2016-07-07 16:51:08 -04:00
Alex Knauth
e3f56c8a25 check for free-vars in types of casted-exprs (#384) 2016-07-07 14:45:07 -04:00
Andrew Kent
39d6a6047a better rec type intersection 2016-07-07 12:53:31 -04:00
Alex Knauth
a1f8908a29 call compute-constraints instead of sc->constraints in get-max-contract-kind (#382)
* call compute-constraints instead of sc->constraints in get-max-contract-kind

* test cast on an intersection type involving Rec

* remove memory limit on sandboxed-unsafe-ops test
2016-07-07 12:16:15 -04:00
Alex Knauth
583a752954 test (and/c vector? any-wrap/c) -like contracts 2016-06-29 17:04:23 -04:00
Sam Tobin-Hochstadt
a10fd9ef45 Fix accidentally-committed debugging code. 2016-06-24 09:15:28 -04:00
Sam Tobin-Hochstadt
8bcfc71fa8 Avoid overlap with sequences of the wrong arity.
Fixes #371.
2016-06-23 12:00:53 -04:00
Asumu Takikawa
ee02c26020 Fix bug in init-envs (need to generate Rows too)
Also simplified Class case a bit
2016-06-22 17:47:46 -04:00
Asumu Takikawa
ca26d22d1c Fix optimizer fix
Fixed spurious syntax quoting and a test
2016-06-22 00:24:31 -04:00
Sam Tobin-Hochstadt
34ff91b3ca Avoid misoptimizing vector-set! of immutable vectors. 2016-06-21 18:05:38 -04:00
Asumu Takikawa
527b233e45 Remove pconvert dependency in tests
The test just tests whether struct properties
work and it's not important what the precise
property is.
2016-06-21 16:43:38 -04:00
Asumu Takikawa
5f311d00c7 Add more init-env tests 2016-06-21 16:37:17 -04:00
Asumu Takikawa
773dab2c24 Reimplement environment initialization
Avoids using mzlib/pconvert in favor of a few big match
clauses. This lets us cut out a package dependency and
makes the code easier to understand.

This commit also removes the use of mzlib/pconvert
in the debug printer in favor of just using the type
serialization performed in init-envs.rkt.

In addition, a few optimizations for type serialization
were implemented that cut a few percent off of zo sizes.

Note that this commit regresses for zo sizes for modules
that heavily use GUI classes, but that is fixed in a
future commit.
2016-06-21 16:37:17 -04:00
Brian Lachance
ec0c8516c2 Add types for combinations, in-combinations 2016-06-15 09:58:36 -04:00
Asumu Takikawa
bc6e9e80cc Don't use number literal types as contracts
Using = for the comparison doesn't work for TR

Fixes bug in 295a4b7e39
2016-06-13 13:25:51 -04:00
Asumu Takikawa
295a4b7e39 Simplify flat contracts for Value types
Potentially speeds up contracts checks for
types like False or Boolean.
2016-06-13 04:08:33 -04:00
Vincent St-Amour
1ba8e5ba33 Fix test that relied on broken mcons printing. 2016-06-11 16:34:23 -05:00
Sam Tobin-Hochstadt
e3863149f5 Increase timeout again. 2016-06-11 15:20:02 -04:00
AlexKnauth
a846514f28 make cast sound 2016-06-06 13:19:33 -04:00
Asumu Takikawa
a984281cdc Add first-order checks to simple-result-> contract
Fixes issue #368
2016-06-03 13:49:26 -04:00
Asumu Takikawa
7aea90242a Adjust contract tests to allow first-order checks 2016-06-03 13:46:00 -04:00
Asumu Takikawa
b338fc6b64 Be less conservative about struct overlap
Fixes issue #366
2016-06-01 14:23:13 -04:00
Andrew Kent
d66816cf76 use match*/no-order to reduce manual code duplication 2016-05-23 18:13:28 -04:00
Asumu Takikawa
71f17f5cb2 Compute struct intersection more conservatively
Fixes GH issue #205
2016-05-20 16:56:49 -04:00
Asumu Takikawa
43dc59bea2 Restrict struct predicate when parent is mutable
Correctly restrict the struct predicate's filter type when
a parent struct is mutable but the child is not and they both
have polymorphic type variables.

See the discussion in GH issue #205
2016-05-20 16:56:49 -04:00
Asumu Takikawa
e800787773 Don't assume built-in structs are mutable
Doing so adds spurious entries in the type environment for
the setters for these structs, even though the setters are
not defined.
2016-05-20 16:55:23 -04:00
Andrew Kent
c7a3fb0cf1 rename restrict to intersect
since 'restrict' will now create intersections when there is
a complex relationship between the two types, calling it
'intersect' makes a lot more sense.
2016-05-20 15:21:41 -04:00
Andrew Kent
b4a4c174e4 initial intersection types addition
Adds intersection types as a better way to handle the the case
when restrict cannot structurally intersect two types (e.g. when
you learn within a polymorphic function a variable x of type A
is also an Integer, but we dont know how A relates to Integer).
This allows for non-lossy refinements of type info while typechecking.
2016-05-20 11:34:04 -04:00
Asumu Takikawa
19e8efec0f Avoid internal error on mismatched values
Fixes issue #342
2016-05-09 18:03:17 -04:00
Asumu Takikawa
31bf61e333 Remove redundant values checks
These can all be done via check-below later in the
typechecking process
2016-05-09 18:03:17 -04:00
Asumu Takikawa
af385d6932 Fix issue #336 2016-04-28 11:29:12 -04:00
Andrew Kent
f9c5a534d0 filter -> prop
This pull request is largely a renaming effort to clean up the TR codebase. There are two primary things I wanted to change:

1. Replace all occurrences of "filter" with "prop" or "proposition"
   - The word "filter" is a meaningless opaque term at this point in the Typed Racket implementation. If anything, it just adds confusion to why things are the way the are. We should use "proposition" instead, since that's what they actually are.

2. Avoid using "Top" and "Bottom" in both the type and proposition realm.
   - Currently the top type is called Univ and the bottom type is called bottom, while the top proposition is called Top and the bottom proposition is called Bot. This is just unnecessarily confusing, doesn't really line up w/ the user-space names, and doesn't line up with the names we use in TR formalisms. Worse, all of the top types of primitive types---e.g. the type of all structs, StructTop--- use Top, so it is really easy to get confused about what name to use for these sorts of things.

With these issues in mind, I made the following changes to names:

Top -> TrueProp
Bot -> FalseProp
TypeFilter -> TypeProp
NotTypeFilter -> NotTypeProp
AndFilter -> AndProp
OrFilter -> OrProp
-filter t o -> -is-type o t
-not-filter t o -> -not-type o t
FilterSet -> PropSet
NoFilter -> #f
NoObject -> #f
-FS -> -PS
-top -> -tt
-bot -> -ff
implied-atomic? q p -> implies-atomic? p q
filter-rec-id -> prop-rec-id
-no-filter -> -no-propset
-top-filter -> -tt-propset
-bot-filter -> -ff-propset
-true-filter -> -true-propset
-false-filter -> -false-propset
PredicateFilter: -> PredicateProp:
add-unconditional-filter-all-args add-unconditional-prop-all-args
2016-04-25 18:36:12 -04:00
Alex Knauth
bacc1b3411 Fix type of syntax->list (#348)
* Fix type of syntax->list

to return `(U False (Listof (Syntaxof Any)))` if it can't prove that the input is a syntax-list.

Fixes https://github.com/racket/typed-racket/issues/347
This makes the type `syntax->list` consistent with the type `stx->list` already has.

* Add tests for syntax->list
2016-04-16 20:18:59 -04:00
ben
f820fac6a0 regression tests
For github issues 111 113 114 115
2016-04-11 17:27:19 -04:00
Vincent St-Amour
b352739131 Always populate type table, regardless of optimizer status.
Closes #343.
2016-04-11 14:43:44 -05:00
Asumu Takikawa
cea5091ee6 Fix struct name in Struct type representation
Closes issue #304
2016-04-07 21:49:50 -04:00
Asumu Takikawa
acef58a5d0 Add an extra test for new random case 2016-04-07 17:00:13 -04:00
Asumu Takikawa
1a11ac53e2 Add typed/racket/random, bump version 2016-04-07 15:03:57 -04:00
Georges Dupéron
e33c902842 Remove debugging output from test case in typed-racket-test/succeed/define-typed-untyped-identifier-syntax-properties.rkt 2016-04-07 16:47:17 +02:00
Asumu Takikawa
2fc669e136 Adjust error message for test
Needed to change due to recent syntax-parse changes
2016-04-07 00:58:09 -04:00
Vincent St-Amour
6dc5b1a994 Fix optimizer test for changes to syntax-parse. 2016-04-06 12:06:42 -05:00
Georges Dupéron
4ab256abf2 Fixes GH issue #315 Syntax properties are not forwarded by define-typed/untyped-identifier, following @samth's suggestions. 2016-04-06 12:06:42 -05:00
Asumu Takikawa
9ec358b665 Update type-table after tc-funapp processing
This makes the tooltip show the more refined function
type after application typechecking. For example, case->
types will be narrowed to the particular case that fits.

Polymorphic function types will be instantiated at the
appropriate type.

Fixes issue #325
2016-03-28 17:53:34 -04:00
Sam Tobin-Hochstadt
7572adb9c2 Fix use of simple-result-> to check the domains.
This broke pict3d, but also disabled some contract checks that
should have been kept.
2016-03-28 17:11:37 -04:00
Asumu Takikawa
812f1a8c79 Add types for box-cas!, unsafe-box*-cas! 2016-03-23 16:02:22 -04:00
Daniel Feltey
425ff47700 Raise errors when signatures are not in the signature environment while typechecking
Closes #319
2016-03-12 16:39:39 -06:00
Asumu Takikawa
d23e05f2c3 Escape "~" in tc-error/fields arguments
Fixes issue #314
2016-03-02 04:43:34 -05:00
Georges Dupéron
350a8bb74e Changed how arguments and other pieces of information are printed, to follow the error message conventions, as suggested by samth in PR #250 .
* PR #250: https://github.com/racket/typed-racket/pull/250
* Error message conventions: http://docs.racket-lang.org/reference/exns.html?q=raise-arg#%28part._err-msg-conventions%29
2016-03-01 15:19:09 -06:00
Asumu Takikawa
72927e2248 Add the rest of Racket's exn hierarchy 2016-02-21 12:20:40 -05:00
Asumu Takikawa
a90f6c46eb Add exn:break:hang-up and exn:break:terminate 2016-02-21 02:48:45 -05:00