Commit Graph

1318 Commits

Author SHA1 Message Date
Brandon Elam Barker
9aefc9f60c Updating info about backends in FAQ 2020-09-23 18:32:46 +01:00
G. Allais
d105dd11a7
[ breaking ] remove List1 related ambiguities (#690) 2020-09-22 15:07:40 +01:00
Jan de Muijnck-Hughes
8adfefa4e5
Added some logging information to the totality checker. (#679) 2020-09-22 12:44:13 +01:00
Ohad Kammar
e77b9b4631
[minor] Make base/Data.Nat.divNatNZ compile-time reducible (#689)
Co-authored-by: Ohad Kammar <ohad.kammar@ed.ac.uk>
2020-09-22 09:53:05 +01:00
MarcelineVQ
c4abdb4480 add Semigroup for Ordering and a -> b
useful items for applying multiple predicates, e.g.
sortBy (comparing length <+> compare)
To sort some lists elements by length and then lexographically
2020-09-21 08:07:21 +01:00
Matus Tejiscak
cf45f43f4e Move the Buffer struct back into the .c file. 2020-09-20 17:49:25 +02:00
Christian Rasmussen
9add729ca3 Add Data.String.Iterator implementation for JavaScript 2020-09-20 17:44:19 +02:00
Matus Tejiscak
0a203f8f52 Add a warning comment. 2020-09-20 17:42:50 +02:00
Matus Tejiscak
b58cc433c8 Index StringIterator with the String it refers to. 2020-09-20 10:36:04 +02:00
Matus Tejiscak
d26a9c55bf Tune Data.String.Iterator. 2020-09-20 10:02:18 +02:00
Matus Tejiscak
f73fa55075 Unpack strings into a lazy list. 2020-09-19 22:48:45 +02:00
Matus Tejiscak
e36c211cc0 Add a missing covering annotation. 2020-09-19 21:57:54 +02:00
Matus Tejiscak
74f592053e Make StringIterator abstract. 2020-09-19 21:54:34 +02:00
Matus Tejiscak
e73c6ae3c6 Add comments to fastUnpack and fastConcat. 2020-09-19 19:19:09 +02:00
G. Allais
8d09ec9c93
[ new ] comma-separated interface parameters (#682) 2020-09-19 15:29:23 +01:00
G. Allais
3008c48f77
Adding a whole bunch of tuple sections (#680) 2020-09-19 14:51:57 +01:00
Matus Tejiscak
7b2d1190a1 Make the comments more explicit. 2020-09-19 15:22:29 +02:00
Matus Tejiscak
bdcb2551b2 Change the way of creating directories. 2020-09-19 15:09:26 +02:00
Matus Tejiscak
f3984fff46 Remove more redundancy. 2020-09-19 14:57:45 +02:00
Matus Tejiscak
63c3ebf124 Remove ambiguities. 2020-09-19 14:46:12 +02:00
Matus Tejiscak
5f9c94a4e1 Fix access modifiers of fastUnpack and fastConcat. 2020-09-19 14:43:08 +02:00
Matus Tejiscak
76c3c3f8e8 Update the Scheme support code. 2020-09-19 14:36:57 +02:00
Matus Tejiscak
38a3145e8e Make the C code more explicit. 2020-09-19 14:32:45 +02:00
Matus Tejiscak
2c7c603f02 Update the idris2 source. 2020-09-19 14:26:34 +02:00
Matus Tejiscak
5360adcc23 String-related stdlib tweaks. 2020-09-19 14:22:54 +02:00
Matus Tejiscak
cf6530b387 Install headers for libidris2_support, use the right mode. 2020-09-19 14:11:12 +02:00
Guillaume ALLAIS
09f418f250 [ admin ] removing useless import 2020-09-16 15:45:16 +01:00
Guillaume ALLAIS
2bb95e59ef [ re #660 ] eta-contract parameter candidates 2020-09-16 15:45:16 +01:00
Guillaume ALLAIS
88c4b4d02f [ new ] mapTermM and mapTerm 2020-09-16 15:45:16 +01:00
Guillaume ALLAIS
1152aa3cdd [ re #660 ] inline lets when detecting parameters 2020-09-16 15:45:16 +01:00
Guillaume ALLAIS
644495f097 [ debug ] logging for datatype parameters 2020-09-16 15:45:16 +01:00
Guillaume ALLAIS
364ff73a1f [ refactor ] split up Core.Context
`Core.Context` is 2k+ LoC and contains a *lot* of different thins.
Rather than moving the `log` functions above of `addData` to be able
to add logging there, fork them off to independent files to make the
whole thing more readable.
2020-09-16 15:45:16 +01:00
Giuseppe Lomurno
98e54cea2f Missing runElab decl check 2020-09-16 12:06:38 +01:00
Chetan Taralekar
3c24bc5ed5
Fix ide mode repl not converting escaped characters (#665) 2020-09-16 10:55:44 +01:00
MarcelineVQ
ea0df039fe change runReader's to take state first to allow easier use
following up on the change made in 5c76053cf3
to encourage people to do it in this manner going forward
2020-09-15 22:46:07 +01:00
MarcelineVQ
19bad79847 change runState's to take state first to allow easier use 2020-09-15 09:23:41 +01:00
MarcelineVQ
5c76053cf3 change StateT, swap result parameters
Nipping this historical artifact in the bud before it roots. It's often
useful to be able to `map` directly to the result of a StateT computation
and due to how Functor works this is made harder when the tuple is
(a,state) vs (state,a)
2020-09-15 09:23:41 +01:00
Guillaume ALLAIS
65e194e9bb [ re #660 ] Positivity checking for empty types 2020-09-14 18:37:47 +01:00
Ohad Kammar
2ae330785b
[contrib] Add misc libraries to contrib (#667)
* [contrib] Add misc libraries to contrib

Expose some `private` function in libs/base that I needed, and seem like
their visibility was forgotten

I'd appreciate a code review, especially to tell me I'm
re-implementing something that's already elsewhere in the library

Mostly extending existing functionality:
* `Data/Void.idr`: add some utility functions for manipulating absurdity.
* `Decidable/Decidable/Extra.idr`: add support for double negation elimination in decidable relations
* `Data/Fun/Extra.idr`:
  + add `application` (total and partil) for n-ary functions
  + add (slightly) dependent versions of these operations
* `Decidable/Order/Strict.idr`: a strict preorder is what you get when
  you remove the diagonal from a pre-order. For example, `<` is the
  associated preorder for `<=` over `Nat`.
  Analogous to `Decidable.Order`. The proof search mechanism struggled
  a bit, so I had to hack it --- sorry.

Eventually we should move `Data.Fun.Extra.Pointwise` to `Data.Vect.Quantifiers` in base
but we don't have any interesting uses for it at the moment so it's not
urgent.

Co-authored by @gallais
2020-09-14 16:22:46 +01:00
MarcelineVQ
5673d188f3 add nicer errors for bad specifiers 2020-09-13 10:10:53 +01:00
MarcelineVQ
5acb306bf9 add ability to target scheme flavor in foreign specifiers
There's some missing flexibility in how foreign specifiers can be used with
scheme that is addressed here with minimal changes to how scheme specifiers
are read. This opens up uses for users that they otherwise would have had
to modify the compiler's support files to accomplish.
2020-09-13 10:10:53 +01:00
Matus Tejiscak
9652179990 Add two cases to the record projection test. 2020-09-11 00:29:14 +02:00
Matus Tejiscak
ba89bbcc0d Fix tests some more. 2020-09-10 20:41:58 +02:00
Matus Tejiscak
d98686d4f8 Update doc, fix tests. 2020-09-10 20:33:08 +02:00
Matus Tejiscak
e491e2969e Re-introduce %prefix_record_projections. 2020-09-10 20:18:51 +02:00
Matus Tejiscak
aebe3c19d9 Revert postfix dotted application. 2020-09-10 19:00:48 +02:00
Ohad Kammar
cdd0d07877 [minor] Remove whitespace 2020-09-10 08:08:59 +01:00
Ohad Kammar
2ece0e438f Refactor dynamic rebinding of default totality requirement
Dynamic rebind now reinstates default if an error is raised

(following @gallais 's code review)
2020-09-10 08:08:59 +01:00
Ohad Kammar
b2a9ba651c Bugfix 654: forgot to update GenerateDef
When synthesizing clauses, allow synthesised clauses to be
partial.

It's an arbitrary choice, the other choice can be the default
option. But if the place in which we're synthesising has a non-default
choice, then we're synthesising under the wrong option too.
2020-09-10 08:08:59 +01:00
Ohad Kammar
2f5ae26f36 Fix typo 2020-09-10 08:08:59 +01:00