Commit Graph

322 Commits

Author SHA1 Message Date
Sören Tempel
c725b11c89 Flush standard out after writing prompt to it
On Unix-like operating systems stdio.h is usually line-buffered. As
putStr uses fputs(3) from stdio.h internally, output will be written to
standard out after a newline character is written to the buffer. Since
the prompt does not contain a newline, it will only be written to
standard output after the user presses return. I encountered this issue
on Alpine Linux which uses musl libc (instead of glibc). However, I
believe this issue is likely also reproducible with glibc. This commit
fixes this issue by flushing standard output after writing the prompt to
it. Surprisingly, `src/Idris/IDEMode/REPL.idr` already does this
correctly, `src/Idris/REPL.idr` does not though.
2021-04-09 15:17:00 +01:00
Ohad Kammar
b65907f770
Support Multi-declarations (#1280) 2021-04-07 12:21:17 +01:00
Edwin Brady
12f40f538f Use correct multiplicity in scope of lets
The scope should be checked at the same multiplicity as the enclosing
expression.
2021-04-04 18:10:34 +01:00
Guillaume ALLAIS
99b87c8156 [ fix #1248 ] Bring qualified names back for data & record types 2021-03-31 23:17:53 +01:00
Atomotron
a1e7221c32
[ fix #1200 ] Switch scheme backend to flonum functions (#1203)
Co-authored-by: Guillaume ALLAIS <guillaume.allais@ens-lyon.org>
2021-03-31 23:17:41 +01:00
G. Allais
238eb62da6
[ fix #1230 ] Better error messages for out-of-scope identifiers (#1233) 2021-03-29 10:45:48 +01:00
Guillaume ALLAIS
2df49ebdd4 [ fix #1224 ] moduleIdent must be capitalised 2021-03-25 15:59:46 +00:00
Guillaume ALLAIS
00929deed6 [ cosmetic ] nicer error messages for uncaught exceptions 2021-03-25 15:59:46 +00:00
Denis Buzdalov
8abd60535b [ fix ] Ability for data types to be operators was restored 2021-03-25 14:16:47 +00:00
G. Allais
97fb5d7b94
[ fix #893 ] proof gadget for with clauses (#1222) 2021-03-25 10:02:06 +00:00
G. Allais
21f2913527
[ fix #710 ] Enforce assumptions about capitalised idents (#1207)
Given we keep getting tripped up by this, here we go:

* Namespaces
* Data names
* Record names
* Data constructor names (except for operators)
* Record constructor names (except for operators)
* Interface constructor names (except for operators)
2021-03-22 13:22:52 +00:00
Guillaume ALLAIS
f600182999 [ close #31 ] test case for the issue 2021-03-18 10:38:48 +00:00
G. Allais
f959987a51
[ fix #834 ] Fix implicitsAs for local definitions (#1199) 2021-03-17 10:54:25 +00:00
André Videla
405c266b5e
[ refactor ] Multiline error report (#1155) 2021-03-16 14:10:45 +00:00
Stefan Höck
591797ebaf
[ new ] Support overloaded floating point literals (#1177) 2021-03-16 14:08:50 +00:00
Stefan Höck
1784593abb
[ new ] Applicative and Monad for Pair (#1188) 2021-03-15 13:42:04 +00:00
Matúš Tejiščák
f4a790ded4
Identify prefix and postfix record projections (#1183)
`.proj` and `proj` are identically defined but separate functions.
This patch fixes it by defining `.proj` only once, and adding `proj = (.proj)`
for every projection.
2021-03-15 13:40:13 +00:00
Raoul Hidalgo Charman
3e85e23638
Namespace checks for resugaring (#1161)
This avoids resugaring to the wrong type when there are user defined
symbols which conflicts with builtins such as Pair.

Changed the test linear002 which was relying on this behaviour for a
user defined Unit.

Fixes #634.
2021-03-15 13:34:54 +00:00
Michael Messer
d08c0c78b3
Change :doc to use PTerm instead of Name (#1178) 2021-03-12 09:46:46 +00:00
Edwin Brady
17cdc7fa88
Merge pull request #1171 from edwinb/fix1163
Correct multiplicities when checking Pi binders
2021-03-09 18:36:08 +00:00
Edwin Brady
04a0f5001f Correct multiplicities when checking Pi binders
We've always just used 0, which isn't correct if the function is going
to be used in a runtime pattern match. Now calculate correctly so that
we're explicit about which type level variables are used at runtime.

This might cause some programs to fail to compile, if they use functions
that calculate Pi types. The solution is to make those functions
explicitly 0 multiplicity. If that doesn't work, you may have been
accidentally trying to use compile-time only data at run time!

Fixes #1163
2021-03-09 17:23:05 +00:00
Ruslan Feizerahmanov
50c60185a7
[ auto ] Ignore hidden names in Core.Context.getSearchData (#1143)
Co-authored-by: Guillaume ALLAIS <guillaume.allais@ens-lyon.org>
2021-03-09 14:13:29 +00:00
Stefan Höck
8d4321eb9a
Add Data.Bits to base (#1033) 2021-03-04 20:59:56 +00:00
Stefan Höck
4c7d0db4bc
[ fix #1148 ] Support hyphenated package names (#1151) 2021-03-04 19:09:15 +00:00
Edwin Brady
c9de4d9f9f Remove more trailing whitespace 2021-03-03 14:36:33 +00:00
Edwin Brady
6a29fab245 Remove trailing whitespace in tests
Keep the linter happy
2021-03-03 13:55:30 +00:00
Edwin Brady
6887a5f95f Record local hints in delayed elaborators
We might not have set up search problems yet when delaying an
elaborator, so we need to know what the local hints were at the point of
delay.
2021-03-03 13:49:32 +00:00
Edwin Brady
a00fc25109
Merge pull request #1146 from edwinb/fix1140
Check current holes only at end of elaboration
2021-03-01 20:51:46 +00:00
Edwin Brady
1d965627c9 Check current holes only at end of elaboration
Fixes #1140 - when checking holes were solved, we checked all of them,
but there may still be some open if there's a local definition.
2021-03-01 19:11:15 +00:00
G. Allais
cee7e38894
[ new ] Proof search from 'Applications of Applicative Proof Search' (#1093) 2021-03-01 08:29:43 +00:00
Edwin Brady
37cc095f7a Merge github.com:idris-lang/Idris2 into versions 2021-02-27 18:00:18 +00:00
Edwin Brady
1d8166b1b2 Version number constraints in 'depends' field
Now reporting an error if we can't find a package that satisfies the
constraints. The version number field can still be a string (as it used
to be) but will give a deprecation warning - and the old style version
string wasn't used anyway.

Version constraints can have an upper and/or lower bound, which can be
inclusive or not.
2021-02-27 17:58:52 +00:00
André Videla
aa27ccbdb6
Merge pull request #1097 from andylokandy/multiline
Implement multi-line string
2021-02-26 21:50:54 +00:00
G. Allais
59de5f1326
[ fix #762 ] Different case tree building strategy (#1125) 2021-02-26 09:33:07 +00:00
Andy Lok
01c5011653 improve error report 2021-02-25 20:14:26 +08:00
Andy Lok
5fed5c2b7a Merge branch 'master' of https://github.com/idris-lang/Idris2 into multiline 2021-02-25 19:52:42 +08:00
Andy Lok
8c23f5944b address comment 2021-02-25 18:43:42 +08:00
G. Allais
fcd834c423
[ fix #110 ] Check the LHS' head is not shadowed (#1121) 2021-02-25 08:51:27 +00:00
Guillaume ALLAIS
da88b80481 [ fix #794 ] missing cases in recoverable 2021-02-24 20:25:04 +00:00
G. Allais
d2eeb7ce86
[ fix #758 ] desugar non-binding sequencing in do blocks to (>>) (#1095) 2021-02-24 11:07:16 +00:00
Denis Buzdalov
95af3cf4be
More compose instances and one usage of them (#1089) 2021-02-23 10:53:43 +00:00
Guillaume ALLAIS
00067e8151 [ fix #637 ] force indentation after a with 2021-02-23 10:52:22 +00:00
G. Allais
c10c1d65a5
[ fix #1022 ] detect more impossible cases (#1108) 2021-02-23 10:51:38 +00:00
Andy Lok
775d7b4bdb Add test for escaping NL 2021-02-23 13:10:36 +08:00
Andy Lok
2358ac0f73 Fix StrError12 test 2021-02-23 02:10:57 +08:00
Guillaume ALLAIS
741960f0d1 [ test ] case for #484
This issue was fixed by #1081
2021-02-22 16:04:04 +00:00
G. Allais
30d402ed7f
[ fix #899 ] Be careful when generating an impossible LHS (#1081) 2021-02-22 09:53:30 +00:00
Andy Lok
83ab4f0f15 Fix lint 2021-02-21 03:13:13 +08:00
Andy Lok
07eb6fda47 Allow hashtag escape for multi-line string 2021-02-21 03:11:01 +08:00
Andy Lok
22a769e6b5 Implement multiline string 2021-02-20 18:05:26 +08:00