Trevor Elliott
0b19e4ab4d
Add a NameDisp to renamer errors
2015-09-28 18:05:09 -07:00
Trevor Elliott
1a6825a5c0
Comput the output of :browse based on what's in scope
2015-09-28 17:32:15 -07:00
Trevor Elliott
45234b15cf
Don't give warnings when imported names are shadowed
2015-09-28 17:31:48 -07:00
Trevor Elliott
2a202fd60a
Makefile wibbles
2015-09-28 13:58:39 -07:00
Trevor Elliott
91eae415c2
Print name origins in a shadowing warning
2015-09-28 13:57:45 -07:00
Trevor Elliott
58da031a90
Sort properties by name
2015-09-28 13:57:34 -07:00
Trevor Elliott
60694f8456
Output format changed
2015-09-28 12:07:42 -07:00
Trevor Elliott
496b87b9d3
Simplify the IfaceDecls type
...
As IfaceDecls is no longer used when interpreting imports, there's no way for
conflicts to appear. As a result, the values of each map no longer need to be
lists, and the mappend operation is greatly simplified.
2015-09-27 19:58:54 -05:00
Trevor Elliott
b4fbec108e
Update some test output
...
The core AST now always prints fully-qualified names.
2015-09-27 19:56:58 -05:00
Trevor Elliott
3a39b271da
Smarter merging of naming environments
...
When merging two entries that are the same, don't duplicate the binding on the
RHS of the map.
2015-09-27 19:42:15 -05:00
Trevor Elliott
0541e80a98
Add a way to print names with their location info
2015-09-27 19:40:36 -05:00
Trevor Elliott
5cb9280530
Fix a display bug in the :help command
...
Don't try to figure out when parens are needed on operators
2015-09-27 19:39:33 -05:00
Trevor Elliott
8bcefbe4dd
Don't produce warnings when shadowing in the dynamic env
2015-09-27 19:39:06 -05:00
Trevor Elliott
a7b7e25c0f
Remove some dead code
2015-09-27 19:38:41 -05:00
Trevor Elliott
694a1ed92d
Fix some bugs with the dynamic env
...
Shadowing wasn't working correctly, as names were being added to the dynamic
naming environment in multiple places.
2015-09-27 18:55:48 -05:00
Trevor Elliott
bc5759f867
Try to have make test
rebuild cryptol less frequently
2015-09-27 18:02:26 -05:00
Trevor Elliott
06b5156b17
Compare two names based on how they would display
2015-09-27 18:02:02 -05:00
Trevor Elliott
454270165f
Case analysis wibbles
2015-09-27 18:01:24 -05:00
Trevor Elliott
8f7e76b714
mkTInfix was rewriting too much
...
`max 1 (a + b) - 1` was rewriting to `max 1 (a + b - 1)`, as mkTInfix was
considering the use of max to be an infix operator. The fix was to make it only
perform this way when the thing in the position of `max` was one of the
un-renamed infix constraints (==,>=,<=).
2015-09-27 18:01:09 -05:00
Trevor Elliott
89aaa72e0b
Start fixing failing tests
2015-09-27 14:35:24 -05:00
Trevor Elliott
c855463fd9
Warning removal
2015-09-26 13:15:10 -05:00
Trevor Elliott
4839de6048
Makefile fix
...
Something wasn't happening correctly with the PKG target, so I altered it to
run everything in one shell.
2015-09-26 13:15:00 -05:00
Trevor Elliott
9e36f05f4b
Merge remote-tracking branch 'origin/master' into wip/name-change
2015-09-26 01:26:23 -05:00
Trevor Elliott
23f520c810
NameMap wasn't actually helping
2015-09-26 01:25:40 -05:00
Trevor Elliott
d9a3bb1de7
Prelude loads again
2015-09-25 16:04:07 -05:00
Trevor Elliott
a133ea0f67
Fixed some renamer problems
2015-09-25 15:02:43 -05:00
Trevor Elliott
c191aebb52
Fix some renamer/parser bugs
2015-09-24 17:32:25 -05:00
Trevor Elliott
55ae1f852d
Major changes plumbed through, now for the bugs!
2015-09-24 16:47:52 -05:00
Trevor Elliott
56171c60ca
Library is building again
2015-09-24 16:30:23 -05:00
Trevor Elliott
e5ab23cfe8
Lib nearly builds
2015-09-23 15:45:42 -07:00
Trevor Elliott
89324e2e95
Fold changes through the parser, and part of the typechecker
2015-09-21 20:08:06 -07:00
Trevor Elliott
2a048be438
More incremental fixes
2015-09-18 17:13:18 -07:00
Trevor Elliott
12cbdabfeb
Continue plumbing new names through
2015-09-18 16:56:23 -07:00
Adam C. Foltzer
c07f97a994
update to Stackage LTS 3.5
...
This snapshot requires GHC 7.10+, but since we're only copying over the
Stackage config on the Jenkins machines, we're not requiring 7.10-only
just yet. The newest 7.8.4 snapshot has a too-old version of SBV,
though, so builds for earlier than 7.8 are more of a wild west.
2015-09-18 16:05:31 -07:00
Trevor Elliott
15bd25c69e
Lots of changes to the renamer
...
The structure is mostly the same, but now it changes the type of the name
parameter to the parsed AST.
2015-09-18 15:45:43 -07:00
Trevor Elliott
88e5965e7f
Parameterize parsed AST on names
2015-09-17 17:30:39 -07:00
Adam C. Foltzer
4df92455e5
update stackage snapshot
2015-09-17 14:45:51 -07:00
Jenkins Builder
75539f9244
add \protect
to fix error on Windows docs
2015-09-16 13:12:37 -07:00
Adam C. Foltzer
2ca0e6f732
switch back to non-latexmk for book (Windows)
...
Windows didn't like the latexmk solution from
5eb5f00d0a
.
2015-09-16 11:37:00 -07:00
Thomas M. DuBuisson
17abf37b7d
A rough cut at SIV.
2015-09-16 09:19:27 -07:00
Adam C. Foltzer
5eb5f00d0a
change how the book is built
2015-09-15 13:38:06 -07:00
Trevor Elliott
98c49b875c
Slight speedup in name comparison
2015-09-14 21:50:26 -07:00
Adam C. Foltzer
28461e7b72
update attribution on SHA512
2015-09-14 17:18:47 -07:00
Thomas M. DuBuisson
7643a17182
Fix #272 Don't full blocks in ChaChaPoly8675309
2015-08-28 15:22:03 -07:00
Brian Huffman
72d8794b9d
Change where integer masking is applied in the interpreter
...
In the Value datatype, we now maintain an invariant for
all packed bitvectors: The value must be within the range
0..2^width-1. Values are normalized when constructed, not
when they are scrutinized.
2015-08-28 13:47:17 -07:00
Aaron Tomb
44f8bcd310
Fix comment in Keccak example
...
Note: this is actually Cryptol 1 code. We should port it. I've started
the process, but don't have a complete port yet.
2015-08-28 09:49:29 -07:00
Adam C. Foltzer
9b6cde1f8d
reexport version number
2015-08-26 11:34:16 -07:00
Trevor Elliott
ec4fb4ab54
Fix #271
...
The renamer wasn't expecting to see anything but (==), (>=), or (<=) show up in
a case where that wasn't a valid assumption.
2015-08-25 10:50:35 -07:00
Brian Huffman
792f8e433f
Allow "-" in infix operator names
2015-08-21 11:00:46 -07:00
Adam C. Foltzer
50ca428018
add regression in #269 to benchmark suite
2015-08-20 15:49:06 -07:00