Leonardo de Moura
9e0b28d8ce
feat(library/init/data/array/basic): improve
2019-02-16 16:08:10 -08:00
Leonardo de Moura
3c73c43ab2
feat(runtime,library/init/data/array/basic): add builtin support for arrays
2019-02-16 15:27:23 -08:00
Leonardo de Moura
c855094ecb
feat(library/init/data/char/basic): mark char.of_nat as noinline
2019-02-16 10:45:52 -08:00
Leonardo de Moura
61274c7d35
feat(library/init/data/char): use uint32 instead of nat for defining char
2019-02-15 18:07:55 -08:00
Leonardo de Moura
542bd432f8
test(tests/compiler): uint constant folding tests
2019-02-15 14:53:29 -08:00
Leonardo de Moura
456ed23cc2
feat(library/init): use extern when declarating nat primitives
2019-02-12 18:12:29 -08:00
Leonardo de Moura
52db59eb87
fix(library/init/data/int/basic): nasty bug at int.repr
2019-02-12 15:58:59 -08:00
Leonardo de Moura
425a4b70d1
feat(library/init/data/int/basic): use extern attribute, and fix div/mod mess
...
Now, int.div and int.mod behave like C++ `/` and `%` for int,
moreover, they satisfy
(a/b)*b + (a%b) = a
2019-02-12 11:41:46 -08:00
Leonardo de Moura
6be47dfb97
feat(library/init/data/string/basic): use extern attribute
2019-02-11 17:54:24 -08:00
Leonardo de Moura
9c3675f58f
feat(library/init/data): use extern when declaring uint and usize primitives
2019-02-11 15:15:57 -08:00
Leonardo de Moura
befa53ec70
feat(library/compiler/emit_cpp): use new extern_attribute module
2019-02-11 11:03:56 -08:00
Leonardo de Moura
348ccf533c
feat(library/compiler): borrowed annotations
2019-02-11 10:08:47 -08:00
Leonardo de Moura
efe64cb2d3
feat(library/init/data/list/basic): tail recursive length
2019-02-08 16:51:44 -08:00
Leonardo de Moura
4734e55f6d
feat(library/init/data): tail recursive append
2019-02-08 16:35:20 -08:00
Leonardo de Moura
afc2569f7c
fix(library/init/data/nat/basic): tail recursive nat.repeat
2019-02-08 11:00:48 -08:00
Leonardo de Moura
ff54bf337a
fix(library/init/data/repr): ultra inefficient to_digits
...
It was very inefficient, and was producing stack overflows for big numbers.
2019-02-05 13:36:13 -08:00
Leonardo de Moura
0918a599ae
feat(*): builtin support for uint functions
...
@kha The VM versions just throw exceptions. They are just stubs to
make sure we can compile Lean.
I implemented the uint functions in the new runtime, but there are a
few missing cases marked with TODO.
I needed these builtins to be able to compile the C++ generated code for
corelib.
2019-02-01 17:04:24 -08:00
Leonardo de Moura
54f501594d
chore(library/init/data/uint,library/init/data/usize): simplify
2019-02-01 15:41:55 -08:00
Sebastian Ullrich
f7a395d035
feat(library/init/data/rbmap/basic): implement rbmap.lower_bound
2018-12-20 15:32:46 +01:00
Sebastian Ullrich
db6b1d6e85
feat(frontends/lean/vm_elaborator,library/init/lean/elaborator): pass parser_state between languages, create parser object on C++ side to existing functions (that don't actually parse anything)
2018-12-18 15:30:38 +01:00
Sebastian Ullrich
cf1aa13a2c
feat(library/init/lean/elaborator): to_pexpr: translate match nodes to pre_equations pexpr
...
Building a proper `equations` pexpr in Lean is not feasible right now because we
need global+local name resolution, [pattern] attribute retrieval, ... to
reimplement `patexpr_to_pattern`. The `pre_equations` will instead be converted
by the old elaborator using this function (TBD).
With this change, `to_pexpr` can now translate every basic definition body in
`core.lean`.
2018-12-05 12:57:41 +01:00
Leonardo de Moura
ed4eeddf0a
feat(runtime/object): add more string primitives
2018-11-14 16:51:10 -08:00
Leonardo de Moura
2fa938220b
chore(library/init/data/string): cleanup
2018-11-14 14:09:45 -08:00
Leonardo de Moura
835b3a10cc
chore(library/init): consistent names
2018-11-14 13:08:57 -08:00
Leonardo de Moura
f404c5c446
refactor(library/init/data): make sure uint* and usize are irreducible
...
Remark: this commit breaks the support for `uint*` and `usize` in the old VM.
2018-11-06 16:55:02 -08:00
Leonardo de Moura
34f01ff32e
chore(library/init/data/nat/basic): missing [specialize]
2018-10-31 14:44:23 -07:00
Leonardo de Moura
cc3767e6a5
refactor(library/init/data): avoid indirection at rbmap
...
Now, the nodes in a `rbmap` contain the key and value, and we avoid
one level of indirection. `rbmap`s are more common than `rbtree`.
We implement `rbtree A` as `rbmap A unit`.
2018-10-26 17:14:09 -07:00
Leonardo de Moura
3ee863da68
feat(library/compiler/erase_irrelevant): eliminate cases_on for builtin types
2018-10-23 14:58:38 -07:00
Leonardo de Moura
8d47d2a026
chore(library/vm,library/init/meta): remove vm_format, and some obsolete meta objects
2018-10-23 11:32:56 -07:00
Leonardo de Moura
89afabae29
refactor(library/init/lean/parser/parsec): make sure custom error message doesn't need to be inhabited
2018-10-21 10:57:23 -07:00
Leonardo de Moura
12d8b0e7ef
feat(library/init/data): annotate rbtree and rbmap functions
2018-10-19 14:57:31 -07:00
Leonardo de Moura
38bc3beffb
fix(library/init): alternative instances
...
Both `alternative` and `monad` implement `applicative`. However,
their default implementations for `seq_right` and `seq_left` are
different. The `alternative` implementation uses the inefficient default
version for `seq_right` available at `applicative`:
```
(seq_right := λ α β a b, const α id <$> a <*> b)
```
instead of the more efficient
```
(seq_right := λ α β x y, x >>= λ _, y)
```
defined at `monad` using the `bind` operator.
This commit makes sure the `applicative` instances for `reader_t`,
`state_t`, `option` and `parsec_t` use the efficient version.
I found the problem when inspecting the generated code for:
```
def symbol (s : string) : parsec' unit :=
(str s *> whitespace) <?> ("'" ++ s ++ "'")
```
2018-10-17 14:25:50 -07:00
Leonardo de Moura
17cb9b4532
chore(library/init/data/repr): remove unnecessary definition
2018-10-17 12:32:38 -07:00
Leonardo de Moura
fa9d6b4ddf
chore(library/init): missing [inline] and remove unnecessary defs
2018-10-17 08:38:30 -07:00
Leonardo de Moura
5683143e8d
chore(library/init/data/to_string): remove unnecessary definition
2018-10-16 16:18:07 -07:00
Leonardo de Moura
effccf9a6d
chore(library/init): mark a few combinators with [specialize]
2018-10-15 13:47:26 -07:00
Sebastian Ullrich
50e6b42f8c
fix(frontends/lean/elaborator): ensure_no_unassigned_metavars: only check mvars in parameter
...
Had forgotten to re-check the standard lib...
2018-10-07 21:11:02 -07:00
Leonardo de Moura
4bcb7051a8
chore(library/init/data/nat/basic): missing @[inline]
2018-10-03 17:24:53 -07:00
Leonardo de Moura
dc937281b3
chore(library/init/string): remove string.iterator_imp
2018-10-02 13:46:01 -07:00
Leonardo de Moura
5ff63c3c96
feat(library/init/data/uint): add uint8
2018-10-01 14:17:11 -07:00
Leonardo de Moura
3a4acbee9e
chore(library/init/data/string): remove string_imp
...
We will use the (to be implemented) `opaque` keyword
2018-10-01 14:17:11 -07:00
Sebastian Ullrich
5123148aa6
feat(library/init/data/string/basic): trim whitespace around symbols
2018-09-28 13:08:24 -07:00
Leonardo de Moura
6a7fbb24f5
feat(library/init/data/dlist): inlining dlist functions is counterproductive
2018-09-24 18:47:10 -07:00
Leonardo de Moura
a31f12d8cd
chore(library/init/core): revert ite+thunks modification
...
We don't need it since we marked `ite` as `[macro_inline]`
2018-09-23 19:27:06 -07:00
Leonardo de Moura
4fd2e71bc9
chore(library/init/data/ordering/basic): mark cmp_using as [inline]
2018-09-17 14:56:31 -07:00
Leonardo de Moura
b07c718425
refactor(library/init/core): change ite signature
2018-09-17 14:27:28 -07:00
Sebastian Ullrich
fea637288d
fix(library/init/data/nat/basic,library/vm/vm_nat): regression in old compiler: primitive for nat equality was ignored
...
Improves parser performance by 26%
2018-09-14 16:33:04 -07:00
Leonardo de Moura
afd54039ab
chore(library): remove >>, we should use *>
2018-09-12 17:31:21 -07:00
Leonardo de Moura
9e305a5f26
chore(library): remove return, we should use pure
2018-09-12 16:51:11 -07:00
Leonardo de Moura
4e6748b55b
feat(library/init/data/rbmap/basic): add rbmap_core low level functions
2018-09-09 18:18:40 -07:00
