1
0
mirror of https://github.com/janet-lang/janet synced 2024-12-24 15:30:27 +00:00
Commit Graph

4038 Commits

Author SHA1 Message Date
primo-ppcg
2a7ea27bb7 do not expand false branch more than once
Fixes #1191
2023-06-11 19:15:48 +07:00
primo-ppcg
32c5b816ae use unquotes instead 2023-06-11 18:38:20 +07:00
Chloe Kudryavtsev
e54ea7a1d8 fixup! peg: add support for "true" and "false" primitives to always/never match 2023-06-11 12:38:40 +02:00
primo-ppcg
1077efd03a update if-let
Fixes #1189
2023-06-11 17:19:29 +07:00
Chloe Kudryavtsev
f9ab91511d peg: add support for "true" and "false" primitives to always/never match
The use cases involve user-expandable grammars.
For example, consider the IRC nickname specification.
> They SHOULD NOT contain any dot character ('.', 0x2E).
> Servers MAY have additional implementation-specific nickname restrictions.

To implement this, we can do something along these lines:
```janet
(def nickname @{:main '(some :allowed)
                :allowed (! (+ :forbidden/dot :forbidden/user))
	        # for lax mode, (put nickname :forbidden/dot false)
	        :forbidden/dot "."
		# to add your own requirements
		# (put nickname :forbidden/user 'something)
		:forbidden/user false})
```

Additionally, it's common in parsing theory to allow matches of the
empty string (epsilon). `true` essentially allows for this.

Note that this does not strictly add new functionality, you could
emulate this previously using `0` and `(! 0)` respectively, but this
should be faster and more intuitive.
The speed improvement primarily comes from `(! 0)` which is now a single
step.
2023-06-11 10:44:39 +02:00
primo-ppcg
2c3ca2984e simplify if-let logic 2023-06-11 12:09:58 +07:00
primo-ppcg
94722e566c if-let better test coverage 2023-06-11 12:07:13 +07:00
Christopher Chambers
163f7ee85d Add test for maclintf in nested macro invocations 2023-06-10 16:52:20 -04:00
Christopher Chambers
52d3470cbe Fix order in which *macro-lints* is set during expansion
Previously, `*macro-lints*` was set after the `macroexpand1` fiber was
resumed, rather than just before.  And, `*macro-lints*` was never
cleared.  This behavior was typically fine since the main users of
`compile` pass the same lint array repeatedly, and the first macro
expansion (somewhere in boot.janet) never produces a lint.  But, when
compiling with a fresh lint array, if the first macro invocation
produced a lint, the lint was always lost.
2023-06-09 12:53:10 -04:00
Calvin Rose
0bd6e85c61 update changelog 2023-06-08 19:54:01 -05:00
Calvin Rose
e35c6b876f
Merge pull request #1183 from primo-ppcg/take-drop
Make take and drop more symmetric
2023-06-08 14:52:51 -05:00
Calvin Rose
9a2897e741 Run through astyle with manual corrections 2023-06-08 13:01:49 -05:00
primo-ppcg
70b2e8179d nitpick performance tweak 2023-06-08 23:57:07 +07:00
primo-ppcg
5317edc65d minor readability change
As suggested by @sogaiu

@zevv forget to push this change in a recent PR (https://github.com/janet-lang/janet/pull/1175#issuecomment-1576128152).

Incidentally, the affected lines were already reformatted in the current PR, via fmt/format-file.
2023-06-08 23:00:05 +07:00
Calvin Rose
866d83579e Address #1165 - Allow for partial ffi support without totally removing testing.
Query at runtime which calling conventions are supported, including
a placeholder :none.
2023-06-08 09:07:16 -05:00
primo-ppcg
a238391b36 take-drop performance tweaks
Increase efficiency for `take` and `drop` with slices.
Check indexed types before bytes types.
2023-06-08 14:50:37 +07:00
Calvin Rose
5e152d30db
Merge pull request #1182 from chris-chambers/channel-close-resumptions 2023-06-07 17:15:45 -05:00
Christopher Chambers
57c954783d Fix resumption values when closing a channel.
When suspended in `ev/give` or `ev/take`, closing the channel should
cause the result of `ev/give` or `ev/take` to be `nil`.

When suspended in `ev/select`, closing the channel should cause the
result of `ev/select` to be `[:close ch]`.

The results were flipped before.
2023-06-07 15:01:56 -04:00
primo-ppcg
b5407ac708 take-drop dictionaries
Return table for `take` of dictionary types.
Allow `drop` of dictionary types.
2023-06-07 19:20:05 +07:00
primo-ppcg
472ec730b5 take-drop symmetry
Allow `take` from the end of bytes or indexed (as `drop` does).
Allow `drop` from fibers (as `take` does).
2023-06-07 18:12:36 +07:00
Calvin Rose
8c819b1f91 Update README.md 2023-06-05 12:55:08 -05:00
Calvin Rose
528a516390 Add more sandbox capabilities.
Add more granularity to ffi sandbox capabilities - distinguish between
using FFI functions, creating FFI functions, and creating executable
memory.
2023-06-04 18:48:34 -05:00
Calvin Rose
6509e37c84 Update CHANGELOG.md and README.md 2023-06-04 16:11:17 -05:00
Calvin Rose
649173f661
Merge pull request #1175 from zevv/quicky
Added JANET_NO_AMALG flag to Makefile
2023-06-04 14:24:35 -05:00
Calvin Rose
1efb0adb35 Add 3 argument form to fiber/new
Allow passing in environment table at fiber creation since
it is a fairly common thing to do.
2023-06-04 14:17:18 -05:00
Calvin Rose
88a8e2c1df Define *task-id* since it is part of the event-loop runtime. 2023-06-04 14:05:37 -05:00
Ico Doornekamp
bb4ff05d35 Added NO_AMALG flag to Makefile to build janet from the individual
source files instead of from the amalgamated janet.c; this considerably
speeds up parallel builds on modern CPUs
2023-06-04 20:02:47 +02:00
Calvin Rose
dd3b601c87 Don't do fiber double arity check. 2023-06-04 12:56:03 -05:00
Calvin Rose
e22d101a62
Merge pull request #1179 from zevv/zevv-tests
Added misc tests to increase test coverage
2023-06-04 12:53:43 -05:00
Calvin Rose
4b3c813f5a Revert to old behavior of janet_fiber returning NULL.
When there is a bad arity function passed in to the fiber
constructor, return NULL so the runtime can choose what to do.
This is not the prettiest API but does work, and gives better error
messages for instance in the compiler.
2023-06-04 11:21:52 -05:00
Calvin Rose
67f375bea2 Small code style change to boot.janet 2023-06-04 11:10:39 -05:00
Calvin Rose
88ba99b87e
Merge pull request #1181 from chris-chambers/ev-gather-cancel
Ensure ev/gather fibers are fully canceled on error
2023-06-04 11:10:24 -05:00
Christopher Chambers
53447e9d0b Ensure ev/gather fibers are fully canceled on error. 2023-06-04 10:49:30 -04:00
Calvin Rose
c4c86f8671 Run boot.janet through janet-format. 2023-06-03 16:47:59 -05:00
Calvin Rose
658941d26d Fix macro declaration. 2023-06-03 14:24:41 -05:00
Calvin Rose
e4bf27b01c Macro hack for meson-min build. 2023-06-03 14:22:16 -05:00
Calvin Rose
7d48b75f81 Update README.md 2023-06-03 14:19:02 -05:00
Calvin Rose
5f56bf836c Update meson.build file. 2023-06-03 13:55:49 -05:00
Ico Doornekamp
c0f5f97ddb Added misc tests to increase test coverage 2023-06-03 19:06:51 +02:00
Calvin Rose
15177ac2e9
Merge pull request #1162 from sogaiu/reorg-tests-new
Reorganize tests
2023-06-03 08:01:34 -05:00
Calvin Rose
8360bc93ac
Merge pull request #1177 from zevv/zevv-isatty
added os/isatty, do not enable colors if stdout is not a tty
2023-06-03 07:58:16 -05:00
Ico Doornekamp
e0ea844d50 added os/isatty, do not enable colors if stdout is not a tty 2023-06-02 16:59:54 +02:00
sogaiu
9675411f35 Reorganize tests 2023-06-02 07:04:07 +09:00
Calvin Rose
e97299fc65 Fix #1174 - bad debug info causing stack traversal to segfault.
Coming from commit 77189b6e66, relating
to changes in source mapping debug info, this caused a segfault when
traversing a stack frame where the birth_pc was incredibly large due
to wrap around. This fix prevents the wrap around and does saturating
subtraction to 0.
2023-06-01 13:01:59 -05:00
Calvin Rose
26a113927e Add handling for new bytecode optimizations. 2023-06-01 12:47:59 -05:00
Calvin Rose
d0aa7ef590 Add a few asserts to quiet some of the -fanalyze calls in gcc 13. 2023-06-01 10:52:34 -05:00
Calvin Rose
5de889419f Rename contains? and contains-key? to has-value? and has-key?
Shorten docstrings to be less like a tutorial. They get put into RAM
and memory ain't free.
2023-05-31 22:35:57 -05:00
Calvin Rose
0fcbda2da7
Merge pull request #1017 from Techcable/feature/helper-func-contains
Add `contains?` helper function to boot.janet
2023-05-31 22:28:41 -05:00
Calvin Rose
14e33c295f Make clock tests less fragile for CI. 2023-05-31 22:21:12 -05:00
Calvin Rose
644ac8caf8 Add compiler optimizations for #1163 - eachp
Should result in much better bytecode in the simple case.
2023-05-31 12:26:27 -05:00