Commit Graph

377 Commits

Author SHA1 Message Date
Bennet Bleßmann
dc495f10f8 change behaviour in supposedly unreachable cases 2025-08-01 20:46:31 +02:00
Bennet Bleßmann
ca52c65902 don't remove the child on wait/kill
- important for wait with timout(0) as we may want to try again until the process has realy exited.

- make process_release release the process instead
2025-08-01 20:33:15 +02:00
Bennet Bleßmann
97e2e5d7f0 implement process_release/1, process_wait/2, process_wait/3, and process_kill/1 2025-08-01 20:27:27 +02:00
Bennet Bleßmann
1ee4f7a55f store child process in machine state 2025-08-01 20:21:36 +02:00
Bennet Bleßmann
43dd1587fb handle some error cases and replace unwrap with expect 2025-08-01 20:21:23 +02:00
Bennet Bleßmann
183761eba4 adjust default cwd 2025-08-01 20:20:41 +02:00
Bennet Bleßmann
82e989f631 add comments and try to fix binding the child process pid 2025-08-01 20:20:20 +02:00
Bennet Bleßmann
2b052bf8dd fix more things 2025-08-01 20:20:09 +02:00
Bennet Bleßmann
003e9d4610 get it working 2025-08-01 20:19:54 +02:00
Bennet Bleßmann
d79ece2497 fix indices 2025-08-01 20:18:44 +02:00
Bennet Bleßmann
ae0baf4893 [WIP] add support to spawn new processes 2025-08-01 20:18:13 +02:00
Bennet Bleßmann
4f04d2dfca replace .skip(n).next() with .nth(n) 2025-07-31 21:47:12 +02:00
Bennet Bleßmann
de89a78cbb remove unecessary cast 2025-07-31 21:47:11 +02:00
Bennet Bleßmann
3365cffa96 prefere for loop 2025-07-31 21:47:10 +02:00
Bennet Bleßmann
df3f1236f7 replace always erroring of_else with map_err 2025-07-31 21:47:09 +02:00
Bennet Bleßmann
7593f88d5a ingore nerver looping loop, but add a todo 2025-07-31 21:47:07 +02:00
Bennet Bleßmann
a639fec153 fix match/if-let can be simplified to ? 2025-07-31 21:47:03 +02:00
Bennet Bleßmann
ae3019d923 fix unecessary reference/dereference 2025-07-31 21:46:51 +02:00
Bennet Bleßmann
8cc74b2af7 fix clippy::uninlined_format_args 2025-07-31 21:26:08 +02:00
Mark Thom
9b293cf1b6 detect end_of_file before end_of_stream in get_char (#2990) 2025-07-22 20:39:52 -07:00
Mark Thom
e303e5a805 correct number_chars (#2976) 2025-07-07 22:44:05 -07:00
Mark Thom
762b63e1f4 use granular hierarchical locks in offset_table.rs 2025-07-07 22:44:02 -07:00
Mark Thom
1332611f83 use OffsetTableImpl without synchronization by default 2025-07-07 22:39:47 -07:00
Bennet Bleßmann
725def07cd fix crash when loading html 2025-07-07 22:39:47 -07:00
Bennet Bleßmann
cbdd0fbf15 make Fixnum::build_with harder to accidentally misuse
change trait bound order for better
2025-07-07 22:39:34 -07:00
Mark Thom
d6b6eda77d fix cont function crashes (#2920) 2025-07-07 22:38:12 -07:00
Mark Thom
4e0493474e replace deprecated unify_complete_string call with allocate_cstr 2025-07-07 22:38:12 -07:00
Markus Triska
d51cbd67e0 ENHANCED: partial_string/3 no longer creates atoms
As a consequence, resulting strings are now quickly reclaimed on
backtracking.

This addresses #2912.

Test case:

    :- use_module(library(iso_ext)).
    :- use_module(library(lists)).

    ab(a).
    ab(b).

Sample query:

    ?- length(Ls, 1_000_000),
       maplist(ab, Ls),
       partial_string(Ls, Es0, []),
       Es0 == Ls.
       Ls = "aaaaaaaaaaaaaaaaaaa ...", Es0 = "aaaaaaaaaaaaaaaaaaa ..."
    ;  Ls = "aaaaaaaaaaaaaaaaaaa ...", Es0 = "aaaaaaaaaaaaaaaaaaa ..."
    ;  Ls = "aaaaaaaaaaaaaaaaaaa ...", Es0 = "aaaaaaaaaaaaaaaaaaa ..."
    ;  Ls = "aaaaaaaaaaaaaaaaaaa ...", Es0 = "aaaaaaaaaaaaaaaaaaa ..."
    ;  Ls = "aaaaaaaaaaaaaaaaaaa ...", Es0 = "aaaaaaaaaaaaaaaaaaa ..."
    ;  ... .

running in constant memory.
2025-07-07 22:38:12 -07:00
Mark Thom
706ffe7ad9 fix functor! size calculations around indexing_code_ptr 2025-07-07 22:38:12 -07:00
Mark Thom
e20363b462 correct create_partial_string (#2593) 2025-07-07 22:38:12 -07:00
Mark Thom
22080f3787 move CodeIndex to F64Table-like table 2025-07-07 22:38:12 -07:00
Mark Thom
9e1e99f961 Revert "remove Term"
This reverts commit 3b5879841aedecba5057c70c71da0ba23e5cd84a.
2025-07-07 22:38:10 -07:00
Mark Thom
fadcfb966e copy partial string blocks properly in all solutions predicates 2025-07-07 22:02:27 -07:00
Mark Thom
ae4d12a123 remove pstr_vec 2025-07-07 22:02:27 -07:00
Mark Thom
c0f72704ec introduce bespoke Heap type for in-heap partial strings 2025-07-07 21:59:50 -07:00
Mark Thom
1ef681bd21 remove Term 2025-07-07 21:54:08 -07:00
Mark Thom
778068129f don't skip '.' char in parse_number_from_string (#2997) 2025-07-07 21:28:07 -07:00
Mark Thom
ce56b5ed6b modify number_token to produce partial lexing results (#2986) 2025-07-05 00:58:28 -07:00
Mark Thom
8e01eea440 be more careful about concatenating '.' in parse_number_from_string (#2986) 2025-07-01 18:40:49 -07:00
Mark Thom
90f4716cc3 Merge branch 'null-stream-safety' of https://github.com/adri326/scryer-prolog into adri326-null-stream-safety 2025-04-29 23:32:48 -07:00
revue_2_presse
5d468d3e19 Replace futures::executor::block_on with tokio::block_in_place 2025-02-25 20:31:19 +01:00
Emilie Burgun
d8213e29c5 Fix set_output/1 and set_input/1 not updating the alias
Before this change, the following set of queries would behave incorrectly:

```
?- open("/tmp/out.log", write, S), set_output(S).
   prints(""), write("/tmp/out.log", "S = stream(...)").
?- write(user_output, hello).
   prints("hello"), unexpected.
   prints(""), write("/tmp/out.log", "hello"). % Expected, but not found.
```

Now, `set_output/1` and `set_input/1` properly bind the `user_output` and
`user_input` aliases, making the queries above behave as expected.
2025-02-06 23:01:38 +01:00
Emilie Burgun
949d316773 Fix realiased streams causing close/1 to leave a dangling stream 2025-02-06 23:00:00 +01:00
Emilie Burgun
0cf46d3ec4 Encapsulate accesses to IndexStore::streams and ::stream_aliases
These two fields are able to hold `Stream` instances, which predicates like `close/1`
expect to be managed properly for their correctness. To ensure that this is the case,
I have removed direct accesses to those two fields, so that they can be properly managed
in one place.
2025-02-06 23:00:00 +01:00
Emilie Burgun
fdc35f8b40 Fix UB caused by interactions with null streams 2025-02-03 00:08:37 +01:00
Emilie Burgun
2fe7b55343 Fix close/1 messing up stream_aliases when user_input or user_output aren't set to Stdin and Stdout 2025-02-03 00:07:39 +01:00
bakaq
590d0daa24 Fix clippy warnings 2025-01-28 16:49:41 -03:00
Emilie Burgun
b76bdd75e4 Fix #2725 by calling load_context/1 in the unspecified branch of strip_module/3
This fixes #2725, by making it so that `strip_module(Pred, M, P), call(M:P)`
doesn't throw an `instanciation_error` when `Pred` isn't in the form `module:predicate`.

Now, `strip_module(hello, M, P)` will call `load_context(M)`, which unifies `M`
with the topmost module (or `user`).

Two new test cases are added: issue2725.pl, which tests the minimal case id(X) --> X.
and the strip_module(P, M, _), call(M:P) scenario, and module_resolution,
which tests the behavior of strip_module in a few scenarios.
2025-01-12 14:08:28 +01:00
Mark Thom
afc7b74bbd qualify module quantification errors better (#2685) 2024-12-25 12:25:01 -08:00
Mark Thom
d287095e61 style corrections 2024-12-25 12:25:01 -08:00