fcaa47986a
Improve error handling for invalid syntax doing [*]T
2024-07-22 16:29:29 +01:00
1e37eaf54d
Begin work for bit_set[...; [N]T] (not working)
2024-07-15 14:49:20 +01:00
0a530b5ce8
Add error for #no_capture being reserved for future use
2024-07-14 13:20:12 +01:00
3533094cc2
Restrict #no_capture to pointer-like types only
2024-07-14 12:44:13 +01:00
3dff83f3dc
Mock out #no_capture for future use
2024-07-14 12:39:30 +01:00
8642d719f0
Imply #no_capture to all variadic parameters
2024-07-14 12:19:47 +01:00
edc793d7c1
Add #no_capture args: ..T to reuse the backing array stack memory
2024-07-14 11:39:05 +01:00
c83307f26f
Give error message suggestion when doing ^x instead of x^ if x is an r-value
2024-07-10 12:39:42 +01:00
f85f6b338a
Change .. variadic logic; comment out unneeded error message
2024-07-10 12:28:15 +01:00
ae2f16edea
Fix #3894
2024-07-09 15:29:36 +01:00
f66b7021a6
Fix not detecting duplicate proc cases
...
Fixes #3864
2024-07-05 00:50:52 +02:00
ec38215842
Fix #3803
2024-06-28 10:09:46 +01:00
29250f2657
fix regression in test_issue_2395
2024-06-07 16:33:38 +02:00
68781f8dd3
Remove unnecessary Wait_Signal checks
2024-06-07 00:11:00 +01:00
7044a7d776
Try to fix a possible race condition with polymorphic record parameters
2024-06-06 23:55:48 +01:00
64bdb3a097
Fix #3630 (first part)
2024-05-26 12:17:03 +01:00
32245e93a1
Fix #3514 along with soa.a[i] bounds checking
2024-05-16 16:18:21 +01:00
8fa20fb875
Extra check for field being nullptr
2024-05-13 14:44:53 +01:00
7905f0533f
Fix #3582 by disallowing it
2024-05-13 12:01:16 +01:00
6bfaf4a093
Fix another oversight for bit_field endian with 1-byte types
2024-05-11 22:36:17 +01:00
811d53b305
Generalize to any 1-byte element in bit_field
2024-05-11 13:48:19 +01:00
aea28d5189
Allow edge-case where backing type of a bit_field is (array of) u8, to allow any endian type
2024-05-11 13:47:33 +01:00
f54977336b
With -vet-style, give suggestion of separating where clauses with a comma rather than '&&'
...
This improves the error messages
2024-05-09 15:56:00 +01:00
d85c8f0b2c
Fix #3555
2024-05-09 10:58:57 +01:00
3f7a369aa1
Check for specialization in typeid/T for parapoly records
2024-05-07 14:53:02 +01:00
0cec2d7827
Fix #3527
2024-05-07 11:51:06 +01:00
0da6a3e214
Fix #3530
2024-05-07 11:42:48 +01:00
6ec7845249
Fix #3531
...
Individual `bit_field` size was not being added to the total size.
Error message was changed to be more explicit.
2024-05-06 18:30:04 -04:00
2201f365a1
Allow #no_alias on multi-pointers
2024-05-03 14:51:02 +01:00
94d35d9918
Disallow mixing endian types within a bit_field
2024-04-24 17:31:31 +01:00
ec5a84a537
Improve code generation for loading bit_field fields
2024-04-24 13:10:58 +01:00
7cd2bc26f4
Clear error message on collisions with using on struct fields
2024-04-17 13:31:32 +01:00
a61ae7c861
Fix #3427
2024-04-16 13:31:49 +01:00
efc3f9916e
Fix #3414
2024-04-12 12:30:16 +01:00
219eb58c08
Fix typo in using parameter error message
2024-04-09 12:03:41 +01:00
b47d73c651
Fix type checking for invalid enum backing type
2024-04-01 13:34:30 +01:00
8e1a2094a7
Fix debug info for map
2024-03-30 13:46:23 +00:00
4edcaa6124
Try storing a pointer to a fake metadata type in the debug info for a map
2024-03-30 10:29:20 +00:00
1009182f7b
Fix another #soa race condition bug
2024-03-26 14:13:55 +00:00
533ba63c82
Fix #3327 #3204 #3200
2024-03-26 13:06:39 +00:00
600ca83386
Merge pull request #3326 from rick-masters/fix_fields_wait_signal
...
Fix fields_wait_signal futex.
2024-03-25 13:29:08 +00:00
e5629dafd0
Potentially fix a race condition with parapoly types (related to #3328 )
2024-03-25 13:23:43 +00:00
6d4f30de1a
Fix fields_wait_signal futex.
2024-03-24 16:28:55 +00:00
1d46adb598
Treat *x as an unary operator to improve error messages for common C-programmer mistakes
2024-03-23 16:51:04 +00:00
61aa4558dc
Fix #3314
2024-03-23 16:41:50 +00:00
6e2efce670
Fix missing ERROR_BLOCK(); calls
2024-03-21 12:05:15 +00:00
29e5f94c2a
Add #no_broadcast procedure parameter to disallow automatic array programming broadcasting on procedure arguments
2024-03-21 11:52:48 +00:00
a750fc0ba6
Add #row_major matrix[R, C]T
...
As well as `#column_major matrix[R, C]T` as an alias for just `matrix[R, C]T`.
This is because some libraries require a row_major internal layout but still want to be used with row or major oriented vectors.
2024-03-19 21:05:23 +00:00
5159f30c9c
Fix error block handling
2024-03-19 13:10:35 +00:00
d5daa9fda5
Minimize error propagation of bad array syntax by treating this like a type
2024-03-19 13:05:28 +00:00
gingerBill
Laytan Laats
Feoramund
Chris
gingerBill
rick-masters