Apple-2-SW/prog8
1
0
Fork 0
mirror of https://github.com/irmen/prog8.git synced 2026-04-21 02:16:41 +00:00
Code Issues Projects Releases Wiki Activity
5,337 Commits 1 Branch 128 Tags
d11386ef262e462eca05e5bded4fad16b2f26e40
Commit Graph

5337 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Irmen de Jong d11386ef26 type check tuning 2025-05-04 23:23:21 +02:00
Irmen de Jong 0e0377d1f0 IR/VM implemented struct allocations and initialization 2025-05-04 14:04:44 +02:00
Irmen de Jong 55e0dbab27 preparing for statically allocating struct instances 2025-05-03 23:44:29 +02:00
Irmen de Jong 4dc82f2c83 preparing for statically allocating struct instances 2025-05-03 19:00:27 +02:00
Irmen de Jong 1ba5587404 allow syntax for declaring variables with struct instance type 2025-05-03 16:14:22 +02:00
Irmen de Jong 835c4b6da3 allow multi-field declarations in structs, get rid of . -> ^^ rewrite 2025-05-03 12:32:29 +02:00
Irmen de Jong dbd955b61e tweak typecheck and better code for constant indexes in pointer derefs 2025-05-02 23:44:26 +02:00
Irmen de Jong d20e2fd88c structs will be a new major version of the compiler 2025-05-02 22:25:26 +02:00
Irmen de Jong e0dea89477 added support for ptr[x].field 2025-05-02 22:16:20 +02:00
Irmen de Jong 6fc2902895 fixing ptr traversal typecheck issues 2025-05-02 00:41:42 +02:00
Irmen de Jong c96e4b40d4 building syntax support for ptr[x].field 
attempting to do this by making '.' an expression operator
 2025-05-02 00:41:42 +02:00
Irmen de Jong 37da3e2170 parser 2025-05-02 00:41:42 +02:00
Irmen de Jong 2661d3c489 allow array syntax on pointers 2025-05-02 00:41:42 +02:00
Irmen de Jong b89bbb9281 allow pointers in subroutines params and return values 2025-05-02 00:41:42 +02:00
Irmen de Jong 696bf636ed better parsing of directive names results in better error messages when an invalid one is found 2025-05-02 00:41:42 +02:00
Irmen de Jong 40952a788a PtSub: params and returns now as children (in PtSignature node) for easier Ast walking 
PtPointerDeref: same but for its start identifier
 2025-05-02 00:41:42 +02:00
Irmen de Jong 0162e7a0c1 fix the scoping problems on subtypes 2025-05-02 00:41:42 +02:00
Irmen de Jong 6ce099f176 IR: fix ptr type checks and struct field assignment errors 2025-05-02 00:41:42 +02:00
Irmen de Jong 476a4bac8e IR: LOADI allows r1 and r2 to be the same for pointer chain dereference optimalization 2025-05-02 00:41:42 +02:00
Irmen de Jong 63a410a6df implicit cast to bool for numeric or pointers as condition arguments (to if, while, until) 2025-05-02 00:41:42 +02:00
Irmen de Jong cca27faa3b fix pointer value assignment (not dereferencing the actual pointer) 2025-05-02 00:41:42 +02:00
Irmen de Jong 803e6bd81a fix uword vs pointer type errors and casts 2025-05-02 00:41:42 +02:00
Irmen de Jong 88269628a2 had to turn ^type syntax into ^^type to avoid confusion with the eor operator once again 2025-05-02 00:41:42 +02:00
Irmen de Jong b920d553a0 make address-of dereference work 2025-05-02 00:41:42 +02:00
Irmen de Jong 5e2d0d0dfc fix param order of AssignTarget 2025-05-02 00:41:42 +02:00
Irmen de Jong 2ae3bd68eb more pointer dereferencing for chains 2025-05-02 00:41:42 +02:00
Irmen de Jong 9c183f27eb pointer dereferencing for chains 2025-05-02 00:41:42 +02:00
Irmen de Jong 8046023e82 pointer dereferencing for simple types (read and write) 2025-05-02 00:41:42 +02:00
Irmen de Jong e328520588 initial struct and typed pointer support 2025-05-02 00:41:40 +02:00
Irmen de Jong 7eb079050c fix strings.hash() on 6502 CPU's. NOTE: now takes string pointer in AY instead of R0 on all platforms v11.3.1 2025-05-01 22:47:43 +02:00
Irmen de Jong 2fdd5543b2 release 11.3.1 2025-04-30 22:40:54 +02:00
Irmen de Jong d04164c0a6 fix const evaluation of bitwise logical expressions (&, |, ^, <<, >>) of signed operands 2025-04-30 22:27:31 +02:00
Irmen de Jong b047731f82 removed some redundant arguments 2025-04-25 23:27:06 +02:00
Irmen de Jong 4d91f92a2e clearer error message 2025-04-25 19:38:45 +02:00
Irmen de Jong 98505d27b1 IR: fix invalid removal of label chunks and include asmsubs when looking for labels 2025-04-24 23:30:51 +02:00
Irmen de Jong cd63a58ad9 IR: get rid of problematic PREPARECALL pseudo instruction 2025-04-24 22:32:40 +02:00
Irmen de Jong 170f8dd092 IR: new CALLFAR instructions to support callfar into another bank 2025-04-24 21:17:48 +02:00
Irmen de Jong 619dcb6a84 fix error in IR register typing 2025-04-24 20:50:19 +02:00
Irmen de Jong 99ae8ea52e code cleanup 2025-04-24 18:51:03 +02:00
Irmen de Jong dc031c30eb add -version option to just print the compiler version and then exit 2025-04-24 18:13:35 +02:00
Irmen de Jong 1e702439b7 avoid JDK 21+/Kotlin method conflict of removeLast()/removeFirst() 
background: see https://www.reddit.com/r/androiddev/comments/1gspjrs/dont_use_kotlins_removefirst_and_removelast_when/
https://youtrack.jetbrains.com/issue/KT-71375/Prevent-Kotlins-removeFirst-and-removeLast-from-causing-crashes-on-Android-14-and-below-after-upgrading-to-Android-API-Level-35

it's about Android but the problem also occurs on desktop JDKs for example when running a Java21 compiled prog8 with Java17
 2025-04-24 18:11:42 +02:00
Irmen de Jong 8debc42381 slightly stricter assignment type compatibility checking 2025-04-24 15:05:35 +02:00
Irmen de Jong 532d719089 more optimal math.randrange() routines 2025-04-24 04:04:14 +02:00
Irmen de Jong b40860aca4 get rid of RTS after JMP 2025-04-23 17:08:16 +02:00
Irmen de Jong 2cbe6b5f7f info message when more optimal goto array[idx] is possible 2025-04-23 16:59:18 +02:00
Irmen de Jong d2cc7ccdfa remove redundant variable=0 initializations (BSS clear takes care of them) 2025-04-23 14:45:38 +02:00
Irmen de Jong 2cb183c6d8 fix regression for goto array[idx] on 6502 cpu which doesn't have jmp (ptr,x) 2025-04-23 02:56:10 +02:00
Irmen de Jong 84026b105f smaller code for goto nosplitptrarray[index] v11.3 2025-04-22 20:01:16 +02:00
Irmen de Jong a4d0589f10 fix errors like parse error still printed in color when -plaintext is set 2025-04-22 19:26:11 +02:00
Irmen de Jong e375f6afce fix diskio.f_read() returning 1 less than the actual size read. 
fix diskio.f_read_all() more robust error status end of loop checking.
 2025-04-21 05:02:16 +02:00