gbdev/rgbds
1
0
Fork 0
mirror of https://github.com/gbdev/rgbds.git synced 2026-08-27 09:05:17 +00:00
Code Issues Packages Projects Releases Wiki Activity
3,643 Commits 3 Branches 52 Tags
a2c52caca24584b732c4e5c13dce78df9f0241d1
Commit Graph
384 Commits
Author SHA1 Message Date
Rangi 604c69e048 Fix palette-packing decant logic to avoid UB from invalid iterator (#2031) 
The "Iterate through the component's color sets, and transfer them"
logic shifted `from.begin()`, which would make subsequent `std::advance`
calls go past `_assigned.end()` and cause UB. This rewrite uses absolute
numeric indexes into `_assigned` to avoid that potential problem.

I haven't been able to craft a test case that actually *reaches*
that piece of logic, and I suspect that it's unreachable given how
we sort and process color sets before decanting, but cannot prove
that; so it stays in. (Maybe if we added 4bpp 16-colors-per-palette
support, it would become reachable?)
 2026-07-22 23:48:02 -04:00
Rangi 9cbfec30c7 Avoid redundant "Unexpected text" error message after "Unknown" error 2026-07-21 13:54:11 -04:00
Rangi 0af28de791 Fix assertion failure on nonzero base palette ID without an attribute map or palette map 
Also fix palette map IDs to wrap at 256, not at 8 like attributes
 2026-07-19 23:59:38 -04:00
Rangi 7e11ab4d8f uint16_t tileIdx type matches options.baseTileIDs 
Tile IDs wrap at 8 bits, but that's handled by `uint8_t tileID`
 2026-07-19 23:33:01 -04:00
Rangi e84bcf83b4 Use Procedure concept to type-check decantOn lambda 2026-07-15 23:27:08 -04:00
Rangi 5aa7067876 Comment on how overflow to 0 is explicitly allowed for tile and palette IDs 2026-07-15 22:28:08 -04:00
Rangi dff176791c Correct grammar "is" vs "are" in RGBGFX reverse mode error messages 2026-07-14 11:29:30 -04:00
Rangi 07f5969005 Consistency check the palette map, not just the attribute map 2026-07-14 11:27:41 -04:00
Rangi d61c0561c1 Correct condition to completely print both palettes and options.palSpec 2026-07-14 11:23:03 -04:00
Rangi 1ed22cc518 Mask blue for consistency (avoids transparent bit bugs) 2026-07-14 10:34:27 -04:00
Rangi 9a15b723f2 Format CGB colors in error/verbose messages as "GB:rr,gg,bb" 
These should be more debuggable than packed "$xxxx" format.
 2026-07-14 10:34:27 -04:00
Rangi e7855ea1b8 Fix reverse mode with -s/--palette-size too small for the tile data 2026-07-14 01:11:11 -04:00
Rangi c0ff3d92ad Fix reverse palette offsets when base palette ID wraps around to 0 2026-07-13 23:08:20 -04:00
Rangi e4583e7ac2 RGBGFX reverse mode checks for too-low palette IDs, not just too-high 2026-07-13 23:08:20 -04:00
Rangi 8b951bc047 Fix a flaky/random-seed rgbgfx_test test failure 
Commit 21eaa2d2cc missed some
intentional unsigned underflow that is needed to set `nbTilesInBank`.
Without it, some of the randomly-seeded RGBGFX tests, such as
`./rgbgfx_test seed8.bin -b 207 -N 54,256`, can fail with "Color
mismatch after round-tripping" errors.
 2026-07-13 21:31:05 -04:00
Rangi 59b6b15106 Run make format 2026-07-13 21:30:23 -04:00
Rangi 304cbb464f Factor out verboseDo to encapsulate magenta output color 2026-07-13 18:05:55 -04:00
Rangi ea3a8ee171 Use magenta color for some verbose output which was missing it 2026-07-13 17:47:38 -04:00
Rangi 4a6ea35299 Avoid overflow in counting tiles for unrealistically large images 2026-07-13 17:47:38 -04:00
Rangi 21eaa2d2cc Fix a > to >=, and avoid underflow in some sanity checks 2026-07-13 13:14:06 -04:00
Rangi 4857ad1e00 Fix crash on image without any colors (all transparent) 2026-07-11 13:18:40 -04:00
Rangi 3b33c942ae Fix a crash and a limitation in RGBGFX unoptimized output 
- If both banks 0 and 1 were completely filled, then
  `assume(bank == 0)` would false and potentially crash.
- If the input image had fully-background tiles, they would be
  incorrectly counted towards `nbTiles` and could cause the
  "Image contains N tiles, exceeding the limit" check to fail.
 2026-07-11 12:32:31 -04:00
Rangi be4c1161bd Fix RGBGFX crash when trimming more tiles than exist after deduplication 2026-07-10 17:35:32 -04:00
Rangi 118f1f7543 Correct option letter in error message 2026-07-10 16:59:30 -04:00
Rangi eaa35ed26d Remove dead return after no-return fatal call 2026-07-10 14:36:46 -04:00
Rangi 7b39bcd412 Refactor expression for clarity 
`tileData[realY + 1 % options.bitDepth]` relied on order of
operations and the fact that `1 % 1 == 0` but `1 % 2 == 1` for
correct indexing. However, it was not obvious at first glance,
and confusable with the more common `(base + offset) % size`
pattern. (And would need changing anyway if we ever support 4bpp.)
 2026-07-10 14:30:25 -04:00
Rangi 255cd7f7f6 Fix a typo in a comment 2026-07-10 14:29:32 -04:00
Rangi 39e7e064c9 Print correct color set limit value in error message 2026-07-09 21:29:02 -04:00
Rangi 7b97d7ce15 Fix RGBGFX crash on deduplicating transparent tiles 2026-07-09 21:24:59 -04:00
Rangi c667bac93a Use trailing return type for begin()/end() 2026-07-08 16:57:52 -04:00
Rangi db96144d49 Use [[nodiscard]] instead of [[gnu::warn_unused_result]] 2026-07-08 16:44:15 -04:00
Rangi 276a1a7952 Use const_cast more precisely to remove const without changing type 2026-07-08 16:42:53 -04:00
Rangi 5aedf31b46 Prevent 32-bit multiplication overflow 
Basically impossible to achieve, but just in case
 2026-07-07 17:02:46 -04:00
Rangi 6e643406a2 Refactor warning diagnostic code to reduce repeated output logic 2026-07-04 23:02:16 -04:00
Rangi 2f604bfa9a Fix error message when an rgbgfx output file cannot be created 2026-07-03 13:08:28 -04:00
vulcandth 998f636495 Fix rgbgfx -Z palette overgeneration on merged color sets (#1912) 
- Fix logic for color set comparison (which affects sorting them)
- Prune color sets which are proper subsets of newly-encountered ones
  (a comment implied we were already doing this, but we weren't)
- Add more verbose logging to debug this behavior
 2026-06-07 09:55:05 -04:00
Rangi 996469ee28 Ensure CRLF line endings are preserved when necessary 
Some test cases need CRLF line endings checked out even on Unix.
Also some source files had inadvertently contained CR bytes.
 2026-05-25 22:00:45 -04:00
Rangi dce14fd4b8 Use C++20 concepts to require constraints on template parameters (#1977) 2026-05-22 17:54:34 -04:00
Rangi 728bed39d5 Do not support GCC 9 (#1978) 
This will let us use C++20 features that GCC 9's experimental
C++20 support did not yet cover, such as "concepts".

This reverts some commits:
- 6bcd79b997
- d5ce5329ea
- 728d14879b
 2026-05-22 16:46:46 -04:00
Rangi 48fcd9a0ca Use templates to reduce the redundant number-lexing functions (#1963) 2026-05-21 23:13:09 +02:00
ISSOtm 6bcd79b997 Revert "Derive operator!= from operator== (#1660)" 
This reverts commit 993879a2ed.
 2026-04-26 23:30:59 +02:00
Rangi42 12186fdccc Keep more non-declaration initialization within the for loop clause 2026-04-19 22:01:56 +02:00
ISSOtm d054f07703 Fix some truncation warnings 
Amusingly, reported by MSVC only
 2026-03-10 01:31:23 -04:00
Rangi d053025070 Avoid writing "XXX" since it's a conventional "TODO/FIXME" comment 2026-01-22 13:01:10 -05:00
Rangi a9ab248fed Improve some RGBGFX error messages (#1876) 
* Improve some RGBGFX error messages

* Fix assertion failure on ambiguous transparent/opaque pixels
 2025-12-19 13:00:05 -05:00
Rangi 2666dcbc26 Remove exclamation marks and periods from error messages (#1874) 2025-12-10 11:50:33 -05:00
Rangi 33475e2c36 Factor out version-printing to usage.cpp (#1870) 2025-12-05 23:04:49 -05:00
Rangi 1eb4eb3339 Reuse the usage.name for printing version info 2025-11-18 22:32:45 -05:00
Rangi42 ca383c91ca Revert "More accurate 8-bit <=> 5-bit RGB color conversion (#1827)" 
This reverts commit 223b3d1921.
 2025-10-24 13:32:59 -04:00
Rangi efb5a88edb Show conventional colored "error:"/"FATAL:" for CLI option errors 2025-10-23 12:40:29 -04:00