A value of another length was looked up as an MD5 and reported NOT FOUND,
a SHA-256 was dropped, and an input with no hash ended on "No valid
hashes found in input". Digests are now read by length (CRC32, MD5 and
its 29-digit Batocera form, SHA-1, SHA-256), SHA-256 is looked up in the
database, and any other value is refused with the formats accepted.
--platform retroarch --system sony-playstaton listed the first ten systems
of the alphabet. The warning now names the closest systems and the
command that lists the platform's own, as --system alone already did.
The region grouping and phase 2 of each pack and manifest called
find_undeclared_files with the same arguments: 30 reports for the 15
manifests of --all --manifest, 2.8 s, now 15 and 1.7 s. The 15 manifests
are byte-identical before and after.
Every member was deflated, archives and PNGs included. On the 2026-10-20
RetroArch pack the .zip, .7z, .chd, .png, .ogg and .pk4 members are 2.1 GB
that took 28 s to deflate again for 9.6 MB saved (0.45 %), and about 16 GB
for 72 MB across all packs. Those formats are now stored. The choice reads
the member name only, so a pack stays byte-reproducible.
_stable_hashes retried through the hash cache, so both retries returned
the first reading and the file was read once before the manifest run
stopped. The retries now read the file again.
Both are written as <platform>.json. --manifest-targets aimed at install/
replaced install/retroarch.json with a target table, and --manifest aimed
at install/targets/ did the reverse, both with exit 0. Each mode now stops
when the output directory holds the other kind and names the directory to
use.
The Required packs and manifests carried every data tree, the FreeIntv
overlays included. A data_directories entry now accepts required: false,
and the platform, emulator and manifest walks leave such a tree out under
--required-only. FreeIntv reads its overlays only with the multi-screen
option and draws a placeholder keypad without one (src/libretro.c:294-296,
350-369).
--verify-packs after an --all-variants build rebuilt only the full pack's
name and opened it alone. The five narrowed packs of the same run passed
without a line. Each one is now named with the hash check the build ran
on its members.
Principles, collection rules and the inverted model in architecture.md, research and citation rules in profiling.md, platform and scraper rules in their pages, tool conventions in tools.md, test and commit expectations in testing-guide.md, installer constraints in installer.md. Each rule sits where a reader looks for it, with its reason. Also fixes the stale release-job line, the suite command without -t ., and the unsourceable definition.
On Windows the partial copy of a file of 16 MB or more is locked with msvcrt.locking, a LockFile region: the locking process cannot access that region through a second handle (Microsoft LockFile remarks). Hashing reopened the path, the read failed with a sharing violation, and every large file was reported as 'folder not writable' (issue 82). Both digests are now read in one pass through the handle that holds the lock. A read-only file already at the destination had the same misleading report: on Windows its attribute is now cleared before the verified copy replaces it, and a refusal that remains is named for what it is.
RetroPie's YAML listed no standalone_cores, so every package was laid
out as a libretro core and the standalone files of amiberry, dolphin,
hatari, mame and ppsspp never reached the pack. Packages without the
lr- prefix (bar the retroarch frontend) are now standalone, and names
shipped both ways are dual-mode.
quasi88 writes its list "N88KNJ2.ROM, and N88SUB.ROM". Extended with
N88EXT0.ROM, which QUASI88kai reads, it came back as "N88SUB.ROM and
N88EXT0.ROM": every list was rewritten in one idiom. A list that puts a
comma before its final and keeps it.
EmuDeck passes Complex_4627v1.03.bin as xemu's flash ROM
(emuDeckXemu.sh:60-61), and xemu's entry Complex_4627.bin states no hash,
only the sizes it maps. The layer matched entries by name or content
only, so without the alias the default name was still asked for. A
hashless entry that checks a size now takes a launcher file of a size
it accepts, when no other entry of that core accepts that size too.
The removal candidates looked for a reading in the profiles only, so a
name the platform's own launch code opens, links or copies was listed
as read by no core: Batocera's DuckStation generator walks 22 names
(duckstationGenerator.py:501-515). Every name a launcher record names
is now a reading on that platform. Recalbox's unexplained entries drop
from 20 to 13, RetroBat's from 24 to 20, EmuDeck's from 25 to 22.
The 43 tos.img aliases in hatari.yml and 10 in hatarib.yml were names Batocera, Recalbox and RetroBat launchers pass to Hatari (hatariGenerator.py, Hatari.Generator.cs, libretroCores.py), now recorded in each platform's launchers: records. With standalone Hatari profiled, those records attach. Measured on Batocera, RetroBat, Recalbox, ROCKNIX and RetroPie: verify statuses, manifest files, omissions and Hatari entries are identical before and after.
Batocera (v2.6.1, hatari.mk), RetroBat (v2.4.1 release), Recalbox and ROCKNIX run standalone Hatari, which the profile did not describe, so their launcher records had no build to attach to. Standalone Hatari reads tos.img from its data directory by default (src/configuration.c:810-812, src/paths.c:316-330), retries a missing name with .img, .rom and .tos (src/file.c:275-308, 481-520), and exits when no TOS loads (src/main.c:191-201). The profile is now standalone + libretro, with the keymap file and the window icon as standalone entries and BOOT.ST scoped to libretro.
apple2gs1.rom and apple2gs3.rom come only from Recalbox's gsplusGenerator.py:107-110, 140, a platform launcher now recorded in platforms/recalbox.yml. No platform resolves gsplus.yml (Recalbox, Batocera and RetroBat run gsplus-legacy), and verify and the manifests of Recalbox, RetroDECK, ROCKNIX and EmuDeck are identical before and after.
RetroPie builds minivmac/minivmac, winterheart/quasi88 and TurtleBazooka/px68k under the package names minivmac, quasi88 and px68k (scriptmodules/emulators/minivmac.sh:12, quasi88.sh:12, px68k.sh:12). With built_by retropie and those names in cores, RetroPie resolves them to minivmac37, quasi88kai and px68k-sdl, while its libretro builds stay reached through retroarch. Measured: RetroPie keeps 532 of 532 platform files OK and gains the 14 files the standalone builds read; no other platform's resolution changes.
bios/Sharp/X68000/cgrom.tmp had cgrom.dat's sha1 (8d72c5b4). px68k reads cgrom.tmp only when cgrom.dat is absent (libretro.c:915-919, SDL/SDL2/winx68k.cpp:317-331), and a real cgrom.tmp is a font mkcgrom generates on the user's machine. No platform declares it.
The launcher layer packed a file under the launcher's name whatever the
platform installs: ROCKNIX's xemu config reads hdd/xbox_hdd.qcow2, and
the profile marks that image installed_on rocknix, yet the pack brought
one. A place whose entry the emulator installs on that platform is now
filled by the install, as validation.deliverable has it for the pack.
RetroBat's ares launcher tries v1.05, v1.02 then v1.01 (Ares.Generator.cs:371-437), now recorded under launchers: in platforms/retrobat.yml. ares itself identifies the image by sha256. No file carries the name, so the database is unchanged.
A names_searched list is one place the launcher fills with the first name it finds, so a core file it would never open once another name of the list is held is not reported missing. A place no declared file and no packed core file fills receives, under the launcher's name, a file the core accepts by content. The layer is applied inside find_undeclared_files, which verify and the pack builder both read, and a test reads their source to keep it there.
The scraper reads each component's prepare, launcher and functions scripts at the components commit the BIOS list comes from, and the presets they copy: the fixed paths set_setting_value writes (melonDS seven, xemu three, Dolphin GBA), the folders handed over, the dir_prep links (LRPS2 pcsx2/bios back to bios/, Dolphin and PrimeHack GC regions under saves/), package copies into bios/, RetroArch core options naming a file (pcsx2_bios, DuckStation per region), XRoar romlists, the PICO-8 exec and the RPCS3 and Vita3K firmware downloads.
platforms/retrodeck.yml is regenerated at fcd0d729: only the launchers section is new.
The scraper walks each package's functions the way RetroPie runs them in one shell (assignments, arrays, for loops, brace expansion, $md_id, calls into other packages) and records every link, copy, move and setting that names $biosdir, the addEmulator flags (sdltrs -romfile3), the BIOS read outside $biosdir (basilisk --rom, gngeopi -B) and the launch scripts that exit without their file (fs-uae kick13.rom, reicast dc_boot.bin).
Each package also gets a package record with its rp_module_repo and the systems it registers, which maps a package to the build it compiles. The schema gains package, repo, systems and the moves and package kinds. platforms/retropie.yml is regenerated at 46128cd2: only the launchers section is new.
The scraper now records, from configgen at the pinned tag, the names each Recalbox launcher tries and where it hands them: the Hatari TOS cascades (standalone, hatari and hatarib cores), GSplus, Amiberry kickstarts, the PCSX2 region choice, the Dolphin IPL presence test, xemu, the mupen64plus 64DD IPL, the Mini vMac boot disk and the o2em_bios option. The system_directory libretroRetroarch gives each core, and the per-system overrides after it, are passes_path records.
Recalbox starts its emulators from a Python configgen too. The parsed source, the record and the ast helpers move to configgen_launchers so both readers use them. Batocera's extractors are unchanged.
The scraper reads the EmuScripts at the commit checkBIOS.sh is pinned to: the BIOS paths changeLine, iniFieldUpdate and RetroArch_setConfigOverride write, the ares and melonDS configs whose paths sed rewrites, and the bios/ directories linked to Flycast, Azahar, the Switch emulators and shadPS4. The script to emulator mapping moves beside the extractors so both read one table.
The scraper reads the start scripts and the configs each package installs at the tag the BIOS list is pinned to: configs that name a BIOS path, the yabasanshiro search, directories handed to DuckStation and the PCSX2 builds, and the ln -s commands that put bios/ where an emulator reads. A package that ships one config per device gets a device field only where the values differ, as melonDS does on AMD64.
The scraper reads emulatorLauncher's generators and es_features.cfg at the commit the BIOS list is pinned to: Ares, Hatari, Mesen, Amiberry, WinUAE, PCSX2, the PSX BIOS cascades, Xemu, openMSX and DuckStation. Each extractor raises when its construct is gone.
The scraper parses the configgen generators with ast at the pinned tag: DuckStation and its legacy build, Hatari, xemu, melonDS, mupen64plus 64DD, x16emu, GSplus, and the o2em and neocd core options. An extractor that no longer recognises its code raises instead of writing an empty list.
"No core opens it" claimed more than the search shows: what was
checked is that no profile of a core the platform runs names the
entry by name, path or hash. The reason and the page sentence now
say that.
The page named the entries a change touched and never the values, so
a maintainer could not check a correction. The report now carries
changes, one object per change with kind, entry, field, old and new,
read from the platform entry and from the value the render writes
(written() per format: record value for System.dat and RomM, one md5
where a single one is checked). The existing label lists are derived
from it. The page writes "md5 X becomes Y" and names each required
flag with its new value. A test parses the original and updated
System.dat and es_bios.xml and checks the list equals the diff.
Where the truth names no value the platform lacks, hashes() still put
the truth's values first and rewrote the list in another order:
Recalbox's msx/CARTS.SHA md5 attribute came out with its two values
swapped, a diff line that changes nothing the frontend checks. The
platform's own list now stands when it already holds every value.
The export keeps every entry a platform declares. An entry the truth
does not pair was still read in 1189 of 2776 cases (alias, hash, a
type: directory tree, a core outside the resolved set), so absence of
a pairing proves nothing. Each report now lists, under
removal_candidates, the unpaired entries that no profile of the cores
the platform runs answers by name, path tail, hash, directory or data
directory, with the cores examined and a reason. A core the platform
names without a profile, or a ROM set under a core that opens sets by
name, leaves the verdict undetermined. The exports page states the
count and lists the entries per platform.
Abdessamad Derraz