[FRLG RNG] Check free party slots in applicable RNG programs by dayfine · Pull Request #61 · dayfine/Arduino-Source

dayfine · 2026-10-29T19:11:55Z

The documentation pages of the RNG programs document the number of free party slots needed. Programs might fail after catching / receiving when there is no free slot, as the pokemon got sent to box.

This PR actually checks for those requirements at program startup, by introducing a general pre-startup hook for programs.

Testing

On a lab Switch with a full party (when catching Mewtwo), Static RNG on this build stopped with that message 5 seconds after starting, before any of its own setup; before, it ran five resets (about 30 minutes) and stopped with the seed error.

dayfine · 2026-10-30T02:25:15Z

 #include "CommonFramework/Options/LabelCellOption.h"
 #include "CommonTools/Options/LanguageOCROption.h"
-#include "NintendoSwitch/NintendoSwitch_SingleSwitchProgram.h"
+#include "PokemonFRLG_RngProgram.h"


This is in the wrong place

should PokemonFRLG_RngProgram not be sorted?

Moved: PokemonFRLG_RngProgram.h now sits with the other same-folder includes at the end, in alphabetical order among them, in all seven program headers (9f36e0d).

Yes, sorted now (9f36e0d).

dayfine · 2026-10-30T02:28:08Z

 using namespace Pokemon;

+void check_party_has_free_slot(ConsoleHandle& console, ProControllerContext& context, StartMenuContext menu_context){
+    console.log("Checking for a free party slot...");


why is not just in SerialPrograms/Source/VideoGames/PokemonFRLG/Programs/RngManipulation/PokemonFRLG_RngProgram.cpp?

Agreed, nothing else uses it. It's now a private helper in PokemonFRLG_RngProgram.cpp, and RngNavigation is back to upstream's version (9f36e0d).

Static, Wild, Gift and Roaming RNG read the Pokemon they catch or receive from the end of the party. With a full party it goes to the PC instead, every attempt reads the wrong Pokemon, and after a few resets the program stops with "The target Seed is missing from the list of nearby seeds", which points at the wrong setting. SingleSwitchProgramInstance gets start_program_game_checks(), a no-op hook called with the controller ready just before program(), for checking that the game is set up the way a program needs. The FRLG RNG programs now derive from RngProgramInstance, which implements it from a required party_requirement() (and start_menu_context() for the Safari Zone): with FREE_SLOT it opens the party and stops with a setup error naming the real problem, or skips the check if the party menu can't be opened. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

…clude The free-slot check is only used by RngProgramInstance, so it lives in PokemonFRLG_RngProgram.cpp instead of RngNavigation. The base class's header joins the other same-folder includes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine · 2026-10-30T04:43:17Z

LGTM

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Take the party check's game steps through a small PartyCheckActions interface, so check_party_requirement() runs against scripted steps in tests, and read the last occupied party slot from an image, so it runs on the party menu screenshots already in CommandLineTests. Behavior is unchanged. Tests: PokemonFRLG::RngPartyCheck (no requirement, free slot, full party, menu not opened) and PokemonFRLG::LastOccupiedPartySlot (Party_1..6 and the held-item/JPN/SPA party screenshots). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine · 2026-10-31T17:07:41Z

Added unit tests in bdbeb41 (new commit on top; behavior unchanged). The check now takes its two game steps (open the party menu, read the last occupied slot) through a small PartyCheckActions interface, so PokemonFRLG::RngPartyCheck runs check_party_requirement() against scripted steps: no requirement (menu never opened), one and five Pokémon (passes), full party (UserSetupError "Your party is full…"), and menu not opened (skips). detect_last_occupied_party_slot() also takes an image now, and PokemonFRLG::LastOccupiedPartySlot runs it on the party screenshots already in CommandLineTests (Party_1–Party_6, held-item, JPN and SPA). CI: Mac, Windows Clang, Ubuntu.

dayfine · 2026-10-31T17:14:34Z

well I was thinking whether we can actually check the program actually perform the party check, with the right setttings -- maybe harder / impossible?

Build each FRLG RNG program, load its settings from JSON, and check the party requirement and start menu its start check uses (WildRng on Route 1 and in the Safari Zone). Make RngProgramInstance's start hook final, so no RNG program can skip the check by overriding it. Behavior is unchanged. Tests: PokemonFRLG::RngProgramPartyRequirement. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

The party screenshot test expected three; the existing empty-slot tests for the same image mark slots 5 and 6 empty. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine · 2026-10-31T17:31:13Z

whether we can actually check the program actually perform the party check, with the right settings

Mostly, yes. fda0033 adds PokemonFRLG::RngProgramPartyRequirement. For each RNG program, it builds the real program, loads its settings from JSON the way a saved profile does, and checks what the start check will ask for. Gift, Static, Roaming and Wild need a free slot. Starter, Egg and RngHelper don't. WildRng uses the standard start menu on Route 1 and the Safari Zone menu at safari_zone_entrance. That commit also makes RngProgramInstance::start_program_game_checks() final, so the compiler stops any RNG program from overriding the hook and skipping the check. The requirement is fixed per program today; no setting changes it. Location only changes which start menu it opens.

What isn't unit tested: that SingleSwitchProgramInstance calls the hook before program(). That needs a running program environment (console, video, controller). It's one line in the shared wrapper, and the lab run that stopped on a full party covered it. SidHelper isn't an RNG program instance and has no party check.

85c066e fixes my expected count for JPN_Party.png (it shows four Pokémon). The CI links in my earlier comment are for the run with that mistake. All three pass at 85c066e, including the 23 new tests: Mac, Windows Clang, Ubuntu.

dayfine · 2026-10-31T17:32:56Z

+    const RngProgramPartyScenario scenarios[] = {
+        {.name = "RngHelper", .make_program = make_rng_program<RngHelper>, .expected_requirement = NONE},
+        {.name = "StarterRng", .make_program = make_rng_program<StarterRng>, .expected_requirement = NONE},
+        {.name = "EggRng", .make_program = make_rng_program<EggRng>, .expected_requirement = NONE},


actually this need 3 open slots: https://pokemonautomation.github.io/Programs/PokemonFRLG/EggRng.html#in-game-instructions

EggRng needs three free slots (the egg and the two wild Pokemon it catches to confirm the seed), and what RngHelper needs depends on its target: gifts, prizes, Togepi and the daycare egg pickup read the received Pokemon from the party, while encounters are only watched. So programs now say how many free slots they need instead of whether they need one, and RngHelper derives it from its Target setting. Tests: PokemonFRLG::RngPartyCheck adds three-slot cases; PokemonFRLG::RngProgramPartyRequirement covers EggRng and RngHelper targets. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E # Conflicts: # SerialPrograms/Source/VideoGames/PokemonFRLG/PokemonFRLG_Tests.cpp

….cpp at its size origin/main grew the file to its code-health allowance (1656 lines over the 1000-line limit); the image overload this PR adds needs three lines. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

# Conflicts: # SerialPrograms/Source/VideoGames/PokemonFRLG/PokemonFRLG_Tests.cpp

Fix the stale function name in the PartyCheckActions comment, name the party size, word the one-slot message for gifts too, sort the new Navigation.h include, and take the test image path as string_view. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

Table-driven input/expectation cases for the party check tests, a comment on each RNG program class and the new overload, and closing namespace comments in the new files. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

dayfine · 2026-11-10T03:16:19Z

+        return "Your party is full. Leave a party slot free so the program can check the "
+            + Pokemon::STRING_POKEMON + " it catches or receives.";
+    }
+    return "Your party has " + std::to_string(party_size) + " " + Pokemon::STRING_POKEMON


strformat + raw string?

Now std::format for both messages (8490e0c); the text is unchanged, so the tests still match. I left out the raw string: the messages have no quotes or backslashes, so R"(...)" would read the same.

dayfine · 2026-11-10T03:17:54Z

    PA_ADD_OPTION(NOTIFICATIONS);
 }

+size_t RngHelper::free_party_slots_needed() const{


do we need this? I hope not?

It's there because RngHelper is the one program whose target is a setting, and only some targets read the received Pokemon from the end of the party (check_for_shiny() → shiny_check_summary()). Options to drop the switch: (a) always ask for 1 free slot except for starters (wild/static users would need a free slot they don't strictly need), (b) no check in RngHelper (return 0; it's the manual tool), (c) one shared shiny_check_reads_party(target) next to check_for_shiny() so the list lives in one place (touches RngNavigation again). Which do you prefer?

Went with (b): RngHelper returns 0 with a one-line comment, the per-target switch is gone, and RngHelper.cpp is back to upstream's version. The party tests keep one RngHelper case (expects 0). (a48b54d)

…g needs no slot Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

dayfine · 2026-11-10T03:38:11Z

Upstream: PokemonAutomation#1881 (branch upstream/frlg-free-party-slots, one squashed commit, same tree as this PR's fixes).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine force-pushed the frlg/party-room-check branch from 5e9ebb5 to 760a5e1 Compare October 29, 2026 19:38

dayfine changed the title ~~FRLG RNG: stop at the start if the party has no free slot~~ FRLG RNG: check the party before the program starts Oct 29, 2026

dayfine force-pushed the frlg/party-room-check branch 2 times, most recently from 13d4322 to 707b17e Compare October 29, 2026 20:57

dayfine commented Oct 30, 2026

View reviewed changes

dayfine and others added 2 commits October 29, 2026 19:30

dayfine force-pushed the frlg/party-room-check branch from 707b17e to 9f36e0d Compare October 30, 2026 02:34

dayfine mentioned this pull request Oct 30, 2026

code_health: class-size and include-layering checks, base from upstream merge-base #70

Merged

dayfine and others added 4 commits October 29, 2026 22:31

Merge remote-tracking branch 'origin/main' into frlg/party-room-check

f194f48

Merge origin/main into frlg/party-room-check

5490a0d

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Merge origin/main into frlg/party-room-check

180a870

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine and others added 2 commits October 31, 2026 10:16

FRLG RNG: JPN_Party.png has four Pokemon

85c066e

The party screenshot test expected three; the existing empty-slot tests for the same image mark slots 5 and 6 empty. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine commented Oct 31, 2026

View reviewed changes

Comment thread SerialPrograms/Source/VideoGames/PokemonFRLG/Programs/RngManipulation/PokemonFRLG_RngHelper.h Resolved

dayfine mentioned this pull request Nov 1, 2026

AutoStory M7 Names and starter RNG #141

Open

dayfine added game:frlg FireRed/LeafGreen rng RNG programs labels Nov 1, 2026

dayfine and others added 5 commits November 2, 2026 12:59

Merge origin/main into frlg/party-room-check

df6210a

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E # Conflicts: # SerialPrograms/Source/VideoGames/PokemonFRLG/PokemonFRLG_Tests.cpp

Merge origin/main into frlg/party-room-check

3618ff3

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Merge origin/main into frlg/party-room-check

1dd5900

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Merge origin/main into frlg/party-room-check

51ae1a6

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

dayfine and others added 6 commits November 7, 2026 07:01

Merge origin/main into frlg/party-room-check

13ce3e5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Merge remote-tracking branch 'upstream/main' into frlg/party-room-check

64af545

# Conflicts: # SerialPrograms/Source/VideoGames/PokemonFRLG/PokemonFRLG_Tests.cpp

Merge remote-tracking branch 'origin/main' into frlg/party-room-check

b240e06

Merge remote-tracking branch 'origin/main' into frlg/party-room-check

3a152c4

dayfine commented Nov 10, 2026

View reviewed changes

dayfine and others added 3 commits November 9, 2026 19:19

FRLG RNG party check: std::format for the message; note why StarterRn…

8490e0c

…g needs no slot Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

FRLG RNG party check: no check in RngHelper, the manual tool

a48b54d

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

FRLG RNG party check tests: std::format for the failure messages

fa6ff20

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

dayfine changed the title ~~FRLG RNG: check the party before the program starts~~ [FRLG RNG] Check free party slots in applicable RNG programs Nov 10, 2026

dayfine and others added 2 commits November 10, 2026 07:02

Merge origin/main into frlg/party-room-check

d3df2bc

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Merge origin/main into frlg/party-room-check

eb3fa61

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AKYzhrMnPK61XHWEch4a6E

Conversation

dayfine commented Oct 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Testing

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dayfine commented Oct 30, 2026

Uh oh!

dayfine commented Oct 31, 2026

Uh oh!

dayfine commented Oct 31, 2026

Uh oh!

dayfine commented Oct 31, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dayfine commented Nov 10, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

dayfine commented Oct 29, 2026 •

edited

Loading