orderby consumes the entire input enumerable in 'sorting' the list;
since we don't care about the order besides a yes/no, just manage the
yield order ourselves.
don't bother using this method more than the spot it's in; was a fun
exercise but other spots are set up for easier debugging
(overall flow is where->deferby->yield), could make things easier to
read but meh
convert all dictionaries to char-byte instead of string (saves 50KB on
compressed dll, lul)
update OT string comparison for pre/post transfer specimens
Showing "TRAINER" for all languages isn't correct, just show a mapped
character
Closes#2049 , thanks @egzonqj & @WEERSOQUEER !
all usages besides fetching a 32bit random value should use rand.next
remove unnecessary do-while loop for calculating random EVs (always
returns 510 in total)
push unsaved IVTotal/EVTotal stuff from prior commit
rework exposed types for evo chain
initial trim now uses much less linq
can probably redo the get initial chain to provide a species to break
on...
continued simplification
core.cs is pretty much the following:
misc one-off junk
moveset fetch
encounter table fetch
evo chain fetch
might eventually move things around a bit further for cleanliness so
that misc meta junk is the only thing that remains in core.
update handling that checked for this case to instead check for
equivalence to ability1 instead of 0
was generating a PGF with ability[1] = 0, which is not correct. Just fix
the binaries to behave and get rid of all the workarounds since future
tables don't have missing values.
bypass savefile language checks if no language set
fix gen2 static gift egg that is really a wild encounter -- egg flag was
set & not cleared -> no wild pkm, bad! Adds test case for this.
(probably isn't a better place to clear, enforces only once per
generation rather than on every yield/end of every static yield)
fix b2w2 egg base levelup reference (not bw, b2w2!), doesn't really
matter much except for better indication (possibly?)
remove some unnecessary linq ToArray() calls
continue relocating code out of Core
Closes#1317
Needs the events crew to produce serialized binaries with hash-flag data
for each generation before proceeding any further. I'm not really
interested in doing all the work for events since it doesn't impact
battle legality.
nickname reset clears except for encountertrades (some have fixed
nicknames).
remove untraded check for HT memory verification, has to be traded if it
has a HT memory.
stored as wc3's, make "Version" from IVersion rather than int so that
"WasXD" recognizes it.
Update handling for fateful encounter trade-away cases.
#1970
Had relearnmoves defined but no current moves; better for generator to
spit out all moves rather than the special move (hold back)
store current moves for all link gifts
Remove edge case handling for level 2 blissey; updates over the past 2
(yay 2 years of legality checking) can now handle that case without
special handling.
Store 4g giratina held item for enc->pkm legality
revise usages of GetSaneGender
Rearrange some logic
Better handle impossible version encounters (gen4)
gen 1-5 done, stuck on 659 - bunnelby egg...?
* B2 has an error in the encounter table for Route 3 Corphish Special
water encounter; GF edited from Basculin-1 to Corphish-1 and forgot to
update the form to 0.
* Ranch Trade egg locations
* Trade gender-PID set correctly
* static roamer glitch does not apply for emerald
* Mirage Island wild Wynaut case now allowed (collision with Gift Wynaut
egg hatched on Mirage Island?)
gen1-3 done; now at 413 (Wormadam) which currently yields an
EncounterEgg AS wormadam, which isn't right...
* first case of EncounterLink, now works (version, ability, OT memory)
* handle gen2 gb events (not even sure if they pop up, just in case)
* handle gen3 ereader shadow mons
* handle gen4 spiky pichu PID type
* handle gen2 crystal egg met level
Handle Faraway Island Mew, handle DP Darkrai/Shaymin & Arceus, handle
not-distributed WC3s.
Now stopping at dex 154 -- one full generation of PKM species IDs can be
completely spawned directly from IEncounterable data!
Exeggutor Island has wild & one static with a fixed ability; the fixed
ability is returned first which causes wild lv40 exeggutors with
ability2 (1/2/H) to be flagged as illegal.
Fixes Krabby / etc.
Fix genderless encounterstatic gender set (Voltorb)
Now at 103, Exeggutor
pretty cool that over 100 species can have every single encounter
generate a legal pkm ;)
might be useful to add metrics to count how many encounters are
generated in a session
Graveler (SM Trade) -> Force applying handling trainer details for
ingame trades
Ponyta (Ranch Trade) Apply correct ability for encountertradePID
Up to 98 : Wild Gen2 Krabby (safari zone impossible location needs to
get filtered from possible encounters)
Fix/add missing gameversion references (BU instead of BW,
pokewalker=HGSS)
increase reuse of pidgenerator & add more generating methods.
seems like it's generating pkm fine besides the PID edge cases, which is
nice
version was set to Encounter_XD, which did not contain the miror b
encounters.
simplify XD check to be O(1) now that version is stored in the encounter
obj
Closes#1895 , re-verified with pkm provided in #1805
there's no shedinja gift, always is an evolution
seal some forms to remove virt call in constructor warning
egg moves also can be taught in gen1, which causes the learn source to
be locked to a tradeback (removing egg moves that have IDs greater than
RBY's moves). Simplify gen<=2 parse preference ordering with this
addition to force egg moves to be acknowledged.
only tested for gen7 egg->pkm (not to eventual set)
pretty sure a good chunk of the egg generation logic can be pulled out
for reuse in other IEncounterable generators (IEncounterable
extensions?)
Test parsing glaceon, input must equal output
Test generating encounters for a specific version (exists only on Moon)
Test generating encounters for the same version (add handling trainer
data to permit Tutors)
fixes error in VC generations-present fetch which allowed movesets from
generations 3-6 lul
exclude negative/invalid gameversions
party stats set when setting a slot to a save file
simplify set/delete slotchange duplicate logic
suggest better met locations beyond VC transfers
hatching a gen6 egg applies memories automatically
Two checks are performed for Gen 3 & 4:
A: Encounterability (wurmple, ball, ability)
B: RNG Compatibility (frame)
Original code only checked 'A'
Recent code ordered by 'A', but returned anything that matched 'B'. If
'B was not satisfied, an 'A' match would get deferred, allowing for an
eventual B & !A to be returned before a A & !B being returned.
yay bool logic
we want to prefer a RNGFrameMatch as well as a SlotCompatible match
original code (no RNG checks) retrieves with SlotCompatible just fine
recent adding in framematch resulted in deferring compatible if not RNG
frame, yielding !A matches before A & !B
maintain 3 queues (A & B get yielded immediately) to track A!B, !AB, and
!A!B
not necessary to check unown case anymore as it'll fall into the A!B
group. better performant to skip that check in favor of only adding for
fail.
thanks theSlayer!
Gen5 does not follow the same convention, 0 = non, 1 = rnd, 2 = always;
not gonna bother updating for just that one
bool? occupies 2 bytes; enum:byte is 1 byte.
should probably move validity checking logic into the IEncounterable
objects instead...
faster to do:
null check & comparison for each
vs
collection add (especially if growing collection)
don't evaluate all slots (faster), capturing the dexlevel[] object is
fine until the enumerable is discarded
WinForms->Core logic absorbing (CommonEdits)
loading ShowdownSet now applies properties to PKM instead of PKMEditor
Contest/IVs for Static/Trades are no longer set by default (less object
allocations), and are now checked by the encounter generator
great job gamefreak
allow current handler to be 0, remove static encounter as the data was
sneakily faked as a WC7 in the events gallery.
Closes#1817 ; still needs to update with new pikachu pid method
allowing shinies (next update).
Level = Level is equivalent to Level = 0
update EggInfoSource to use Level instead of LevelMin (LevelMin returns
Level, just be clear on usage)
update VerifyCurrentMoves so that only gen1/2 do the gen1/2 move update
methods. Recently extended all encounters to have a generation property;
we only care about gen1/2 here.
https://projectpokemon.org/home/forums/topic/43655-pokemon-crystal-egg-mark-as-not-legit/
closes#1755
also:
fix vc transfer checks not being called (EncounterOriginalGB is no
longer GBEncounterData).
remove usages of GBEncounterData -- class is (as of a ~~month ago) now
never leaked out of the EncounterGenerator -- always returns the actual
encounter data as the wrapper is not needed.
simplify logic flow / references
indicates met location for transferred mons
most gen1/2 encounters won't show values due to the location not being
stored respective to the string tables (anyone wanna do a location remap
after initial load?)
previously was only defined for gen1/2, why not all gens
can now query encounters in Core for encounters per generation via
reflection; can help group encounters for data vis or be faster than
getting individual gen numbers
extract EncounterStatic match determination; in doing so fixes "lvl" by
passing byval.
Pichu encounters (2) come before the Gift level 25 pikachu, causing lvl
to be dropped to 5. When pikachu comes around, it compares 5==25 -> bad;
pulling out the method and passing byval allows it to be 25 at the start
of every new encounterstatic checked.
get language list now doesn't return new objects (or re-enumerate)
update rand usage to be inclusive for top bound, extend shuffle to
collections
remove unnecessary location overrides (already overriden in legal fetch)
#1700
dp personal: add wormadam/deoxys manual forme count values
generator: bcc handling for pkm present in & out of bcc, and static/wild
duplicates
util: fix incorrect variable reference
using certain slot modifiers (swarm/game pak) causes different
arrangements of slots for different static/magnet pull groups to pull
from
store a list of permuted/different slots and add them to the table at
the end with the rest
move static/magnet pull marking into gen4 methods (only leave for gen3)
move trophy slot generation into gen4dppt area fetch (necessary for
static/magnet permuting)
fix electric off-by-one (yay for curse ??? type shifting everything)
dynamic level -> calc level
no sweet scent -> proc available (rock smash/fish)
change frame checking in encountergenerator to save which frame it was
found on (unused but possibly later can capture this for output)
Turns out there was no documentation anywhere on encounterslot numbers
for the individual forms; some sites had overall rates.
Assuming a human filled out the encounter slots, forms would repeat left
to right until a new form takes over. Below slot numbers match the rates
when summed up.
https://pastebin.com/raw/bQt3ZMbq
via pret/pokeemerald (and pokeruby); rate is set at 50%
swarm pokemon have the ability to have 4 fixed moves, so derive from
EncounterSlot to have moves, and remove special handling from
VerifyCurrentMoves (now picked up by IMoveset pattern
pull out transfer locations to const references
add vc2 & crown beast/celebi met location suggesting
add vc2 crystal sensitive detection
add 2 more usum trainer stats (thanks holla!)
move MarkEncounterTypeData back above the area size reduction so that
the mark-by-locationID special handling will work
prevent slottype from becoming overwritten if already defined by special
case handling
add examples provided to automated test cases
closes#1648
GetValid already filters by species
move some common non-location code out of the method; can be done for
the dexlevel fetching too -- future improvement.
don't bother converting to pkm then generating the sprite off that; use
the overload with int/bool
removes sav/pkm dependency for generating gift descriptions, and
standardizes the names a little. maybe a PKMBase interface would be
useful...
fix roamer3 not setting IVs
fix usum z items rightmost pixel getting chopped off
use Gen* instead of GenNumber for specific cases (faster)
add WasGiftEgg location case for Gen7
remove some unnecessary array allocations
sourced from Ultra Moon, location IDs are not known (need manual docs).
Many of these may be unused, like last gen. Will eventually remove
unused ones once crowdsourced effort finishes.
force colo starters to be male only
use some c# lang features for EncounterType flags
add edge case PIDIV detection for starter gender/shiny lock scenario
adds a little documentation for those unfamiliar with the NPC PKM
generation quirks.
Colosseum Starters will now be recognized with a different PIDIV type
which is specific to them & them only.
checks were previously only performed when finding the encounter, once
found, the GB encounter was validated. the transfer encounter is derived
after all encounter checks, thus remaining unchecked (really, only for
Locations).
Move generating logic to the generator, and double check the locations
are valid.
#1535
Thanks @DJPanda065 !
fetch is exclusively done with version already known; there are cases
where version can change for eggs so it's safer this way.
version change reference (gen6+):
212a489335
move must be removed 2->1 so cannot be permitted as egg move.
remove unreachable code (ParseMovesRelearn only is called for Generation
>= 6, and thus Format != 2).
- G1/G2 only have Hiragana letter he -> fixed Houndoom(ヘルガー) /Heracross (ヘラクロス) default name mismatch
- Fixed Stadium2 Farfetch'd move (Thrash -> Slash)
- Added VC Mew TID check
- Update Chinese legality strings
Korean g/s cannot access gen1 movesets, so pruning gen1 was necessary.
Rather than rebuilding the array (slow), just futureproof the entire
method to not rely on '7' (in case this code is ever used on Gen8).
Card 0525 was accidentally uploaded without the 3 flawless IVs setting;
sometimes when the internal event binary is built, the 3IV card comes
before the 0IV.
Play favorites with the 0IV.
Restriction happens because Korean can not trade with non-Korean GB era games
- There is no Korean release for gen 1 pokemon, included VC
- With no gen1 pokemon means any Korean gen2 is Gen2_NotTradeback, that means no gen1 origin nor moves are Legal
- Crystal was never released in Korean
- Pokemon Stadium 2 was never released in Korean, that means no move reminder for gen 2 korean pokemon
- Generation 4 can not trade between Korean and not Korean games, but Korean games can use the palpark with any language
Chinese language restrictions
There is no Chinese release for gen 1 and 2 pokemon games, VC Chinese games are in Japanese
farfetch'd apostrophe mapping disabled for gen5 and prior
Closes#1474
flag gen2 new moves as not obtainable if transferred from RBY (VC1) as
they must be removed before transferring 2->1.
Initialize move array for WC3 if none are present
Closes#1472
More gracefully handle invalid version pkms in a given format (gen5
origin version pkm on gen4 format no longer throws an exception).
I think egg hatch location checking is wrong for traded eggs (see gen3
vs future games, origin game shouldn't change but could hatch ORAS egg
on XY at XY location?)
Checks for max language ID on transferred pkm (gen5 pkm on gen7 can't
have CHS/CHT etc)
handles edge case for JPN B/W ingame trades
update pk5->pk6 transfer logic to bump langID 0 to 1.
Thanks HaxAras & Britty for providing samples!
Copied binaries from sm, references set up for saving time later
I assume not much will change in terms of structures, so it should just
be a file replacement & filling in static encounter table => 'working'
legality checks
Allows evolution met level check to be skipped
Adds minimum level constraint for evolution chain generating
Closes#1430
also adds deferral to the gen3 encounter generator as encounterstatic
gifts can collide with wild encounters when transferred
personalinfo setup wasn't populating info for form indexes
remove erroneous pid-type mismatch message on gen3/4 non-matching if
there are no deferred encounters
The encounter generator was returning always RBY and GSC and that make the ParseMovesGenGB to check only for rb and gs learnset for initial moves, the game returned should be the game for the encounter to make ParseMovesGenGB use the correct learnset table for the initial moves
Closes#1417
Add global link mission stats (thanks Holla!)
remove some ToArray() linq in favor of direct copies
Relocate encounter suggestion logic to separate class
Closes#1396, addresses other edge cases like entree-non HA & happiny
egg.
Fix misc refactoring errors
Don't early abort gen4->5+ encounter generation (need to check ingame
trades for abra/gengar). Can avoid by PID dictionary to check for trades
first, but unimplemented.
Closes#1377 , not fixing Pokewalker PID mismatch as algorithm needs to
be ironed out separately.
Add some documentation for the high-level encounter finding
Fix encounterstatics not being filtered for the associated game version
(closes#1372)
was a side effect of refactoring this week
shifts encounter data from Legal's split tables to individual files for
easier maintenance and initialization. Legal Core's init is so much
simpler now.
fix resource name typo
GetGenMovesCheckOrder doesn't respect generation restrictions, while
EncounterMoves does; filter afterwards.
Add compiler flags in the Analysis file to toggle try/catch off for
easier debugging. Just uncomment the #define SUPPRESS line.
Closes#1345
PGF for movie victini has OTgender=3; only eggs are supposed to set OT
details from the SAV.
update pattern matchers to abuse this fact.
Thanks @odaxis !
Rarely ever used, the only case where an unfavorable match is returned
first is Emerald Lati@s transferred to a future generation, as there are
2 encounters (one with and one without Fateful).
Closes#1326
Thanks @sora10pls !
Indicates why another type of encounter to match to was rejected, mostly
useful for diagnosing past gen specimens to figure out what changes need
to be made for that encounter to persist.
supply met level so the static encounter has a level to work with
instead of 0
Closes#1316
in other areas:
level 48 charizard VC1 transfer changed to mewtwo causes minlvl >
maxlevel, handle instead of throwing exception.
* Added location to encounter slot to make verification againts the location of the encounter when the pokemon has lost met location, like generation 2 heabutt tree, jhoto surfing in route 45 and in the future generation 4 munchlax tree
Added version to generation 1 and 2 encounter locations to filter by catch rate based of the version of the encounter and check initial moves of the encounter only for the game that match the encounter
Filter generation 2 pokemon for crystal who have met location based of the time of day when it was captured
Completed version to static and traded encounters for gen 1 pokemon, to avoid check a red encounter with yellow initial moves, if an encounter is possible in both games with diferent moves it is duplicated (like eevee), if it is possible in both games with the same moves is left as RBY, the encounter will only use red/blue moveset
Verify some invalid gen 2 encounters. Crystall heabutt encounters based on the TID, using the tree selection algorithm of the game to determine if the encounter is possible for the TID (implemented base on https://bulbapedia.bulbagarden.net/wiki/Headbutt_tree#Mechanics). Coordinates of Crystal trees obtained with the programa G2Map
Added checks for fishing encounters for unreacheable water tiles in gen 2, route 14, national park and the beta safari zone.
* Fix gen 1 static encounters and trade encounters filter by version
* Missing strings
* Fixes various gen 3 event templates.
* Adds more checks to PokePark 2005 eggs for DS Download Play.
* Fixes various gen 3 event checks. Still missing japanese events, but will add those another time.
* Adds 10 ANIV Celebi.
* Fixes PCJP 2003 eggs. Still doesn't use the exact formula though, but BACD-R is something to start with.
Add special ribbon detection (for wc3),
speed up seeds from IVs calc (flip top bit of seed to get the result for
other iteration)
Add a bunch of gen3 event data
Switch encounter slot generator to yield style to not require the entire
method to finish iterating
Although the pressure slot may be best-matching, it isn't the typical
match. Defer pressure slot clone to the end (defers the orderby
calculation)
There's probably a better way to do the pre-work (gen 5+ only operate
off one encounterarea... usually. Past Gens end up repeating the same
work for each area tested
slot matching runs slightly faster now due to the deferred execution
reduce 'as' usages for pk1 (cast & check once)
reorder some slot generation checks (early return prior to slot
filtering; not crucial as the linq wasn't evaluated yet).
Closes#1257
Disallow static encounters that aren't eggs or are for eggs (mismatch)
Check egg encounter level for IsWithinRange
Check egg met level while still is egg
Valid species/movepools are already determined via the dexlevel and
prior methods, no need to check again with current species.
edge case: sing froslass from CXD won't check gen3 moves (ie, sing)
Previous approach was 'most likely to find valid', this approach is a
little greedier to try to find a more fitting match. Will result in a
small increase in computation time, but trying out EncounterTrades and
static encounters first is preferred.
Fix Rhydon ingame trade gender. Bulbapedia is wrong, that's a Male
Rhydon with those IVs.
* Fix issue #1200
Minimum level for evolution moves in generation 7 is 2 thanks to the move reminder, also the evolution could be trigger at the minimum level and not the next level, except when the minimum level is the met level
* Gen 2 VC Speculation, it wont have Move Reminder because the only gen 2 move reminder was in Pokemon Stadium 2
* Missing variable
* Fix update generation 2 level moves
* Add move source to the check result for current moves, it will be used for analysis of evolution with move to determine how many egg moves had the pokemon and determine if the evolution move could be a egg move that was forgotten
* Verify evolution for species that evolved leveling up with an specific move learned, the evolution must be at least one level after the pokemon could legally learn the move or one level after transfer to the first generation where it can evolve
* Check to detect traded unevolved Kadabra based in catch rate and moves exclusive from yellow or red/blue
If pokemon have data exclusive from one version but is in another version that means it should be evolved
* Check no tradeback moves for preevolutions, like Pichu exclusive non tradeback moves for a Pikachu, that Pikachu could not have at the same time Pichu gen 2 moves and gen 1 moves because move reminder do not allow to relearn Pichu moves and gen 2 moves must be forgotten to trade into generation 1 games
* Legallity strings for non tradeback checks
* https://bulbapedia.bulbagarden.net/wiki/Pok%C3%A9mon_breeding#Passing_moves_down
Eggs only inherit a level up move if both parents know the move, that means genderless and male only moves could not have any level up move as an egg except the base egg moves because Ditto is one parent
Nidoran male and Volbeat excluded because they can breed with Nidoran female and Illusime
* Small check to not search for egg moves in genderless pokemon, generation 2 data include egg moves for Starmie
* Fix female only species
* Stomp is not a possible egg moves of Stanee
* Fix Steenee evolution move, it cant be inherited as an egg
All gens that store version number do not change the version ID on
trade/hatch; this info is preserved and can directly imply which game
generated the egg.
the 'version' is ignored anyway, just useful to track for gen2 eggs.
encounterinvalid isegg no longer throws exceptions when parsing moves
dropping invalid file paths to main window are now ignored (downloads
from chrome can be manually deleted but still dragged from the downloads
bar)
add pidiv case for manaphy gift
More elegant handling for PKM without nicknames that have another
language species name; use that language ID instead of the pkm's
language ID. Since ToLower is not called, Ho-Oh from 3/4->6 now
transfers correctly.
fix gen3 encounterstatic eggs while still on gen3 games; only check the
egg location if still an egg
tighten up gen3 PIDIV compatibility checks, roamer glitch is fixed on
Emerald, and static encounters can't get method 2, and can only have
method 4 if under software emulation (RS Box) since the emu speed is
faaaaast 🚤 /s
Fixes HGSS egg moves, and improves encounter detail detection.
example: togepi from FRLG transferred up, old detection couldn't
differentiate the gift egg from a regular egg; however, PIDIVs can lend
a hand -- eggs don't have a PIDIV!
fixed shiny BACD spread detection (in regards to antishiny rerolls).
Shiny WISHMKR now detected correctly.
Add exclusion for gen3/4 balls since met location requires a less
precise match; static encounters will never be in safari/sport balls.
Fix wurmple species ref, isn't smeargle
eevee gen7 egg has no moves; always generate eggs first for gen4+
Closes#1179Closes#1176 (prior commit)
there's probably a better way to optimize this (don't check wild
encounters if 100% sure it's an egg); best left for future optimization
since it's working fine atm.
Track gen1/2 origin game as well as generation
fix pk2 gender setting (incorrect bitmask & wasn't set from tabs)
extend female gender check for gen2 to apply to GS encounters -- if met
location is missing, was traded to gen1 which clears gender.
exercise in deferred execution/state machine, only calculate possible
matches until a sufficiently valid match is obtained. Previous setup
would try to calculate the 'best match' and had band-aid workarounds in
cases where a subsequent check may determine it to be a false match.
There's still more ways to improve speed:
- precalculate relationships for Encounter Slots rather than iterating
over every area
- yielding individual slots instead of an entire area
- group non-egg wondercards by ID in a dict/hashtable for faster
retrieval
reworked some internals:
- EncounterMatch is always an IEncounterable instead of an object, for
easy pattern matching.
- Splitbreed checking is done per encounter and is stored in the
EncounterEgg result
- Encounter validation uses Encounter/Move/RelearnMove/Evolution to
whittle to the final encounter.
As a part of the encounter matching, a lazy peek is used to check if an
invalid encounter should be retained instead of discarded; if another
encounter has not been checked, it'll stop the invalid checks and move
on. If it is the last encounter, no other valid encounters exist so it
will keep the parse for the invalid encounter.
If no encounters are yielded, then there is no encountermatch. An
EncounterInvalid is created to store basic details, and the parse is
carried out.
Breaks some legality checking features for flagging invalid moves in
more detail, but those can be re-added in a separate check (if
splitbreed & any move invalid -> check for other split moves).
Should now be easier to follow the flow & maintain 😄
