Mirrors/PKHeX
2
0
Fork 0
mirror of https://github.com/kwsch/PKHeX synced 2024-12-20 17:33:25 +00:00
Code Issues Projects Releases Packages Wiki Activity
PKHeX/PKHeX.Core/Legality/Encounters/Verifiers/VerifyCurrentMoves.cs

750 lines
34 KiB
C#
Raw Normal View History

Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
using System;
using System.Collections.Generic;
using System.Linq;
using static PKHeX.Core.LegalityCheckStrings;
Move species/move names to parsesettings
2020-11-28 02:45:06 +00:00
using static PKHeX.Core.ParseSettings;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Misc verbosity reductions hide enum name
2019-02-22 04:44:31 +00:00
using static PKHeX.Core.MoveSource;
using static PKHeX.Core.Severity;
using static PKHeX.Core.CheckIdentifier;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
namespace PKHeX.Core
{
add more pkhex.core xml documentation adds a bunch of documentation useful for those unfamiliar with the core library
2017-10-24 06:12:58 +00:00
/// <summary>
/// Logic to verify the current <see cref="PKM.Moves"/>.
/// </summary>
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
public static class VerifyCurrentMoves
{
Misc clean split some methods with optional parameters=null add more xmldoc replace some magic numbers -> enum/const references consolidate common array operations (span soon maybe?)
2019-09-10 07:21:51 +00:00
/// <summary>
/// Verifies the current moves of the <see cref="pkm"/> data based on the provided <see cref="info"/>.
/// </summary>
/// <param name="pkm">Data to check</param>
/// <param name="info">Encounter conditions and legality info</param>
/// <returns>Validity of the <see cref="PKM.Moves"/></returns>
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
public static void VerifyMoves(PKM pkm, LegalInfo info)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var parse = info.Moves;
Array.ForEach(parse, p => p.Reset());
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
var currentMoves = pkm.Moves;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesForEncounters(pkm, parse, info, currentMoves);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
// Duplicate Moves Check
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
VerifyNoEmptyDuplicates(currentMoves, parse);
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
if (currentMoves[0] == 0) // Can't have an empty move slot for the first move.
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[0].FlagIllegal(LMoveSourceEmpty, CurrentMove);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesForEncounters(PKM pkm, CheckMoveResult[] parse, LegalInfo info, int[] currentMoves)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Use enum for species comparisons slightly easier to read with named values
2019-12-09 01:39:19 +00:00
if (pkm.Species == (int)Species.Smeargle) // special handling for Smeargle
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
{
ParseMovesForSmeargle(pkm, parse, currentMoves, info); // Smeargle can have any moves except a few
return;
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
// gather valid moves for encounter species
Move min level check for gen1/2 closer to move adder Check per-evolution min level rather than basing it on the original encounter data. Correctly parses stuff like a captured level 23 Nidorina that immediately evolves into Nidoqueen, then learns Body Slam. Passing zero is better than passing 1, as 0 requires no value pushing. tidy up method signatures in MoveLevelUp so that species is followed by form, and the lookups are textually aligned. Clarify "maxLevel" instead of max/lvl, same as min.
2021-07-23 04:25:15 +00:00
info.EncounterMoves = new ValidEncounterMoves(pkm, info.EncounterMatch, info.EvoChainsAllGens);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (info.Generation >= 6)
ParseMoves3DS(pkm, parse, currentMoves, info);
else
ParseMovesPre3DS(pkm, parse, currentMoves, info);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesForSmeargle(PKM pkm, CheckMoveResult[] parse, int[] currentMoves, LegalInfo info)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
if (!pkm.IsEgg)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
{
ParseMovesSketch(pkm, parse, currentMoves);
return;
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
// can only know sketch as egg
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
var levelup = new int[info.EvoChainsAllGens.Length][];
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
levelup[pkm.Format] = new[] { (int)Move.Sketch };
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
info.EncounterMoves = new ValidEncounterMoves(levelup);
Trailing commas No functional change
2021-08-20 20:49:20 +00:00
var source = new MoveParseSource { CurrentMoves = currentMoves };
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMoves(pkm, source, info, parse);
Refactoring mostly renaming things, includes a little bit of added sugar and splitting methods to simplify the codebase. all methods are now PascalCase
2017-06-18 01:37:19 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesWasEggPreRelearn(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> currentMoves, LegalInfo info, EncounterEgg e)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Don't generate move check order for an invalid ver
2018-06-09 03:36:52 +00:00
// Level up moves could not be inherited if Ditto is parent,
Simplify and enhance expression cache fetch of PersonalInfo (avoid re-fetching) If event, the pkm wasn't bred with ditto If genderless or male-only (except nido/volb), assume ditto
2019-04-17 06:07:47 +00:00
// that means genderless species and male only species (except Nidoran-M and Volbeat; they breed with Nidoran-F and Illumise) could not have level up moves as an egg
var pi = pkm.PersonalInfo;
Revise pichu volt tackle egg handling Clean up ParseMovesWasEggPreRelearn as only EncounterEgg call into it, and no EncounterEgg has a special moves interface implemented.
2021-09-16 01:43:18 +00:00
var AllowLevelUp = !pi.Genderless && !(pi.OnlyMale && Breeding.MixedGenderBreeding.Contains(e.Species));
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
int BaseLevel = AllowLevelUp ? 100 : e.LevelMin;
Extract EvoBase and MoveList from Core These are really old bloated parts of the original legality checking logic; separating for now to rewrite.
2020-06-21 00:44:05 +00:00
var LevelUp = MoveList.GetBaseEggMoves(pkm, e.Species, e.Form, e.Version, BaseLevel);
Misc tweaks 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
2017-11-18 06:19:23 +00:00
Revise pichu volt tackle egg handling Clean up ParseMovesWasEggPreRelearn as only EncounterEgg call into it, and no EncounterEgg has a special moves interface implemented.
2021-09-16 01:43:18 +00:00
var TradebackPreevo = pkm.Format == 2 && e.Species > 151;
Remove trailing whitespace
2018-05-12 15:13:39 +00:00
var NonTradebackLvlMoves = TradebackPreevo
Revise pichu volt tackle egg handling Clean up ParseMovesWasEggPreRelearn as only EncounterEgg call into it, and no EncounterEgg has a special moves interface implemented.
2021-09-16 01:43:18 +00:00
? MoveList.GetExclusivePreEvolutionMoves(pkm, e.Species, info.EvoChainsAllGens[2], 2, e.Version).Where(m => m > Legal.MaxMoveID_1).ToArray()
Remove unnecessary empty array allocations yay net 4.6 read more: http://justinvp.com/2015/07/20/array-empty/
2018-08-02 01:30:51 +00:00
: Array.Empty<int>();
Misc tweaks 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
2017-11-18 06:19:23 +00:00
Move more breeding logic to Breeding Simplify some method signatures, reduce checks for MoveEgg.GetEggMoves ran legality checks for 100,000 pkm, no more exceptions (Internal Error) -- added a "CanGameGenerateEggs" which filters out unused gameIDs
2021-01-01 23:01:22 +00:00
var Egg = MoveEgg.GetEggMoves(pkm.PersonalInfo, e.Species, e.Form, e.Version, e.Generation);
Flag gen2 egg moves when transferred from gen1 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).
2017-10-17 00:24:19 +00:00
if (info.Generation < 3 && pkm.Format >= 7 && pkm.VC1)
Replace more linq usage
2021-08-06 05:39:38 +00:00
Egg = Array.FindAll(Egg, m => m <= Legal.MaxMoveID_1);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Revise pichu volt tackle egg handling Clean up ParseMovesWasEggPreRelearn as only EncounterEgg call into it, and no EncounterEgg has a special moves interface implemented.
2021-09-16 01:43:18 +00:00
var specialMoves = e.CanHaveVoltTackle ? new[] { (int)Move.VoltTackle } : Array.Empty<int>(); // Volt Tackle for bred Pichu line
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
var source = new MoveParseSource
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
CurrentMoves = currentMoves,
Minor annotation tweaks no functional change
2021-01-13 06:26:08 +00:00
SpecialSource = specialMoves,
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
NonTradeBackLevelUpMoves = NonTradebackLvlMoves,
EggLevelUpSource = LevelUp,
EggMoveSource = Egg,
};
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMoves(pkm, source, info, parse);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesSketch(PKM pkm, CheckMoveResult[] parse, ReadOnlySpan<int> currentMoves)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = parse.Length - 1; i >= 0; i--)
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[i];
Disallow Thousand Arrows/Waves for Format6 Smeargle moves unavailable in that format, same as light of ruin Closes #3182 ty @DaWoblefet
2021-04-05 02:39:31 +00:00
var move = currentMoves[i];
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (move == 0)
r.Set(None, pkm.Format, Valid, LMoveSourceEmpty, CurrentMove);
else if (Legal.IsValidSketch(move, pkm.Format))
r.Set(Sketch, pkm.Format, Valid, L_AValid, CurrentMove);
else
r.Set(Unknown, pkm.Format, Invalid, LMoveSourceInvalidSketch, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMoves3DS(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> currentMoves, LegalInfo info)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Remove unnecessary empty array allocations yay net 4.6 read more: http://justinvp.com/2015/07/20/array-empty/
2018-08-02 01:30:51 +00:00
info.EncounterMoves.Relearn = info.Generation >= 6 ? pkm.RelearnMoves : Array.Empty<int>();
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (info.EncounterMatch is IMoveset)
ParseMovesSpecialMoveset(pkm, currentMoves, info, parse);
else
ParseMovesRelearn(pkm, currentMoves, info, parse);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesPre3DS(PKM pkm, CheckMoveResult[] parse, int[] currentMoves, LegalInfo info)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
if (info.EncounterMatch is EncounterEgg e)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (pkm.IsEgg)
VerifyRelearnMoves.VerifyEggMoveset(e, parse, currentMoves, CurrentMove);
else
ParseMovesWasEggPreRelearn(pkm, parse, currentMoves, info, e);
return;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Merge IGeneration into IEncounterable all but egg exposed it; now, just make egg expose it and remove the unnecessary interface we still need to Set generation for non-eggs/mgift, so have a separate Settable interface for internal purposes.
2020-05-17 19:32:28 +00:00
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
// Not all games have a re-learner. Initial moves may not fill out all 4 slots.
Merge IGeneration into IEncounterable all but egg exposed it; now, just make egg expose it and remove the unnecessary interface we still need to Set generation for non-eggs/mgift, so have a separate Settable interface for internal purposes.
2020-05-17 19:32:28 +00:00
int gen = info.EncounterMatch.Generation;
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
if (gen == 1 || (gen == 2 && !AllowGen2MoveReminder(pkm)))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesGenGB(pkm, currentMoves, info, parse);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesSpecialMoveset(pkm, currentMoves, info, parse);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesGenGB(PKM pkm, IReadOnlyList<int> currentMoves, LegalInfo info, CheckMoveResult[] parse)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
var enc = info.EncounterMatch;
Update PKHeX.Core abstractions with latest logic Co-Authored-By: Matt <17801814+sora10pls@users.noreply.github.com> Co-Authored-By: SciresM <8676005+SciresM@users.noreply.github.com> Co-Authored-By: Lusamine <30205550+Lusamine@users.noreply.github.com>
2022-02-05 01:35:15 +00:00
var evos = info.EvoChainsAllGens[enc.Generation];
Refactor EvoCriteria to be a struct, reduce allocation (#3483) * Make EvolutionCriteria struct 8 bytes per object instead of 26 Unify LevelMin/LevelMax to match EncounterTemplate bubble up precise array type for better iteration * Inline queue operations, less allocation * Inline some logic * Update EvolutionChain.cs * Improve clarity on duplicate move check * Search reverse For a dual stage chain, finds it first iteration rather than second.
2022-04-24 04:33:17 +00:00
var level = evos.Length > 0 ? evos[^1].LevelMin : enc.LevelMin;
Remove unnecessary empty array allocations yay net 4.6 read more: http://justinvp.com/2015/07/20/array-empty/
2018-08-02 01:30:51 +00:00
var InitialMoves = Array.Empty<int>();
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
var games = enc.Generation == 1 ? GBRestrictions.GetGen1Versions(enc) : GBRestrictions.GetGen2Versions(enc, pkm.Korean);
Refactoring remove some unneeded logic moves in gen1/2 can be reordered, sequence match should be intersect (full) contains match.
2018-06-19 02:57:32 +00:00
foreach (var ver in games)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Merge extended property patterns
2021-12-05 02:37:47 +00:00
var VerInitialMoves = enc is IMoveset {Moves.Count: not 0 } x ? (int[])x.Moves : MoveLevelUp.GetEncounterMoves(enc.Species, 0, level, ver);
Refactor EvoCriteria to be a struct, reduce allocation (#3483) * Make EvolutionCriteria struct 8 bytes per object instead of 26 Unify LevelMin/LevelMax to match EncounterTemplate bubble up precise array type for better iteration * Inline queue operations, less allocation * Inline some logic * Update EvolutionChain.cs * Improve clarity on duplicate move check * Search reverse For a dual stage chain, finds it first iteration rather than second.
2022-04-24 04:33:17 +00:00
if (VerInitialMoves.SequenceEqual(InitialMoves))
return; // Already checked this combination, and it wasn't valid. Don't bother repeating.
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
var source = new MoveParseSource
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
CurrentMoves = currentMoves,
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
Base = VerInitialMoves,
};
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMoves(pkm, source, info, parse);
Remove Tradeback state caching in PKM data Since we have more metadata with move learn sourcing, we can check if it was traded to gen2 to get new moves / deleted. Adjust call sites appropriately might have some issues, to be ironed out maybe
2021-08-20 22:59:54 +00:00
// Must have a minimum count of moves, depending on the tradeback state.
if (pkm is PK1 pk1)
{
int count = GBRestrictions.GetRequiredMoveCount(pk1, source.CurrentMoves, info, source.Base);
if (count == 1)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
return;
Remove Tradeback state caching in PKM data Since we have more metadata with move learn sourcing, we can check if it was traded to gen2 to get new moves / deleted. Adjust call sites appropriately might have some issues, to be ironed out maybe
2021-08-20 22:59:54 +00:00
Lessen severity of Gen1 missing moves if leveled up once Closes #3277
2021-10-30 04:25:48 +00:00
// Reverse for loop and break instead of 0..count continue -- early-breaks for the vast majority of cases.
// We already flag for empty interstitial moveslots.
for (int m = count - 1; m >= 0; m--)
Remove Tradeback state caching in PKM data Since we have more metadata with move learn sourcing, we can check if it was traded to gen2 to get new moves / deleted. Adjust call sites appropriately might have some issues, to be ironed out maybe
2021-08-20 22:59:54 +00:00
{
var move = source.CurrentMoves[m];
Lessen severity of Gen1 missing moves if leveled up once Closes #3277
2021-10-30 04:25:48 +00:00
if (move != 0)
break;
// There are ways to skip level up moves by leveling up more than once.
// https://bulbapedia.bulbagarden.net/wiki/List_of_glitches_(Generation_I)#Level-up_learnset_skipping
// Evolution canceling also leads to incorrect assumptions in the above used method, so just indicate them as fishy in that case.
// Not leveled up? Not possible to be missing the move slot.
var severity = enc.LevelMin == pkm.CurrentLevel ? Invalid : Fishy;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].Set(None, pkm.Format, severity, LMoveSourceEmpty, CurrentMove);
Remove Tradeback state caching in PKM data Since we have more metadata with move learn sourcing, we can check if it was traded to gen2 to get new moves / deleted. Adjust call sites appropriately might have some issues, to be ironed out maybe
2021-08-20 22:59:54 +00:00
}
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (Array.TrueForAll(parse, z => z.Valid))
return;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
InitialMoves = VerInitialMoves;
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesSpecialMoveset(PKM pkm, IReadOnlyList<int> currentMoves, LegalInfo info, CheckMoveResult[] parse)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
var source = new MoveParseSource
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
CurrentMoves = currentMoves,
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
SpecialSource = GetSpecialMoves(info.EncounterMatch),
};
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMoves(pkm, source, info, parse);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Change interface for method call to less restricted Don't need "Name" or "LongName" for these methods.
2021-07-14 00:33:23 +00:00
private static IReadOnlyList<int> GetSpecialMoves(IEncounterTemplate enc)
Add emerald swarm special move detection I don't think SlotType.Grass->Swarm changes anything, best to use it for differentiating vs the default 1% slot. Closes #1362
2017-07-26 00:35:07 +00:00
{
Minor clean
2021-01-11 02:15:33 +00:00
if (enc is IMoveset mg)
misc updates use arrays instead of list when capacity known use enums for const decoration remove unnecessary aliases
2018-09-02 02:55:08 +00:00
return mg.Moves;
Remove unnecessary empty array allocations yay net 4.6 read more: http://justinvp.com/2015/07/20/array-empty/
2018-08-02 01:30:51 +00:00
return Array.Empty<int>();
Add emerald swarm special move detection I don't think SlotType.Grass->Swarm changes anything, best to use it for differentiating vs the default 1% slot. Closes #1362
2017-07-26 00:35:07 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesRelearn(PKM pkm, IReadOnlyList<int> currentMoves, LegalInfo info, CheckMoveResult[] parse)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
var source = new MoveParseSource
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
CurrentMoves = currentMoves,
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
SpecialSource = GetSpecialMoves(info.EncounterMatch),
};
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Refactoring readability++
2017-09-04 20:48:10 +00:00
if (info.EncounterMatch is EncounterEgg e)
Move more breeding logic to Breeding Simplify some method signatures, reduce checks for MoveEgg.GetEggMoves ran legality checks for 100,000 pkm, no more exceptions (Internal Error) -- added a "CanGameGenerateEggs" which filters out unused gameIDs
2021-01-01 23:01:22 +00:00
source.EggMoveSource = MoveEgg.GetEggMoves(pkm.PersonalInfo, e.Species, e.Form, e.Version, e.Generation);
Flag gen2 egg moves when transferred from gen1 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).
2017-10-17 00:24:19 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMoves(pkm, source, info, parse);
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
var relearn = pkm.RelearnMoves;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = parse.Length - 1; i >= 0; i--)
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[i];
if (!r.IsRelearn && !pkm.IsEgg)
continue;
if (relearn.Contains(currentMoves[i]))
continue;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.FlagIllegal(string.Format(LMoveRelearnFMiss_0, r.Comment));
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMoves(PKM pkm, MoveParseSource source, LegalInfo info, CheckMoveResult[] parse)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Initial Isle of Armor support Co-Authored-By: sciresm <sciresm@users.noreply.github.com> Co-Authored-By: Matt <sora10pls@users.noreply.github.com> Co-Authored-By: Archit Date <architdate@gmail.com>
2020-06-19 23:51:15 +00:00
// Special considerations!
Replace more linq usage
2021-08-06 05:39:38 +00:00
const int NoMinGeneration = 0;
int minGeneration = NoMinGeneration;
Invert some "!is" to "is not"
2020-12-29 08:58:08 +00:00
if (pkm is IBattleVersion {BattleVersion: not 0} v)
Initial Isle of Armor support Co-Authored-By: sciresm <sciresm@users.noreply.github.com> Co-Authored-By: Matt <sora10pls@users.noreply.github.com> Co-Authored-By: Archit Date <architdate@gmail.com>
2020-06-19 23:51:15 +00:00
{
Replace more linq usage
2021-08-06 05:39:38 +00:00
minGeneration = ((GameVersion) v.BattleVersion).GetGeneration();
Move reset action to class
2020-10-04 19:08:16 +00:00
source.ResetSources();
Initial Isle of Armor support Co-Authored-By: sciresm <sciresm@users.noreply.github.com> Co-Authored-By: Matt <sora10pls@users.noreply.github.com> Co-Authored-By: Archit Date <architdate@gmail.com>
2020-06-19 23:51:15 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// Check empty moves and relearn moves before generation specific moves
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Replace more linq usage
2021-08-06 05:39:38 +00:00
var move = source.CurrentMoves[m];
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[m];
Replace more linq usage
2021-08-06 05:39:38 +00:00
if (move == 0)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(None, pkm.Format, Valid, LMoveSourceEmpty, CurrentMove);
Replace more linq usage
2021-08-06 05:39:38 +00:00
else if (minGeneration == NoMinGeneration && info.EncounterMoves.Relearn.Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Relearn, info.Generation, Valid, LMoveSourceRelearn, CurrentMove);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (Array.TrueForAll(parse, z => z.IsParsed))
return;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// Encapsulate arguments to simplify method calls
PKHeX.Core Nullable cleanup (#2401) * Handle some nullable cases Refactor MysteryGift into a second abstract class (backed by a byte array, or fake data) Make some classes have explicit constructors instead of { } initialization * Handle bits more obviously without null * Make SaveFile.BAK explicitly readonly again * merge constructor methods to have readonly fields * Inline some properties * More nullable handling * Rearrange box actions define straightforward classes to not have any null properties * Make extrabyte reference array immutable * Move tooltip creation to designer * Rearrange some logic to reduce nesting * Cache generated fonts * Split mystery gift album purpose * Handle more tooltips * Disallow null setters * Don't capture RNG object, only type enum * Unify learnset objects Now have readonly properties which are never null don't new() empty learnsets (>800 Learnset objects no longer created, total of 2400 objects since we also new() a move & level array) optimize g1/2 reader for early abort case * Access rewrite Initialize blocks in a separate object, and get via that object removes a couple hundred "might be null" warnings since blocks are now readonly getters some block references have been relocated, but interfaces should expose all that's needed put HoF6 controls in a groupbox, and disable * Readonly personal data * IVs non nullable for mystery gift * Explicitly initialize forced encounter moves * Make shadow objects readonly & non-null Put murkrow fix in binary data resource, instead of on startup * Assign dex form fetch on constructor Fixes legality parsing edge cases also handle cxd parse for valid; exit before exception is thrown in FrameGenerator * Remove unnecessary null checks * Keep empty value until init SetPouch sets the value to an actual one during load, but whatever * Readonly team lock data * Readonly locks Put locked encounters at bottom (favor unlocked) * Mail readonly data / offset Rearrange some call flow and pass defaults Add fake classes for SaveDataEditor mocking Always party size, no need to check twice in stat editor use a fake save file as initial data for savedata editor, and for gamedata (wow i found a usage) constrain eventwork editor to struct variable types (uint, int, etc), thus preventing null assignment errors
2019-10-17 01:47:31 +00:00
var moveInfo = new LearnInfo(pkm, source);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
// Check moves going backwards, marking the move valid in the most current generation when it can be learned
Extract generation traversal list fetch to static class
2021-04-17 20:09:09 +00:00
int[] generations = GenerationTraversal.GetVisitedGenerationOrder(pkm, info.EncounterOriginal.Generation);
Fix gen1 VC legality checking error 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
2017-07-18 23:21:31 +00:00
if (pkm.Format <= 2)
Replace more linq usage
2021-08-06 05:39:38 +00:00
generations = Array.FindAll(generations, z => z < info.EncounterMoves.LevelUpMoves.Length);
if (minGeneration != NoMinGeneration)
generations = Array.FindAll(generations, z => z >= minGeneration);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
Replace more linq usage
2021-08-06 05:39:38 +00:00
if (generations.Length != 0)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Replace more linq usage
2021-08-06 05:39:38 +00:00
int lastgen = generations[^1];
foreach (var gen in generations)
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesByGeneration(pkm, parse, gen, info, moveInfo, lastgen);
if (Array.TrueForAll(parse, z => z.IsParsed))
return;
Replace more linq usage
2021-08-06 05:39:38 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Use enum for species comparisons slightly easier to read with named values
2019-12-09 01:39:19 +00:00
if (pkm.Species == (int)Species.Shedinja && info.Generation <= 4)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseShedinjaEvolveMoves(pkm, parse, source.CurrentMoves, info.EvoChainsAllGens);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
foreach (var r in parse)
{
if (!r.IsParsed)
r.Set(Unknown, info.Generation, Invalid, LMoveSourceInvalid, CurrentMove);
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesByGeneration(PKM pkm, CheckMoveResult[] parse, int gen, LegalInfo info, LearnInfo learnInfo, int last)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
Span<bool> HMLearned = stackalloc bool[parse.Length];
bool KnowDefogWhirlpool = GetHMCompatibility(pkm, parse, gen, learnInfo.Source.CurrentMoves, HMLearned);
ParseMovesByGeneration(pkm, parse, gen, info, learnInfo);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
if (gen == last)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesByGenerationLast(parse, learnInfo, info.EncounterMatch);
New legallity checks (#1196) * 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
2017-06-07 03:10:05 +00:00
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
switch (gen)
{
Condense some switch expressions Should be it for now; exhaustively searched for all that made sense to change
2020-12-25 18:58:33 +00:00
case 1 or 2:
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseMovesByGeneration12(pkm, parse, learnInfo.Source.CurrentMoves, gen, info, learnInfo);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
break;
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
Condense some switch expressions Should be it for now; exhaustively searched for all that made sense to change
2020-12-25 18:58:33 +00:00
case 3 or 4:
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
if (pkm.Format > gen)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
FlagIncompatibleTransferHMs45(parse, learnInfo.Source.CurrentMoves, gen, HMLearned, KnowDefogWhirlpool);
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
break;
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesByGeneration(PKM pkm, CheckMoveResult[] parse, int gen, LegalInfo info, LearnInfo learnInfo)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
var moves = learnInfo.Source.CurrentMoves;
bool native = gen == pkm.Format;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[m];
if (IsCheckValid(r)) // already validated with another generation
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Misc tweaks (apostrophe & vc2 pre) 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.
2017-09-20 23:19:50 +00:00
int move = moves[m];
if (move == 0)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Fix 4 egg move gen2 oddish 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.
2018-04-01 22:04:53 +00:00
if (gen <= 2)
{
if (gen == 2 && !native && move > Legal.MaxMoveID_1 && pkm.VC1)
Fix broken gen2 tests add oddish to the mix
2018-04-01 22:51:55 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Unknown, gen, Invalid, LMoveSourceInvalid, CurrentMove);
Fix broken gen2 tests add oddish to the mix
2018-04-01 22:51:55 +00:00
continue;
}
if (gen == 2 && learnInfo.Source.EggMoveSource.Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(EggMove, gen, Valid, LMoveSourceEgg, CurrentMove);
Fix 4 egg move gen2 oddish 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.
2018-04-01 22:04:53 +00:00
else if (learnInfo.Source.Base.Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Initial, gen, Valid, native ? LMoveSourceDefault : string.Format(LMoveFDefault_0, gen), CurrentMove);
Fix 4 egg move gen2 oddish 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.
2018-04-01 22:04:53 +00:00
}
Fix broken gen2 tests add oddish to the mix
2018-04-01 22:51:55 +00:00
if (info.EncounterMoves.LevelUpMoves[gen].Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(LevelUp, gen, Valid, native ? LMoveSourceLevelUp : string.Format(LMoveFLevelUp_0, gen), CurrentMove);
Misc tweaks (apostrophe & vc2 pre) 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.
2017-09-20 23:19:50 +00:00
else if (info.EncounterMoves.TMHMMoves[gen].Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(TMHM, gen, Valid, native ? LMoveSourceTMHM : string.Format(LMoveFTMHM_0, gen), CurrentMove);
Misc tweaks (apostrophe & vc2 pre) 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.
2017-09-20 23:19:50 +00:00
else if (info.EncounterMoves.TutorMoves[gen].Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Tutor, gen, Valid, native ? LMoveSourceTutor : string.Format(LMoveFTutor_0, gen), CurrentMove);
Misc tweaks (apostrophe & vc2 pre) 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.
2017-09-20 23:19:50 +00:00
else if (gen == info.Generation && learnInfo.Source.SpecialSource.Contains(move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Special, gen, Valid, LMoveSourceSpecial, CurrentMove);
Fix shared egg moves for non-galar origin Use the private method; we already know the gen and that the egg moves are permitted Relocate method to MoveEgg
2020-02-17 03:47:57 +00:00
else if (gen >= 8 && MoveEgg.GetIsSharedEggMove(pkm, gen, move))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(Shared, gen, Valid, native ? LMoveSourceShared : string.Format(LMoveSourceSharedF, gen), CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (gen >= 3 || !IsCheckValid(r))
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Minor simplification if conditional checks >=3, and all subsequent checks require Valid; add this to the continue condition, and remove unnecessary comparisons update some comments
2019-03-17 03:07:30 +00:00
// Gen1/Gen2 only below
if (gen == 2 && learnInfo.Source.NonTradeBackLevelUpMoves.Contains(m))
Misc clean split some methods with optional parameters=null add more xmldoc replace some magic numbers -> enum/const references consolidate common array operations (span soon maybe?)
2019-09-10 07:21:51 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.Gen2PreevoMoves.Add(m);
Misc clean split some methods with optional parameters=null add more xmldoc replace some magic numbers -> enum/const references consolidate common array operations (span soon maybe?)
2019-09-10 07:21:51 +00:00
}
Minor simplification if conditional checks >=3, and all subsequent checks require Valid; add this to the continue condition, and remove unnecessary comparisons update some comments
2019-03-17 03:07:30 +00:00
else if (gen == 1)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.Gen1Moves.Add(m);
Refactoring de-linq some areas where direct accessing is possible (list/array)
2017-12-05 04:16:54 +00:00
if (learnInfo.Gen2PreevoMoves.Count != 0)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.MixedGen12NonTradeback = true;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesByGeneration12(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> currentMoves, int gen, LegalInfo info, LearnInfo learnInfo)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
// Mark the gen 1 exclusive moves as illegal because the pokemon also have Non tradeback egg moves.
if (learnInfo.MixedGen12NonTradeback)
{
foreach (int m in learnInfo.Gen1Moves)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LG1MoveExclusive, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
foreach (int m in learnInfo.Gen2PreevoMoves)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LG1TradebackPreEvoMove, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
New legallity checks (#1196) * 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
2017-06-07 03:10:05 +00:00
Remove Tradeback state caching in PKM data Since we have more metadata with move learn sourcing, we can check if it was traded to gen2 to get new moves / deleted. Adjust call sites appropriately might have some issues, to be ironed out maybe
2021-08-20 22:59:54 +00:00
if (gen == 1 && pkm.Format == 1 && !AllowGen1Tradeback)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseRedYellowIncompatibleMoves(pkm, parse, currentMoves);
ParseEvolutionsIncompatibleMoves(pkm, parse, currentMoves, info.EncounterMoves.TMHMMoves[1]);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseMovesByGenerationLast(CheckMoveResult[] parse, LearnInfo learnInfo, IEncounterTemplate enc)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Remove gen1 flagged moves if they can originate from egg Revises the tradeback status to remove any index that could have been sourced from egg moves instead. really should refactor this TradebackStatus away, but Move validation still remains one of the un-refined parts of the legality checks :P
2020-10-04 19:21:05 +00:00
int gen = enc.Generation;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
ParseEggMovesInherited(parse, gen, learnInfo);
ParseEggMoves(parse, gen, learnInfo);
ParseEggMovesRemaining(parse, learnInfo, enc);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseEggMovesInherited(CheckMoveResult[] parse, int gen, LearnInfo learnInfo)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
var moves = learnInfo.Source.CurrentMoves;
// Check higher-level moves after all the moves but just before egg moves to differentiate it from normal level up moves
// Also check if the base egg moves is a non tradeback move
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[m];
Remove gen1 flagged moves if they can originate from egg Revises the tradeback status to remove any index that could have been sourced from egg moves instead. really should refactor this TradebackStatus away, but Move validation still remains one of the un-refined parts of the legality checks :P
2020-10-04 19:21:05 +00:00
if (IsCheckValid(r)) // already validated
{
if (gen == 2 && r.Generation != 1)
continue;
}
int move = moves[m];
if (move == 0)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Remove gen1 flagged moves if they can originate from egg Revises the tradeback status to remove any index that could have been sourced from egg moves instead. really should refactor this TradebackStatus away, but Move validation still remains one of the un-refined parts of the legality checks :P
2020-10-04 19:21:05 +00:00
if (!learnInfo.Source.EggLevelUpSource.Contains(move)) // Check if contains level-up egg moves from parents
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Remove gen1 flagged moves if they can originate from egg Revises the tradeback status to remove any index that could have been sourced from egg moves instead. really should refactor this TradebackStatus away, but Move validation still remains one of the un-refined parts of the legality checks :P
2020-10-04 19:21:05 +00:00
if (learnInfo.IsGen2Pkm && learnInfo.Gen1Moves.Count != 0 && move > Legal.MaxMoveID_1)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(InheritLevelUp, gen, Invalid, LG1MoveTradeback, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.MixedGen12NonTradeback = true;
New legallity checks (#1196) * 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
2017-06-07 03:10:05 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
else
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(InheritLevelUp, gen, Valid, LMoveEggLevelUp, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.LevelUpEggMoves.Add(m);
Remove gen1 flagged moves if they can originate from egg Revises the tradeback status to remove any index that could have been sourced from egg moves instead. really should refactor this TradebackStatus away, but Move validation still remains one of the un-refined parts of the legality checks :P
2020-10-04 19:21:05 +00:00
if (gen == 2 && learnInfo.Gen1Moves.Contains(m))
learnInfo.Gen1Moves.Remove(m);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseEggMoves(CheckMoveResult[] parse, int gen, LearnInfo learnInfo)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
var moves = learnInfo.Source.CurrentMoves;
// Check egg moves after all the generations and all the moves, every move that can't be learned in another source should have preference
// the moves that can only be learned from egg moves should in the future check if the move combinations can be breed in gens 2 to 5
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[m];
if (IsCheckValid(r))
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
int move = moves[m];
if (move == 0)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
bool wasEggMove = learnInfo.Source.EggMoveSource.Contains(move);
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
if (wasEggMove)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// To learn exclusive generation 1 moves the pokemon was tradeback, but it can't be trade to generation 1
// without removing moves above MaxMoveID_1, egg moves above MaxMoveID_1 and gen 1 moves are incompatible
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
if (learnInfo.IsGen2Pkm && learnInfo.Gen1Moves.Count != 0 && move > Legal.MaxMoveID_1)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(EggMove, gen, Invalid, LG1MoveTradeback, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.MixedGen12NonTradeback = true;
}
else
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(EggMove, gen, Valid, LMoveSourceEgg, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.EggMovesLearned.Add(m);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
if (!learnInfo.Source.EggEventSource.Contains(move))
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
continue;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
if (!wasEggMove)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Misc updates extract some enums to core, use Enum.GetNames to do string fetch make stamp listbox taller clean up Display*ID setters
2019-02-22 05:54:41 +00:00
if (learnInfo.IsGen2Pkm && learnInfo.Gen1Moves.Count != 0 && move > Legal.MaxMoveID_1)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(SpecialEgg, gen, Invalid, LG1MoveTradeback, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.MixedGen12NonTradeback = true;
}
else
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(SpecialEgg, gen, Valid, LMoveSourceEggEvent, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
learnInfo.EventEggMoves.Add(m);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseEggMovesRemaining(CheckMoveResult[] parse, LearnInfo learnInfo, IEncounterTemplate enc)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
// A pokemon could have normal egg moves and regular egg moves
// Only if all regular egg moves are event egg moves or all event egg moves are regular egg moves
Replace more linq usage
2021-08-06 05:39:38 +00:00
var RegularEggMovesLearned = learnInfo.EggMovesLearned.FindAll(learnInfo.LevelUpEggMoves.Contains);
Refactoring de-linq some areas where direct accessing is possible (list/array)
2017-12-05 04:16:54 +00:00
if (RegularEggMovesLearned.Count != 0 && learnInfo.EventEggMoves.Count != 0)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// Moves that are egg moves or event egg moves but not both
minor clean
2021-04-19 01:18:09 +00:00
var IncompatibleEggMoves = RegularEggMovesLearned.Except(learnInfo.EventEggMoves).Union(learnInfo.EventEggMoves.Except(RegularEggMovesLearned));
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
foreach (int m in IncompatibleEggMoves)
{
minor clean
2021-04-19 01:18:09 +00:00
bool isEvent = learnInfo.EventEggMoves.Contains(m);
if (isEvent)
{
if (!learnInfo.EggMovesLearned.Contains(m))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LMoveEggIncompatibleEvent, CurrentMove);
minor clean
2021-04-19 01:18:09 +00:00
}
else
{
if (learnInfo.EggMovesLearned.Contains(m))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LMoveEggIncompatible, CurrentMove);
minor clean
2021-04-19 01:18:09 +00:00
else if (learnInfo.LevelUpEggMoves.Contains(m))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LMoveEventEggLevelUp, CurrentMove);
minor clean
2021-04-19 01:18:09 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Invert some "!is" to "is not"
2020-12-29 08:58:08 +00:00
else if (enc is not EncounterEgg)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Minor simplification if conditional checks >=3, and all subsequent checks require Valid; add this to the continue condition, and remove unnecessary comparisons update some comments
2019-03-17 03:07:30 +00:00
// Event eggs cannot inherit moves from parents; they are not bred.
Track a PKM's Box,Slot,StorageFlags,Identifier metadata separately (#3222) * Track a PKM's Box,Slot,StorageFlags,Identifier metadata separately Don't store within the object, track the slot origin data separately. Batch editing now pre-filters if using Box/Slot/Identifier logic; split up mods/filters as they're starting to get pretty hefty. - Requesting a Box Data report now shows all slots in the save file (party, misc) - Can now exclude backup saves from database search via toggle (separate from settings preventing load entirely) - Replace some linq usages with direct code * Remove WasLink virtual in PKM Inline any logic, since we now have encounter objects to indicate matching, rather than the proto-legality logic checking properties of a PKM. * Use Fateful to directly check gen5 mysterygift origins No other encounter types in gen5 apply Fateful * Simplify double ball comparison Used to be separate for deferral cases, now no longer needed to be separate. * Grab move/relearn reference and update locally Fix relearn move identifier * Inline defog HM transfer preference check HasMove is faster than getting moves & checking contains. Skips allocation by setting values directly. * Extract more met location metadata checks: WasBredEgg * Replace Console.Write* with Debug.Write* There's no console output UI, so don't include them in release builds. * Inline WasGiftEgg, WasEvent, and WasEventEgg logic Adios legality tags that aren't entirely correct for the specific format. Just put the computations in EncounterFinder.
2021-06-23 03:23:48 +00:00
var gift = enc is EncounterStatic {Gift: true}; // otherwise, EncounterInvalid
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
foreach (int m in RegularEggMovesLearned)
{
if (learnInfo.EggMovesLearned.Contains(m))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(gift ? LMoveEggMoveGift : LMoveEggInvalidEvent, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
else if (learnInfo.LevelUpEggMoves.Contains(m))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(gift ? LMoveEggInvalidEventLevelUpGift : LMoveEggInvalidEventLevelUp, CurrentMove);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseRedYellowIncompatibleMoves(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> currentMoves)
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
var incompatible = GetIncompatibleRBYMoves(pkm, currentMoves);
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
if (incompatible.Count == 0)
return;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
if (incompatible.Contains(currentMoves[m]))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(LG1MoveLearnSameLevel, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
}
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
private static IList<int> GetIncompatibleRBYMoves(PKM pkm, IReadOnlyList<int> currentMoves)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Minor simplification if conditional checks >=3, and all subsequent checks require Valid; add this to the continue condition, and remove unnecessary comparisons update some comments
2019-03-17 03:07:30 +00:00
// Check moves that are learned at the same level in Red/Blue and Yellow, these are illegal because there is no Move Reminder in Gen1.
// There are only two incompatibilities for Gen1; there are no illegal combination in Gen2.
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
switch (pkm.Species)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
// Vaporeon in Yellow learns Mist and Haze at level 42, Mist can only be learned if it leveled up in the daycare
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// Vaporeon in Red/Blue learns Acid Armor at level 42 and level 47 in Yellow
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
case (int)Species.Vaporeon when pkm.CurrentLevel < 47 && currentMoves.Contains((int)Move.AcidArmor):
Reduce allocations for gen1 move parse use shared interfaces
2018-10-07 22:36:32 +00:00
{
var incompatible = new List<int>(3);
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (currentMoves.Contains((int)Move.Mist)) incompatible.Add((int)Move.Mist);
if (currentMoves.Contains((int)Move.Haze)) incompatible.Add((int)Move.Haze);
Refactoring de-linq some areas where direct accessing is possible (list/array)
2017-12-05 04:16:54 +00:00
if (incompatible.Count != 0)
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
incompatible.Add((int)Move.AcidArmor);
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
else
return Array.Empty<int>();
Reduce allocations for gen1 move parse use shared interfaces
2018-10-07 22:36:32 +00:00
return incompatible;
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// Flareon in Yellow learns Smog at level 42
// Flareon in Red Blue learns Leer at level 42 and level 47 in Yellow
Minor clean Use some enums, save a few virtual/static fetches
2021-01-16 20:01:40 +00:00
case (int)Species.Flareon when pkm.CurrentLevel < 47 && currentMoves.Contains((int)Move.Leer) && currentMoves.Contains((int)Move.Smog):
return new[] { (int)Move.Leer, (int)Move.Smog };
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
Minor simplification if conditional checks >=3, and all subsequent checks require Valid; add this to the continue condition, and remove unnecessary comparisons update some comments
2019-03-17 03:07:30 +00:00
default: return Array.Empty<int>();
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Move relearn required flag to checkmoveresult reduce usage for non gen6+ relearn move cases save contains result for later usage instead of recomputing the inverse
2018-07-02 04:34:17 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void ParseEvolutionsIncompatibleMoves(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> moves, IReadOnlyList<int> tmhm)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Extract gb specific core logic to separate class
2018-10-10 04:07:13 +00:00
GBRestrictions.GetIncompatibleEvolutionMoves(pkm, moves, tmhm,
out var prevSpeciesID,
out var incompatPrev,
out var incompatCurr);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Extract gb specific core logic to separate class
2018-10-10 04:07:13 +00:00
if (prevSpeciesID == 0)
return;
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Extract gb specific core logic to separate class
2018-10-10 04:07:13 +00:00
var prev = SpeciesStrings[prevSpeciesID];
var curr = SpeciesStrings[pkm.Species];
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Reduce allocations for gen1 move parse use shared interfaces
2018-10-07 22:36:32 +00:00
{
Extract gb specific core logic to separate class
2018-10-10 04:07:13 +00:00
if (incompatCurr.Contains(moves[m]))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(string.Format(LMoveEvoFLower, curr, prev), CurrentMove);
Extract gb specific core logic to separate class
2018-10-10 04:07:13 +00:00
if (incompatPrev.Contains(moves[m]))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[m].FlagIllegal(string.Format(LMoveEvoFHigher, curr, prev), CurrentMove);
Reduce allocations for gen1 move parse use shared interfaces
2018-10-07 22:36:32 +00:00
}
}
Eliminate boxing on encounter search (criteria) struct implementing interface is boxed when passed to method that accepts interface (not generic method). Removes IDexLevel (no other inheritors but EvoCriteria) and uses the primitive the data is stored (array, not IReadOnlyList) for slightly better perf.
2022-05-08 01:29:36 +00:00
private static void ParseShedinjaEvolveMoves(PKM pkm, CheckMoveResult[] parse, IReadOnlyList<int> currentMoves, EvoCriteria[][] evos)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
int shedinjaEvoMoveIndex = 0;
More shedinja edge cases Don't allow shedinja evolution moves if it wasn't evolved in that generation
2020-04-12 18:28:03 +00:00
var format = pkm.Format;
for (int gen = Math.Min(format, 4); gen >= 3; gen--)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Eliminate boxing on encounter search (criteria) struct implementing interface is boxed when passed to method that accepts interface (not generic method). Removes IDexLevel (no other inheritors but EvoCriteria) and uses the primitive the data is stored (array, not IReadOnlyList) for slightly better perf.
2022-05-08 01:29:36 +00:00
if (evos[gen].Length != 2)
More shedinja edge cases Don't allow shedinja evolution moves if it wasn't evolved in that generation
2020-04-12 18:28:03 +00:00
continue; // Was not evolved in this generation
Allow Ninjask moves on Gen 4 Sport Ball Shedinja
2022-01-03 03:12:35 +00:00
if (gen == 4 && pkm.Ball != 4 && !(pkm.Ball == (int)Ball.Sport && pkm.HGSS))
More shedinja edge cases Don't allow shedinja evolution moves if it wasn't evolved in that generation
2020-04-12 18:28:03 +00:00
continue; // Was definitively evolved in Gen3
Handle g4 shedinja learn case too change 291/292 to Shedinja/Ninjask for easy enum reference check
2020-01-12 22:34:29 +00:00
var maxLevel = pkm.CurrentLevel;
Extract EvoBase and MoveList from Core These are really old bloated parts of the original legality checking logic; separating for now to rewrite.
2020-06-21 00:44:05 +00:00
var ninjaskMoves = MoveList.GetShedinjaEvolveMoves(pkm, gen, maxLevel);
More shedinja edge cases Don't allow shedinja evolution moves if it wasn't evolved in that generation
2020-04-12 18:28:03 +00:00
bool native = gen == format;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int m = parse.Length - 1; m >= 0; m--)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[m];
if (IsCheckValid(r)) // already validated
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
continue;
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
if (!ninjaskMoves.Contains(currentMoves[m]))
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
continue;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
// Can't have more than one Ninjask exclusive move on Shedinja
if (shedinjaEvoMoveIndex != 0)
{
r.FlagIllegal(LMoveNincada, CurrentMove);
continue;
}
Reduce allocations for gen1 move parse use shared interfaces
2018-10-07 22:36:32 +00:00
var msg = native ? LMoveNincadaEvo : string.Format(LMoveNincadaEvoF_0, gen);
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.Set(ShedinjaEvo, gen, Valid, msg, CurrentMove);
shedinjaEvoMoveIndex = m;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (shedinjaEvoMoveIndex == 0)
Refactoring where!contains -> except relocate special tutor
2018-06-09 23:04:06 +00:00
return;
Update shedinja ninjask move checks check level of source moves for any incompatibility in non-bred cases. Closes #1805
2018-02-02 03:57:02 +00:00
// Double check that the Ninjask move level isn't less than any Nincada move level
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
var r = parse[shedinjaEvoMoveIndex];
if (r.Source != LevelUp)
return;
var move = currentMoves[shedinjaEvoMoveIndex];
int levelS = MoveList.GetShedinjaMoveLevel((int)Species.Shedinja, move, r.Generation);
Update shedinja ninjask move checks check level of source moves for any incompatibility in non-bred cases. Closes #1805
2018-02-02 03:57:02 +00:00
if (levelS > 0)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
return;
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
int levelN = MoveList.GetShedinjaMoveLevel((int)Species.Nincada, move, r.Generation);
int levelJ = MoveList.GetShedinjaMoveLevel((int)Species.Ninjask, move, r.Generation);
Update shedinja ninjask move checks check level of source moves for any incompatibility in non-bred cases. Closes #1805
2018-02-02 03:57:02 +00:00
if (levelN > levelJ)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
r.FlagIllegal(string.Format(LMoveEvoFHigher, SpeciesStrings[(int)Species.Nincada], SpeciesStrings[(int)Species.Ninjask]), CurrentMove);
Update shedinja ninjask move checks check level of source moves for any incompatibility in non-bred cases. Closes #1805
2018-02-02 03:57:02 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static bool GetHMCompatibility(PKM pkm, IReadOnlyList<CheckMoveResult> parse, int gen, IReadOnlyList<int> moves, Span<bool> HMLearned)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
// Check if pokemon knows HM moves from generation 3 and 4 but are not valid yet, that means it cant learn the HMs in future generations
if (gen == 4 && pkm.Format > 4)
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
FlagIsHMSource(HMLearned, Legal.HM_4_RemovePokeTransfer);
return moves.Where((m, i) => IsDefogWhirl(m) && IsCheckInvalid(parse[i])).Count() == 2;
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
if (gen == 3 && pkm.Format > 3)
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
FlagIsHMSource(HMLearned, Legal.HM_3);
return false;
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
void FlagIsHMSource(Span<bool> flags, ICollection<int> source)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = parse.Count - 1; i >= 0; i--)
flags[i] = IsCheckInvalid(parse[i]) && source.Contains(moves[i]);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
}
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Fix gen4 hm indexing
2021-01-26 04:58:56 +00:00
private static bool IsDefogWhirl(int move) => move is (int)Move.Defog or (int)Move.Whirlpool;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static int GetDefogWhirlCount(IReadOnlyList<CheckMoveResult> parse, IReadOnlyList<int> moves)
{
int ctr = 0;
for (int i = moves.Count - 1; i >= 0; i--)
{
if (!IsDefogWhirl(moves[i]))
continue;
var r = parse[i];
if (!r.Valid || r.Generation >= 5)
continue;
ctr++;
}
return ctr;
}
private static bool IsCheckInvalid(CheckMoveResult chk) => chk.IsParsed && !chk.Valid;
private static bool IsCheckValid(CheckMoveResult chk) => chk.IsParsed && chk.Valid;
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void FlagIncompatibleTransferHMs45(IReadOnlyList<CheckMoveResult> parse, IReadOnlyList<int> currentMoves, int gen, ReadOnlySpan<bool> HMLearned, bool KnowDefogWhirlpool)
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
{
Don't generate move check order for an invalid ver
2018-06-09 03:36:52 +00:00
// After all the moves from the generations 3 and 4,
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// including egg moves if is the origin generation because some hidden moves are also special egg moves in gen 3
Don't generate move check order for an invalid ver
2018-06-09 03:36:52 +00:00
// Check if the marked hidden moves that were invalid at the start are now marked as valid, that means
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
// the hidden move was learned in gen 3 or 4 but was not removed when transfer to 4 or 5
if (KnowDefogWhirlpool)
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
int invalidCount = GetDefogWhirlCount(parse, currentMoves);
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
if (invalidCount == 2) // can't know both at the same time
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = parse.Count - 1; i >= 0; i--) // flag both moves
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Minor renaming/type change of passed args Next commit will have more changes where we remove some reliance on PKM.GenNumber, in favor of the passed encounter's generation
2020-05-31 19:12:07 +00:00
if (IsDefogWhirl(currentMoves[i]))
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[i].FlagIllegal(LTransferMoveG4HM, CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
// Flag moves that are only legal when learned from a past-gen HM source
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = HMLearned.Length - 1; i >= 0; i--)
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
if (HMLearned[i] && IsCheckValid(parse[i]))
parse[i].FlagIllegal(string.Format(LTransferMoveHM, gen, gen + 1), CurrentMove);
Misc style & minor tweaks Remove move combobox flicker hack (no longer necessary) Add more Array.Empty usages cache mysterygift sizes seal some classes no functionality changes
2018-08-03 03:11:42 +00:00
}
Refactoring split up current move parsing a little
2017-09-02 06:15:57 +00:00
}
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void VerifyNoEmptyDuplicates(ReadOnlySpan<int> moves, ReadOnlySpan<CheckMoveResult> parse)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
bool emptySlot = false;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = 0; i < parse.Length; i++)
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
{
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
var move = moves[i];
if (move == 0)
{
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
emptySlot = true;
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
continue;
}
// If an empty slot was noted for a prior move, flag the empty slots.
if (emptySlot)
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
FlagEmptySlotsBeforeIndex(moves, parse, i);
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
emptySlot = false;
continue;
}
// Check for same move in next move slots
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
FlagDuplicateMovesAfterIndex(moves, parse, i, move);
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
}
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void FlagDuplicateMovesAfterIndex(ReadOnlySpan<int> moves, ReadOnlySpan<CheckMoveResult> parse, int index, int move)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
{
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
for (int i = parse.Length - 1; i > index; i--)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
{
Minor tweaks Pass evo chain to memory learning machine move Pass minimum level for GB encounters if present, rather than enc min level
2021-07-23 02:08:56 +00:00
if (moves[i] != move)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
continue;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[index].FlagIllegal(LMoveSourceEmpty);
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
return;
}
}
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
private static void FlagEmptySlotsBeforeIndex(ReadOnlySpan<int> moves, ReadOnlySpan<CheckMoveResult> parse, int index)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
{
Minor tweaks Pass evo chain to memory learning machine move Pass minimum level for GB encounters if present, rather than enc min level
2021-07-23 02:08:56 +00:00
for (int i = index - 1; i >= 0; i--)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
{
Minor tweaks Pass evo chain to memory learning machine move Pass minimum level for GB encounters if present, rather than enc min level
2021-07-23 02:08:56 +00:00
if (moves[i] != 0)
Remove usage of linq for duplicate count Check successive move slots after the current index rather than doing a count eq >= 2. Flag all empty slots prior to a filled move slot, rather than flagging a full slot as invalid because "empty move".
2020-08-30 22:43:29 +00:00
return;
Performance: Slightly reduce allocations in moveset validation (#3460) * Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object. * Ensures the objects are never-null, and makes cleanup easier. Slightly adjusts some other parts of the moveset validation to reduce allocations.
2022-03-13 01:06:03 +00:00
parse[i].FlagIllegal(LMoveSourceEmpty);
Refactor encounter matching 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 :smile:
2017-05-28 04:17:53 +00:00
}
}
}
}
Reference in a new issue Copy permalink