We implement simple state machine iterators to iterate through every split type encounter array, and more finely control the path we iterate through. And, by using generics, we can have the compiler generate optimized code to avoid virtual calls.
In addition to this, we shift away from the big-5 encounter types and not inherit from an abstract class. This allows for creating a PK* of a specific type and directly writing properties (no virtual calls). Plus we can now fine-tune each encounter type to call specific code, and not have to worry about future game encounter types bothering the generation routines.
Relocate checksum adders to Checksums class, improve performance by eliminating slice calls
Improve HOME sharing for GO8 and IMoveset encounters: actually sanity check GO8, and skip other non-PK7/PB7 cases.
Instead of looping, if the moveset is full and a new move is added, the game shifts all arr[1..] down one slot then adds the move at the end.
Since we don't need to keep track of PP/PP Ups, we can just defer the shifting and do n % 4 rotations at the end instead of n rotations (one on each move added).
## Issue
We want to discard-but-remember any slots that aren't a perfect fit, on the off chance that a better one exists later in the search space. If there's no better match, then we gotta go with what we got.
## Example:
Wurmple exists in area `X`, and also has a more rare slot for Silcoon, with the same level for both slots.
* We have a Silcoon that we've leveled up a few times.
Was our Silcoon originally a Wurmple, or was it caught as a Silcoon?
* To be sure, we have to check the EC/PID if the Wurmple wouldn't evolve into Cascoon instead.
* We don't want to wholly reject that Wurmple slot, as maybe the Met Level isn't within Silcoon's slot range.
---
Existing implementation would store "deferred" matches in a list; we only need to keep 1 of these matches around (less allocation!). We also want to differentiate between a "good" deferral and a "bad" deferral; I don't think this is necessary but it's currently used by Mystery Gift matching (implemented for the Eeveelution mystery gifts which matter for evolution moves).
The existing logic didn't use inheritance, and instead had static methods being reused across generations. Quite kludgy. Also, the existing logic was a pain to modify the master encounter yield methods, as one generation's quirks had to not impact all other generations that used the method.
---
The new implementation splits out the encounter yielding methods to be separate for each generation / subset. Now, things don't have to check `WasLink` for Gen7 origin, because Pokémon Link wasn't a thing in Gen7.
---
## Future
Maybe refactoring yielders into "GameCores" that expose yielding behaviors / properties, rather than the static logic. As more generations and side-gamegroups get added (thanks LGPE/GO/GameCube), all this switch stuff gets annoying to maintain instead of just overriding/inheritance.
## Conclusion
This shouldn't impact any legality results negatively; if you notice any regressions, report them! This should reduce false flags where we didn't defer-discard an encounter when we should have (wild area mons being confused with raids).
* Rework gen1 slot loading
Slot templates are precomputed from ROM data and just loaded straight in, with tight coupling to the encounter area (grouped by slot types).
* Revise fuzzy met check for underleveled wild evos
Example: Level 23 poliwhirl in RBY as a level 50 poliwhirl, will assume the chain is 25-50 for poliwhirl (as poliwag evolves at 25). Instead of revising the origin chain, just ignore the evo min level in the comparison.
Previous commit fixed it for gen1.
* Rework gen2-4 slot loading
Gen4 not finished, Type Encounter data and some edge encounters not recognizing yet...
* Add feebas slots for old/good encounters
* Begin moving properties
Great news! Gen5-7 need to be de-dumbed like Gen1-4.
Then I can remove the bang (!) on the Area accessor and ensure that it's never null!
* Split off XD pokespot slot encounter table type
* Set area in constructor
* Deduplicate g3 roaming encounters
* Deduplicate xd encounter locations (rebattle)
Only difference is met location; no need to create 500 extra encounter objects. A simple contains check is ok (rarely in gen3 format).
* Make all slots have a readonly reference to their parent area
* Minor clean
* Remove "Safari" slot type flag
Can be determined via other means (generation-location), allows us to reduce the size of SlotType member to a byte
Output of slot binaries didn't preserve the Safari flag anyway.
* Update SlotType.cs
* Handle type encounters correctly
* Merge safari area into regular xy area
* Merge dexnav accessor logic
* fix some logic so that tests pass again
rearrange g5 dw init to be done outside of static constructor (initializer instead)
PIDGenerator: friend safari slots now generate with required flawless IV count
* Add cianwood tentacool gift encounter
* Remove unnecessary abstractions
Fake area just returned a slot; since Slots have a non-null reference to the area, we can just return the slot and use the API to grab a list of possible slots for the chain.
Increase restrictiveness of location/type get-set operations
* Minor tweaks, pass parameters
DexNav observed state isn't necessary to use, only need to see if it's possible to dexnav. Now that we have metadata for slots, we can.
* Remove unused legality tables
remove linq for characteristic calc (get max IV) and others
remove some reliance on new[] for iv/ev
pk5 replace bitmagic with property get->set copy for cleanliness
Starter Pikachu & Eevee have different growth rates than their base
forms (seriously WHY?)
remove old api surface in PKX as a breaking change as adding the
parameter is necessary.
V### names weren't enjoyable to work with; use similar verbose style as
the program message strings.
updating the translation files with the remapped variable names shortly
remap list: https://pastebin.com/jybkVDAK
Checks.cs initially started out small, but over the years it has grown
to handle multiple types of checks. With all these checks next to
eachother, it's hard to see the overall groups. Splitting them up
(potentially further?) allows for more focused maintenance &
understanding.
Not sure if I'm happy with the overall bandaids used (checks no longer
done within LegalityAnalysis so variable repointing is excessively
used), but I'm happier the way it is now compared to the huge Checks.cs
