Existing `get`/`set` logic is flawed in that it doesn't work on Big Endian operating systems, and it allocates heap objects when it doesn't need to.
`System.Buffers.Binary.BinaryPrimitives` in the `System.Memory` NuGet package provides both Little Endian and Big Endian methods to read and write data; all the `get`/`set` operations have been reworked to use this new API. This removes the need for PKHeX's manual `BigEndian` class, as all functions are already covered by the BinaryPrimitives API.
The `StringConverter` has now been rewritten to accept a Span to read from & write to, no longer requiring a temporary StringBuilder.
Other Fixes included:
- The Super Training UI for Gen6 has been reworked according to the latest block structure additions.
- Cloning a Stadium2 Save File now works correctly (opening from the Folder browser list).
- Checksum & Sanity properties removed from parent PKM class, and is now implemented via interface.
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
* 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.
AltForm & Form & Forme => Form
GenNumber & Generation => Generation
Extract out SpeciesForm interface, and re-add IGeneration
For those using PKHeX as a dependency, this should be a pretty straightforward manual replacement... GenNumber and AltForm should be quick find-replace`s.
Introducing a new PKM format: SK2
Split ICaughtData2 off of PK2 so it can be shared with SK2 when type-checks occur
Add conversion for PK2<->SK2
Split the split-buffer handling for GBPKM to GBPKML (what a name), so that I can reuse shared accessor logic for SK2.
Move logic closer to where it is used rather than in larger static classes
EncounterStatic(7): move VC transfer template creation to class, simplify some sanity checks
EvoChain: g==2 case is never hit as the generation check at the top of the loop already skips
Results in a little more code, but each path is less tangled
simplify some expressions
remove RBDragonair content in favor of a strict filter for catch rate
Move gen1 trade trainer names to stringconverter, since pk1/pk2 shouldn't refer to legality classes
All logic in PokeCrypto is separate from the rest of the PKHeX.Core
library; makes it easy to just rip this portion out and reuse in other
projects without needing the entirety of PKHeX.Core logic
optimize out the CheckEncrypted to the actual path, separate methods.
Only usages of this method were with hardcoded Format values, so no
impact
Extract Species Name logic to SpeciesName
Extract Language logic to Language
Remove FormConverter wrapper for string[] fetch
Rearrange some logic to more appropriate locations, update access
modifiers / types
Move some pkm array methods to arrayutil, make generic
PKX.GetVCLanguage was a dupe of _K12.GuessedLanguage() so just expose
the method
PKX is now back to pkm data manip only
reduces loading time (don't have to allocate conversion arrays when
launching a gen7 game), and separates things to easier to manage
locations
reworks gen3 string encode/decode, no longer does 3->4->5 and 5->4->3;
instead goes straight to the end result without an intermediary format.
String sanitization should probably be broken up rather than reused, oh
well.
PK2 doesn't like getting passed null for the byte[] parameter (rightly
so); don't depend on the src code order having the first constructor
being the one with the least arguments
fix it in pk1/2 so that the first constructor has the least args, anyway
remove Identifier param. 99% of the time, identifier is not provided,
resulting in a useless call
end result:
provide 'create new' and 'create from' constructors
Logic is essentially identical; implement a base class and have the
generation specific structures implment the differences.
Reduce the verbosity a little
pk1 boxdata stores current level & current hp, which is only present in
pk2 party data.
if the user drops in pk2(boxdata), the transfer leaves 0 for both
values, which isn't correct.
detect stat_level to determine if values should be regenerated or not.
Thanks HaxAras for finding this :)
Remove unnecessary floor operation, don't fetch stat arrays for each
stat
Current HP is a box stat in pk1 format (offset 0x1), which isn't stored
in box pk2's. If the hp is zero, set it to the current HP.
Thanks HaxAras for the conversion tip!
convert all dictionaries to char-byte instead of string (saves 50KB on
compressed dll, lul)
update OT string comparison for pre/post transfer specimens
Showing "TRAINER" for all languages isn't correct, just show a mapped
character
Closes#2049 , thanks @egzonqj & @WEERSOQUEER !
all usages besides fetching a 32bit random value should use rand.next
remove unnecessary do-while loop for calculating random EVs (always
returns 510 in total)
gender: take top 4 bits of gr:
31 = 0x1F = 1
63 = 0x3F = 3
127 = 0x7F = 7
191 = 0xBF = 11
See the pattern? If we change the compares from >= to >, we -1. All
numbers match except for the 25/75 ratio pkm... which unveils the
problem.
Simplify the calc for these using the logic above, which fixes the error
and makes the code easier to read!
Thanks SystemError for assisting :)
