Forge loads all files found on the setlookup folder.
Example J21.txt (J21 set code) contains a list of set codes (sorted from Newest to Oldest beforehand) to lookup for images on other sets.
The new implementation now contains a completely refactored version of the algorithm, also including the invocation of the latest method for getAlternativeCardPrint in StaticData (with extra statistics paramenters gathered from current pool). This include optimisation for card frame, and cards from Expansion sets.
Moreover, the new implementation now automatically distribute card arts for **all** the cards in the pool - regardless they need to be "replaced" with alternative print or not.
THE ONLY case when this is not happening is when there is NO card to update in the current pool. This makes sense as this means editions has been specified, so the Deck won't receive any UNDESIRED alteration.
CardPool API has been extended by including utility methods to gather specific statistics about a cardPool. These statistics include:
- Distribution (card count frequency) of the Card Edition included in the Pool
- Distribution of the Card Edition Type included in the Pool
- Retrieving the most common Card Edition Type among those included in the Pool
- Determine whether or not the Pool "isModern" (i.e. the majority of cards is gathered from Modern Sets)
- get the PivotCardEdition: this is the cornerstone of card art optimisation for decks
The PivotCardEdition is the edition that will be considered the threshold boundary for cards in the pool. Any decision of card arts for other cards will be considered based on the PivotEdition, that is "alternativeCardPrint" released BEFORE or AFTER (depending on the current Card Art Preference) the PIVOT EDITION RELEASE DATE.
Also, this commit includes an optimisation in add method implementation, getting rid of lots of duplicated code!
Methods related to Card Art Preference Settings, and for Card Art Smart Selection have been renamed with better wording.
Most importantly, StaticData now includes a new method to retrieve the number of art count for a given PaperCard, as well as a completely refactored and improved method for getAlternativeCardPrint.
In particular, the class now provides three different implementation of getAlternativeCardPrint, with the more sophisticated one also including normalisation wrt. the card frame, and whether or not the chosen alternative print should be gathered from an Expansion edition.
All these specialisation are part of the "Card Art Optimisation" algorithm included in Deck, and instructed by statistics gathered from Card Pool (next commit)
Now the smart card art selection is also included in the set of checkings that are performed to see whether or not card arts in Pools require any update.
Deferred sections for deck are only used once, and just when the deck needs to be loaded. At that time, if any optimisation to card art is enabled (and required, that is "there are cards with no pre-specified edition").
After that, sections are memorised and later updated if anything needs updates, and the last used card art preference was different.
CardEdition now includes a new Map data strcuture that is allocated only if necessary (i.e. when lazy card loading is enabled) and it will be used for faster card in set lookup by card name. The Map will return a List of cards to account for multiple versions (prints) of a card with the same name within a CardEdition
The new `getAlternativeCardArt` method builds on the previous `getCardFromEdition` which generates alternative card art given a PaperCard and a reference date.
Extension to this method is also incorporated to force the options to adopt for card art preference policy.
If None provided, default options set up in Game preferences will be used!
CardPool now includes the logic to (a) retrieve the edition frequency map of itself, as well as retrieving its corresponding PivotEdition (depending on cards in the pool).
This refactoring optimises the code in Deck class for "smartCardArt" selection
`getTheLatestOfAllTheOriginalEditionsOfCardsIn` is the new name for `getEarliestEditionWithAllCards` which better clarifies the intent of the method.
The implementation has been optimised, according to the new CardDb behaviour and APIs.
The tests have been extended to verify that the output of the method is always consistent regardless of the edition of cards in the input CardPool.
This method will be also re-used in Decks when inspecting the list of cards to add in a Deck Section.
This refactoring avoids code duplications, and improves modularization.
New tests (and fixed updated CardRequest implementation) verify that any method used for Card Retrieval when receiving directly CardRequest formatted strings as cardName won't have any side effect. In particular, the code is reliable to be robust enough to support passing in both card name (only) or a full cardRequest string.
If the latter, any parameter in the request will be updated - accordingly - only when explicit params in function calls are provided.
Also, new tests demonstrate that the current implementation has NO side effect when testing for multiple combinations. Therefore, whatever is requested in requestString won't be overruled, unless really requested too.
