Implement source?

[ZeldaZach]

Do we want where the card came from for promos and such?

[dgant]

I love MTGJSON's coverage of promo cards. Generally, any data that MTGJSON includes, I'll find a use for, and promo card source would be no exception. Here are some of the ways that ComboDeck uses or would like to use promo card data:

• Displaying information about promo cards
• Indexing promo cards by all the keywords a person might use to search for it (for example, a person may be more likely to search for "Scrye" than "Media Inserts")
• Listing promo cards that haven't been released yet
• Translating MTGJSON promo classification to other data sources, like the TCGPlayer/MKM price APIs, or alternative card art sources

For example, suppose you're pulling prices for a card from several stores (like the TCGPlayer or MKM price APIs), etc. Most of the time, you can match their card+set names straight from the MTGJSON data. Sometimes, you have to make some small adjustments (like "Commander" instead of "Magic: the Gathering: Commander").

But their categorization of promo cards often differs greatly from each other and from MTGJSON. Compare http://combodeck.net/Card/Char/15th_Anniversary ("15th Anniversary") to https://www.magiccardmarket.eu/Products/Singles/DCI+Promos/Char ("DCI Promos") to http://shop.tcgplayer.com/magic/pro-tour-promos/char ("Pro Tour Promos").

In that example, knowing that this Char is both a Pro Tour card, as well as specifically for the 15th Anniversary, would make it easier to construct rules to match this MTGJSON printing to other data sources.

I know there's tons to do and this is likely a low priority. Just wanted to shine some light on why MTGJSON consumers might find this information useful!

[Psithief]

Source as an array of one or more strings?

[xiapoy]

I confirm you that if we had every mtgjsonSetName matching a mkmSetName, this would remarkably improve the matching ratio between the 2 sources. Currently, 70 mtgjsonSetName have no mkmSetName equivalent. Original work on this issue was made by @florentdouine (see: mtgjson/mtgjson#93). I guess that for 70 missing values, we can analyze them and update manually.