jjallaire
commented
5 years ago Here's my take on what we'd need to do here. All of the action happens in the upload.js file here: https://github.com/jjallaire/draftpod/blob/master/src/components/navigator/cardpool/upload.js
First, we'll need to update our extraction/validation logic to look for the "Count", "Edition", and "Card Number" fields. That code is here:
We'll want to read the Edition and then validate that it matches the passed set_code. We can probably do this for the existing sets by comparing "Edition" to the return value of set.name(set_code). If it doesn't match then we'll want to set valid to false and push an appropriate error.
In terms of the other fields, we'll probably want to add support for "Card Number" and "Count" to the readers for id and quantity respectively.
Finally, we'll want the code in completeCardpoolUpload() to be updated to do conversions from collector number ("Card Number") to Multiverse ID (we do that in this function b/c that's where we currently get access to the full Scryfall dataset: https://github.com/jjallaire/draftpod/blob/master/src/components/navigator/cardpool/upload.js#L136
I think the best way to do the conversion is probably to build a hashtable of collector number => card then just do lookups. We'll also need a way to determine when the conversion is actually necessary. This could be done a number of different ways, including using the presence of certain field names (e.g. "Card Number") in the CSV as an indicator, or even just the fact that the id's don't fall into the normal Mvid range. I don't have a strong opinion on how this should work.
dci2112
It would be great to be able to import from Deckbox as well as Decked Builder. From what I can see the standard CSV export from Deckbox has the info we need to match up to Scryfall data:
We will need the "Count", "Edition", and "Card Number" fields to do the import. I'll polk around a bit to see if anything in our current code needs to change significantly to accommodate this.