stegriff / magic-api Goto Github PK

Assume Ætherling is in the database (first character is unicode combined AE). If a user searches for Aetherling (using regular ASCII), the database search does not find the existing entry. As a result, Ætherling is downloaded again, and put in a new record.

This also happens with partial names of long, unambiguous card names. For example, 'Our Market Research Shows' will keep downloading new copies of the long-name card.

Remedy ideas:

• DB field for 'original query' to store the search string which downloaded the record. Can compare against this before card name when searching for cards.
• Do character conversion some way to fix comparisons with Unicode, and use LIKE statements to catch long card names

~~I made a design assumption that card texts would not be longer than 255 characters. Some cards, for example, "Varolz, the Scar-Striped", exceed this limit.

I'll write a database alteration to expand card text to a new limit.~~

The character limit is now 500, but it remains to be seen if there are cards with longer text. This is not well documented.

'Dispel' cannot be found because it's name is inside 'Dispeller's capsule'
'Forest' cannot be found for a similar reason

The longest card name is "Our Market Research Shows That Players Like Really Long Card Names So We Made this Card to Have the Absolute Longest Card Name Ever Elemental" from Unhinged.

This is 141 characters long. The field is currently varchar(50), so it needs to be expanded.

[http://gatherer.wizards.com/Pages/Card/Details.aspx?multiverseid=74237]

Since Gatherer's latest redesign sometime in 2014/2015, magic-api no longer correctly gets the line breaks in card text, so instead of having an _ character, it just mushes the lines together.

Some searches like 'Staunch hearted warrior' are running endlessly. Real card name is 'staunch-hearted warrior'. Will investigate -- probably a bug in the card_extractor.

Having the multiverse ID available in the API would be a huge help. This would allow a developer to link to the gatherer page, look up card images from mtgimage, etc.

The Scry frontend tries to load images from mtgimage.com which was taken down due to a copyright claim by WotC (fair enough).

Remove this functionality