TIAPI
Overview
TIAPI = TI-API = API for tango.info
Ideally this will be a RESTful API. It will provide the ability to list collections and retrieve tracks from a collections. The object of the API is to provide a dynamic way to read from the tango.info library.
- All API quries are sent to a base URI (i.e. http://tango.info/API/ )
- Only GET methods are used as the API isn't planned to support modifying the database.
- All responses are a single internet media type (standard for the site is still undecided).
- Any response making reference to related resources are hypertext links.
Listing Collections
- HTTP GET method is sent from the user to a collection URI (i.e. http://tango.info/API/artist="D'Arienzo'/track_name="El Flete"/ ).
- The tango.info library is quired for the specified collection.
- A internet media type object is created matching the collection requested containing:
- Track Name
- Orchestra
- Artist
- Vocalist
- Album
- TINP as a hypertext link
- Year
- Duration
- TINT as a hypertext link
- (possibly more as needed)
- An HTTP response is sent back to the user containing the collection results.
Retrieving A Track's Meta Info
- HTTP GET method is sent from the user to a track URI (i.e. http://tango.info/API/00743217137229-1-4 ).
- The tango.info library is quired for the specified track.
- A internet media type object is created containing all the tracks meta information.
- An HTTP response is sent back to the user containing the collection results.
URI Structure
The structure of the URI could take on many forms:
- Parsed as a directory structure:
- Parsed as query string:
Return Format
Any Internet media format works as a return format. The key is that it is a standard form so the user knows what to expect. If there is a need to the output to be in a different format there are many conversion tools to do this (i.e. https://code.google.com/p/xml2json-xslt/ )
Options
- csv?
- xml - could be used for the whole site
- json?