/instabot

A first go at building something in Clojure. Tried different ways structuring the project.

Primary LanguageClojureGNU General Public License v2.0GPL-2.0

instabot

A Clojure app designed to ... monitor hashtags and users on instagram and fetch new media.

Deploy

  1. Make an uberjar
  2. Start with java -jar and provide the following in the ENV: client-id : The instagram client id client-secret : The instagram client secret redirect-uri : The instagram redirect uri
  3. Thats it. But I recommend to use behind nginx proxy.

Notes

Problemet med get-tagged-medias är att den returnar x antal media + en pagination. Det kan vara sjukt många poster som har gjorts på hashtagen. Detta löses genom att räkna ut antal sidor och dela upp så det blir mindre än 5.000 requests per timme.

Dock går det inte att räkna ut detta innan ut vi måste ha en counter som ser till att antal requests inte blir för många per timme.

Search funkar bara i 7dagar max. /media/search Search for media in a given area. The default time span is set to 5 days. The time span must not exceed 7 days. Defaults time stamps cover the last 5 days. Can return mix of image and video types.

För att göra denna app krävs följande API endpoints hos instagram:

Limits: Unauthenticated Calls 5,000 / hour per application

ENDPOINT UNSIGNED CALLS (PER TOKEN) SIGNED CALLS (PER TOKEN) POST /media/media-id/likes 30 / hour 100 / hour POST /media/media-id/comments 15 / hour 60 / hour POST /users/user-id/relationships 20 / hour 60 / hour

Verkar som denna endpoint är starten: http://instagram.com/developer/endpoints/tags/#get_tags_media_recent

Och sen: http://instagram.com/developer/endpoints/users/#get_users Går den att använda med endast ett klient id?

Usage

FIXME

License

TODO

  • ✓ Get the user information.
  • ✓ Save the individual images in a database
  • ✓ Save the user information.
  • ✓ Fix an update user function, which uses the instagram ID as UID.
  • ✓ Create "Spaningar", which is a that every [time] the script goes out to find the new images with the hashtag, untill the DateTime of the last fetch for the spaning.
  • ✓ Fix something that regurlary runs the Spaningar. Probably CRON, easy to run.
  • ✓ Create an web-interface to view the images of each "Spaning".
  • ✓ Create an web-interface to view the data about each user.
  • ✓ Create an web-interface to manage each "Spaning". Basic CRUD.
  • ✓ Fix throttling.
  • ✓ Fix start date on spaningar.

  • Maybe save the images every loop, instead of gathering all of them?

  • Actually use the end date.

  • Update spaningar to be either hashtag or user.

  • Save the images.

  • Döp om spaningar till: searches

  • Google maps integration

  • Filtrering på taggar, och sånt

  • Spara ner allt som zip(?)

  • Lägg till date-picker

  • Video-stöd

Copyright © 2014 FIXME

Distributed under the Eclipse Public License either version 1.0 or (at your option) any later version.