Client-ID will be required in August

[DallasNChains]

Heads up! We’re going to require the Client-ID header along with your API requests as of August 8th. Thus far, it has been a soft requirement, but we’re going to start enforcing it.

More details on the blog.

[scottrc]

Thanks for the heads up. Is there any way to test that I am sending the headers correctly? I am working with Java.

[DallasNChains]

The API doesn’t send back anything in particular to let you know. The best way to test this is to examine your network traffic (using DevTools, Fiddler, or equivalent) to ensure the headers are being sent.

[sundheden]

Will it also work by suffixing the get-url with the client id, like it does today with token?
Example: https://api.twitch.tv/kraken/channels/channelname/subscriptions?oauth_token=XXXXXX&client_id=XXXX

[3ventic]

Yes, though it will be useless for GET /streams, where it has a different meaning (unless that changes).

[night]

You can now check if you’re properly identifying by pulling

https://api.twitch.tv/kraken

with your client ID specified. If properly identified, the identified field will return true.

[DallasNChains]

Thanks for that @night! I was just going to make a post about it.

[BtbN]

Is the clientID I get for my “App” safe to commit to a public repository on github?

[Dev]

Displaying the ClientID shouldn’t be to dangerous, as long as you keep your Client Secret confidential.

[sundheden]

I assume this is a way for you to “track” API requests, is there any plan on adding a query per minute limit on requests too? (or is there one already?)

[DallasNChains]

Do you mean a rate limit on API requests? Currently, if you don’t include Client-ID, you’re subject to rate limiting. See the rate limits section here.

[ARTI]

Thank you for keeping us informed Dallas!

[twitchdev]

I am not getting the “identified” value, either True or False, back when making requests in Python to the “kraken” endpoint, while supplying the client id value.

https_request(
    "api.twitch.tv",
    "GET",
    "/kraken/user",
    { "Accept":        "application/vnd.twitchtv.v3+json",
      "Authorization": "OAuth " + accessToken,
      "Client-ID":     CLIENT_ID }

Reply

{
  "_id": 00000000,
  "_links": {
    "self": "https://api.twitch.tv/kraken/users/XXXXXXX"
  },
  "bio": null,
  "created_at": "XXXXXXX",
  "display_name": "XXXXXXX",
  "email": "XXXXXXX",
  "logo": null,
  "name": "XXXXXXX",
  "notifications": {
    "email": false,
    "push": false
  },
  "partnered": false,
  "type": "user",
  "updated_at": "XXXXXXX"
}

[DallasNChains]

You’re sending the request to kraken/users. It should be just the root kraken as shown in the FAQ here: Client-ID Requirement FAQs