Calling the Twitch API
The API methods are namespaced into the different API sections
like Kraken and Helix, and below that, into the different
subresources like /users
and /streams
.
All API methods are async
and thus can be
await
ed. Here's a quick example:
async function isStreamLive(userName: string) {
const user = await apiClient.helix.users.getUserByName(userName);
if (!user) {
return false;
}
return await apiClient.helix.streams.getStreamByUserId(user.id) !== null;
}
We make extensive use of convenience methods that fetch related resources, so this can also be written a bit easier:
async function isStreamLive(userName: string) {
const user = await apiClient.helix.users.getUserByName(userName);
if (!user) {
return false;
}
return await user.getStream() !== null;
}
In Helix, some resources are paginated using a cursor. To
faciliate easy pagination, there are special methods suffixed
with Paginated
that wrap the result in a
HelixPaginatedRequest
object. There are multiple ways to use this object to get your
results.
- Using
getNext()
:
async function getAllClipsForBroadcaster(userId: string) {
const request = apiClient.helix.clips.getClipsForBroadcasterPaginated(userId);
let page: HelixClip[];
const result: HelixClip[] = [];
while ((page = await request.getNext()).length) {
result.push(...page);
}
return result;
}
- Using
getAll()
:
async function getAllClipsForBroadcaster(userId: string) {
const request = apiClient.helix.clips.getClipsForBroadcasterPaginated(userId);
return request.getAll();
}
-
Using a
for await ... of
loop:
async function findClipFromBroadcasterWithTitle(userId: string, searchTerm: string) {
for await (const clip of apiClient.helix.clips.getClipsForBroadcasterPaginated(userId)) {
if (clip.title.includes(searchTerm)) {
return clip;
}
}
return null;
}
WARNING: Uncontrolled use of the paginator may lead to quick exhaustion of your rate limit with big, unfiltered result sets. Please use filters and exit your loops as soon as you have enough data.