]> git.immae.eu Git - github/Chocobozzz/PeerTube.git/blame - server/lib/activitypub/crawl.ts
Fix video upload with a capitalized ext
[github/Chocobozzz/PeerTube.git] / server / lib / activitypub / crawl.ts
CommitLineData
edb4ffc7 1import * as Bluebird from 'bluebird'
a1587156 2import { URL } from 'url'
db4b15f2
C
3import { ActivityPubOrderedCollection } from '../../../shared/models/activitypub'
4import { logger } from '../../helpers/logger'
5import { doJSONRequest } from '../../helpers/requests'
6import { ACTIVITY_PUB, REQUEST_TIMEOUT, WEBSERVER } from '../../initializers/constants'
8fffe21a 7
2ba92871
C
8type HandlerFunction<T> = (items: T[]) => (Promise<any> | Bluebird<any>)
9type CleanerFunction = (startedDate: Date) => (Promise<any> | Bluebird<any>)
10
db4b15f2
C
11async function crawlCollectionPage <T> (argUrl: string, handler: HandlerFunction<T>, cleaner?: CleanerFunction) {
12 let url = argUrl
13
14 logger.info('Crawling ActivityPub data on %s.', url)
8fffe21a
C
15
16 const options = {
8fffe21a 17 activityPub: true,
dcd75f78 18 timeout: REQUEST_TIMEOUT
8fffe21a
C
19 }
20
2ba92871
C
21 const startDate = new Date()
22
db4b15f2 23 const response = await doJSONRequest<ActivityPubOrderedCollection<T>>(url, options)
8fffe21a
C
24 const firstBody = response.body
25
a1587156 26 const limit = ACTIVITY_PUB.FETCH_PAGE_LIMIT
8fffe21a
C
27 let i = 0
28 let nextLink = firstBody.first
29 while (nextLink && i < limit) {
ee79b60e 30 let body: any
c0e71e84 31
ee79b60e
C
32 if (typeof nextLink === 'string') {
33 // Don't crawl ourselves
a1587156 34 const remoteHost = new URL(nextLink).host
ee79b60e
C
35 if (remoteHost === WEBSERVER.HOST) continue
36
db4b15f2 37 url = nextLink
ee79b60e 38
db4b15f2 39 const res = await doJSONRequest<ActivityPubOrderedCollection<T>>(url, options)
ee79b60e
C
40 body = res.body
41 } else {
42 // nextLink is already the object we want
43 body = nextLink
44 }
8fffe21a 45
8fffe21a
C
46 nextLink = body.next
47 i++
48
49 if (Array.isArray(body.orderedItems)) {
50 const items = body.orderedItems
db4b15f2 51 logger.info('Processing %i ActivityPub items for %s.', items.length, url)
8fffe21a
C
52
53 await handler(items)
54 }
55 }
2ba92871
C
56
57 if (cleaner) await cleaner(startDate)
8fffe21a
C
58}
59
60export {
61 crawlCollectionPage
62}