From 5e17144cb756345eae945ba3f6413c60472d04ec Mon Sep 17 00:00:00 2001 From: dumbmoron <136796770+dumbmoron@users.noreply.github.com> Date: Fri, 11 Aug 2023 11:43:07 +0000 Subject: [PATCH] fix instagram downloads --- src/modules/processing/services/instagram.js | 70 +++++++++++++++----- 1 file changed, 52 insertions(+), 18 deletions(-) diff --git a/src/modules/processing/services/instagram.js b/src/modules/processing/services/instagram.js index 8544d103..d1714393 100644 --- a/src/modules/processing/services/instagram.js +++ b/src/modules/processing/services/instagram.js @@ -1,35 +1,69 @@ -import got from "got"; +import { createStream } from "../../stream/manage.js"; +import { genericUserAgent } from "../../config.js"; export default async function(obj) { - // i hate this implementation but fetch doesn't work here for some reason (i personally blame facebook) - let html; + let data; try { - html = await got.get(`https://www.instagram.com/p/${obj.id}/`) - html.on('error', () => { - html = false; - }); - html = html ? html.body : false; + const url = new URL('https://www.instagram.com/graphql/query/'); + url.searchParams.set('query_hash', 'b3055c01b4b222b8a47dc12b090e4e64') + url.searchParams.set('variables', JSON.stringify({ + child_comment_count: 3, + fetch_comment_count: 40, + has_threaded_comments: true, + parent_comment_count: 24, + shortcode: obj.id + })) + + data = await fetch(url, { + headers: { + 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9', + 'User-Agent': genericUserAgent, + 'X-Ig-App-Id': '936619743392459', + 'X-Asbd-Id': '129477', + 'Sec-Fetch-Dest': 'empty', + 'Sec-Fetch-Mode': 'cors', + 'Sec-Fetch-Site': 'same-origin', + 'upgrade-insecure-requests': '1', + 'accept-encoding': 'gzip, deflate, br', + 'accept-language': 'en-US,en;q=0.9,en;q=0.8' + } + }) + data = (await data.json()).data; } catch (e) { - html = false; + data = false; } - if (!html) return { error: 'ErrorCouldntFetch' }; - if (!html.includes('application/ld+json')) return { error: 'ErrorEmptyDownload' }; - let single, multiple = [], postInfo = JSON.parse(html.split('script type="application/ld+json"')[1].split('">')[1].split('')[0]); - if (Array.isArray(postInfo)) postInfo = postInfo[0]; + if (!data) return { error: 'ErrorCouldntFetch' }; - if (postInfo.video.length > 1) { - for (let i in postInfo.video) { multiple.push({type: "video", thumb: postInfo.video[i]["thumbnailUrl"], url: postInfo.video[i]["contentUrl"]}) } - } else if (postInfo.video.length === 1) { - single = postInfo.video[0]["contentUrl"] + let single, multiple = []; + const sidecar = data?.shortcode_media?.edge_sidecar_to_children; + if (sidecar) { + sidecar.edges.forEach(e => { + // todo: allow downloading images once frontend supports it + if (e.node?.is_video) { + multiple.push({ + type: "video", + // thumbnails have `Cross-Origin-Resource-Policy` set to `same-origin`, so we need to proxy them + thumb: createStream({ + service: "instagram", + type: "default", + u: e.node?.display_url, + filename: "image.jpg" + }), + url: e.node?.video_url + }) + } + }) + } else if (data?.shortcode_media?.video_url) { + single = data.shortcode_media.video_url } else { return { error: 'ErrorEmptyDownload' } } if (single) { return { urls: single, filename: `instagram_${obj.id}.mp4`, audioFilename: `instagram_${obj.id}_audio` } - } else if (multiple) { + } else if (multiple.length) { return { picker: multiple } } else { return { error: 'ErrorEmptyDownload' }