feat(embed): add upcloud and streamsb embed scrapers

src/backend/embeds/streamsb.ts

@@ -0,0 +1,212 @@
+import Base64 from "crypto-js/enc-base64";
+import Utf8 from "crypto-js/enc-utf8";
+
+import { MWEmbedType } from "@/backend/helpers/embed";
+import { proxiedFetch } from "@/backend/helpers/fetch";
+import { registerEmbedScraper } from "@/backend/helpers/register";
+import {
+  MWCaptionType,
+  MWStreamQuality,
+  MWStreamType,
+} from "@/backend/helpers/streams";
+
+const qualityOrder = [
+  MWStreamQuality.Q1080P,
+  MWStreamQuality.Q720P,
+  MWStreamQuality.Q480P,
+  MWStreamQuality.Q360P,
+];
+
+async function fetchCaptchaToken(domain: string, recaptchaKey: string) {
+  const domainHash = Base64.stringify(Utf8.parse(domain)).replace(/=/g, ".");
+
+  const recaptchaRender = await proxiedFetch<any>(
+    `https://www.google.com/recaptcha/api.js?render=${recaptchaKey}`
+  );
+
+  const vToken = recaptchaRender.substring(
+    recaptchaRender.indexOf("/releases/") + 10,
+    recaptchaRender.indexOf("/recaptcha__en.js")
+  );
+
+  const recaptchaAnchor = await proxiedFetch<any>(
+    `https://www.google.com/recaptcha/api2/anchor?ar=1&hl=en&size=invisible&cb=flicklax&k=${recaptchaKey}&co=${domainHash}&v=${vToken}`
+  );
+
+  const cToken = new DOMParser()
+    .parseFromString(recaptchaAnchor, "text/html")
+    .getElementById("recaptcha-token")
+    ?.getAttribute("value");
+
+  if (!cToken) throw new Error("Unable to find cToken");
+
+  const payload = {
+    v: vToken,
+    reason: "q",
+    k: recaptchaKey,
+    c: cToken,
+    sa: "",
+    co: domain,
+  };
+
+  const tokenData = await proxiedFetch<string>(
+    `https://www.google.com/recaptcha/api2/reload?${new URLSearchParams(
+      payload
+    ).toString()}`,
+    {
+      headers: { referer: "https://www.google.com/recaptcha/api2/" },
+      method: "POST",
+    }
+  );
+
+  const token = tokenData.match('rresp","(.+?)"');
+  return token ? token[1] : null;
+}
+
+registerEmbedScraper({
+  id: "streamsb",
+  displayName: "StreamSB",
+  for: MWEmbedType.STREAMSB,
+  rank: 150,
+  async getStream({ url, progress }) {
+    /* Url variations
+    - domain.com/{id}?.html
+    - domain.com/{id}
+    - domain.com/embed-{id}
+    - domain.com/d/{id}
+    - domain.com/e/{id}
+    - domain.com/e/{id}-embed
+    */
+    const streamsbUrl = url
+      .replace(".html", "")
+      .replace("embed-", "")
+      .replace("e/", "")
+      .replace("d/", "");
+
+    const parsedUrl = new URL(streamsbUrl);
+    const base = await proxiedFetch<any>(
+      `${parsedUrl.origin}/d${parsedUrl.pathname}`
+    );
+
+    progress(20);
+
+    // Parse captions from url
+    const captionUrl = parsedUrl.searchParams.get("caption_1");
+    const captionLang = parsedUrl.searchParams.get("sub_1");
+
+    const basePage = new DOMParser().parseFromString(base, "text/html");
+
+    const downloadVideoFunctions = basePage.querySelectorAll(
+      "[onclick^=download_video]"
+    );
+
+    const dlDetails = [];
+    for (const func of downloadVideoFunctions) {
+      const funcContents = func.getAttribute("onclick");
+      const regExpFunc = /download_video\('(.+?)','(.+?)','(.+?)'\)/;
+      const matchesFunc = regExpFunc.exec(funcContents ?? "");
+      if (matchesFunc !== null) {
+        const quality = func.querySelector("span")?.textContent;
+        const regExpQuality = /(.+?) \((.+?)\)/;
+        const matchesQuality = regExpQuality.exec(quality ?? "");
+        if (matchesQuality !== null) {
+          dlDetails.push({
+            parameters: [matchesFunc[1], matchesFunc[2], matchesFunc[3]],
+            quality: {
+              label: matchesQuality[1].trim(),
+              size: matchesQuality[2],
+            },
+          });
+        }
+      }
+    }
+
+    progress(40);
+
+    let dls = await Promise.all(
+      dlDetails.map(async (dl) => {
+        const getDownload = await proxiedFetch<any>(
+          `/dl?op=download_orig&id=${dl.parameters[0]}&mode=${dl.parameters[1]}&hash=${dl.parameters[2]}`,
+          {
+            baseURL: parsedUrl.origin,
+          }
+        );
+
+        const downloadPage = new DOMParser().parseFromString(
+          getDownload,
+          "text/html"
+        );
+
+        const recaptchaKey = downloadPage
+          .querySelector(".g-recaptcha")
+          ?.getAttribute("data-sitekey");
+        if (!recaptchaKey) throw new Error("Unable to get captcha key");
+
+        const captchaToken = await fetchCaptchaToken(
+          parsedUrl.origin,
+          recaptchaKey
+        );
+        if (!captchaToken) throw new Error("Unable to get captcha token");
+
+        const dlForm = new FormData();
+        dlForm.append("op", "download_orig");
+        dlForm.append("id", dl.parameters[0]);
+        dlForm.append("mode", dl.parameters[1]);
+        dlForm.append("hash", dl.parameters[2]);
+        dlForm.append("g-recaptcha-response", captchaToken);
+
+        const download = await proxiedFetch<any>(
+          `/dl?op=download_orig&id=${dl.parameters[0]}&mode=${dl.parameters[1]}&hash=${dl.parameters[2]}`,
+          {
+            baseURL: parsedUrl.origin,
+            method: "POST",
+            body: dlForm,
+          }
+        );
+
+        const dlLink = new DOMParser()
+          .parseFromString(download, "text/html")
+          .querySelector(".btn.btn-light.btn-lg")
+          ?.getAttribute("href");
+
+        console.log(dlLink);
+
+        return {
+          quality: dl.quality.label as MWStreamQuality,
+          url: dlLink,
+          size: dl.quality.size,
+          captions:
+            captionUrl && captionLang
+              ? [
+                  {
+                    url: captionUrl,
+                    langIso: captionLang,
+                    type: MWCaptionType.VTT,
+                  },
+                ]
+              : [],
+        };
+      })
+    );
+    dls = dls.filter((d) => !!d.url);
+    dls = dls.sort((a, b) => {
+      const aQuality = qualityOrder.indexOf(a.quality);
+      const bQuality = qualityOrder.indexOf(b.quality);
+      return aQuality - bQuality;
+    });
+
+    progress(60);
+
+    // TODO: Quality selection for embed scrapers
+    const dl = dls[0];
+    if (!dl.url) throw new Error("No stream url found");
+
+    return {
+      embedId: MWEmbedType.STREAMSB,
+      streamUrl: dl.url,
+      quality: dl.quality,
+      captions: dl.captions,
+      type: MWStreamType.MP4,
+    };
+  },
+});

src/backend/embeds/upcloud.ts

@@ -0,0 +1,93 @@
+import { AES, enc } from "crypto-js";
+
+import { MWEmbedType } from "@/backend/helpers/embed";
+import { registerEmbedScraper } from "@/backend/helpers/register";
+import {
+  MWCaptionType,
+  MWStreamQuality,
+  MWStreamType,
+} from "@/backend/helpers/streams";
+
+import { proxiedFetch } from "../helpers/fetch";
+
+interface StreamRes {
+  server: number;
+  sources: string;
+  tracks: {
+    file: string;
+    kind: "captions" | "thumbnails";
+    label: string;
+  }[];
+}
+
+function isJSON(json: string) {
+  try {
+    JSON.parse(json);
+    return true;
+  } catch {
+    return false;
+  }
+}
+
+registerEmbedScraper({
+  id: "upcloud",
+  displayName: "UpCloud",
+  for: MWEmbedType.UPCLOUD,
+  rank: 200,
+  async getStream({ url }) {
+    // Example url: https://dokicloud.one/embed-4/{id}?z=
+    const parsedUrl = new URL(url.replace("embed-5", "embed-4"));
+
+    const dataPath = parsedUrl.pathname.split("/");
+    const dataId = dataPath[dataPath.length - 1];
+
+    const streamRes = await proxiedFetch<StreamRes>(
+      `${parsedUrl.origin}/ajax/embed-4/getSources?id=${dataId}`,
+      {
+        headers: {
+          Referer: parsedUrl.origin,
+          "X-Requested-With": "XMLHttpRequest",
+        },
+      }
+    );
+
+    let sources:
+      | {
+          file: string;
+          type: string;
+        }
+      | string = streamRes.sources;
+
+    if (!isJSON(sources) || typeof sources === "string") {
+      const decryptionKey = await proxiedFetch<string>(
+        `https://raw.githubusercontent.com/enimax-anime/key/e4/key.txt`
+      );
+
+      const decryptedStream = AES.decrypt(sources, decryptionKey).toString(
+        enc.Utf8
+      );
+
+      const parsedStream = JSON.parse(decryptedStream)[0];
+      if (!parsedStream) throw new Error("No stream found");
+      sources = parsedStream as { file: string; type: string };
+    }
+
+    return {
+      embedId: MWEmbedType.UPCLOUD,
+      streamUrl: sources.file,
+      quality: MWStreamQuality.Q1080P,
+      type: MWStreamType.HLS,
+      captions: streamRes.tracks
+        .filter((sub) => sub.kind === "captions")
+        .map((sub) => {
+          return {
+            langIso: sub.label,
+            url: sub.file,
+            type: sub.file.endsWith("vtt")
+              ? MWCaptionType.VTT
+              : MWCaptionType.UNKNOWN,
+          };
+        }),
+    };
+  },
+});