Some tweaks + more tests

src/Utility.ts

@@ -180,10 +180,13 @@ export function createNotesFromMoebooruBoxes(document: Document, boxSize: [numbe
   return notes;
 }
 
-export function parseResolutionString(str: string): [number, number] | undefined {
+export function parseResolutionString(str: string | undefined | null): [number, number] | undefined {
+  if (!str) return undefined;
+
   // Example string: 1600x2200
   const res = str
     .split("x") // Split on 'x' character
+    .map((v, _) => v.replace(/\D/g, "")) // Remove all non-digits
     .map((v, _) => parseInt(v)) // Parse ints
     .filter(Number); // Filter NaN. Also removes the number 0, but that's fine because the resolution can't be 0.
 

src/engines/Gelbooru.test.ts

@@ -62,3 +62,17 @@ test("get rule34.xxx post 2740994", async () => {
   expect(res.posts[0].pageUrl).toBe(pageUrl);
   expect(res.posts[0].contentUrl).toContain(".rule34.xxx//images/2472/35bf31c98f53a369c047baf9bae496fb.jpeg");
 });
+
+test("get xbooru.com post 834083", async () => {
+  // Don't worry this isn't a cursed image.
+  const pageUrl = "https://xbooru.com/index.php?page=post&s=view&id=834083";
+  const res = await scrapeUrl(pageUrl);
+  expect(res.posts[0]).toHaveTag("friendship_is_magic", "copyright");
+  expect(res.posts[0]).toHaveTag("nurse_redheart", "character");
+  expect(res.posts[0]).toHaveTag("nurse");
+  expect(res.posts[0]).toHaveTag("standing");
+  expect(res.posts[0]).toHaveResolution([1024, 1878]);
+  expect(res.posts.length).toBe(1);
+  expect(res.posts[0].pageUrl).toBe(pageUrl);
+  expect(res.posts[0].contentUrl).toBe("https://img.xbooru.com//images/584/040dc1625c094a7d5d7b0918aa0d8324.png");
+});

src/engines/Gelbooru.ts

@@ -21,7 +21,11 @@ export default class Gelbooru extends ScrapeEngineBase {
 
   canImport(url: Location): boolean {
     return (
-      url.host == "safebooru.org" || url.host == "gelbooru.com" || url.host == "rule34.xxx" || url.host == "tbib.org"
+      url.host == "safebooru.org" ||
+      url.host == "gelbooru.com" ||
+      url.host == "rule34.xxx" ||
+      url.host == "tbib.org" ||
+      url.host == "xbooru.com"
     );
   }
 
@@ -32,6 +36,7 @@ export default class Gelbooru extends ScrapeEngineBase {
       case "safebooru.org":
       case "rule34.xxx":
       case "tbib.org":
+      case "xbooru.com":
         version = Version.v020;
         break;
       case "gelbooru.com":

src/engines/Rule34us.test.ts

@@ -0,0 +1,18 @@
+import { scrapeUrl } from "../test/engine";
+
+test("get rule34.us post 4646579", async () => {
+  // Kinda sketchy.
+  const pageUrl = "https://rule34.us/index.php?r=posts/view&id=4646579";
+  const res = await scrapeUrl(pageUrl);
+  expect(res.posts.length).toBe(1);
+  expect(res.posts[0]).toHaveTag("sciamano240", "artist");
+  expect(res.posts[0]).toHaveTag("mario", "character");
+  expect(res.posts[0]).toHaveTag("nintendo", "copyright");
+  expect(res.posts[0]).toHaveTag("crown");
+  expect(res.posts[0]).toHaveTag("solo");
+  expect(res.posts[0]).not.toHaveTag("original");
+  expect(res.posts[0]).not.toHaveTag("edit");
+  expect(res.posts[0]).toHaveResolution([2100, 3000]);
+  expect(res.posts[0].pageUrl).toBe(pageUrl);
+  expect(res.posts[0].contentUrl).toBe("https://img2.rule34.us/images/91/16/91160933dc1b3c737b3ef4a25b12680d.png");
+});

src/engines/Rule34us.ts

@@ -4,14 +4,14 @@
 import { ScrapeEngine, ScrapeResult, ScrapedPost, ScrapedTag } from "../ScrapeEngine";
 import { TagCategory } from "../BooruTypes";
 import { CategoryMap } from "./Common";
-import { guessContentType } from "../Utility";
+import { guessContentType, parseResolutionString } from "../Utility";
 
 export default class Gelbooru implements ScrapeEngine {
   name = "rule34us";
 
   private readonly classNameToCategoryMap: CategoryMap = {
     "character-tag": "character",
-    "artist-tag": "character",
+    "artist-tag": "artist",
     "copyright-tag": "copyright",
     "metadata-tag": "meta",
   };
@@ -40,6 +40,13 @@ export default class Gelbooru implements ScrapeEngine {
     // Set content type
     post.contentType = guessContentType(post.contentUrl);
 
+    // Set resolution
+    const statEls = Array.from(document.querySelectorAll<HTMLLIElement>(".container > ul > div > li"));
+    const sizeEl = statEls.find((x) => x.innerText.indexOf("Size:") != -1);
+    if (sizeEl) {
+      post.resolution = parseResolutionString(sizeEl.innerText);
+    }
+
     // Patch contentUrl if contentType == "video"
     // This is because the "Original" button does link to "302 Found" for videos. Idk why.
     // Shit's fucked, there are multiple workarounds for that and this is one of them.

src/engines/SankakuComplex.test.ts

@@ -14,8 +14,8 @@ test("get chan.sankakucomplex.com post 19614481 with notes", async () => {
   expect(res.posts[0]).toHaveNote("What are you looking at...");
   expect(res.posts[0]).toHaveNote("What...?");
   expect(res.posts[0].pageUrl).toBe(pageUrl);
-  expect(res.posts[0].contentUrl).toBe(
-    "https://s.sankakucomplex.com/data/50/7e/507e986a62cc5b866d92ad491dcfb1a4.jpg?e=1637625160&m=e0n_DCg-3fnmrhGzpI0Ftg"
+  expect(res.posts[0].contentUrl).toContain(
+    "https://s.sankakucomplex.com/data/50/7e/507e986a62cc5b866d92ad491dcfb1a4.jpg"
   );
 });
 
@@ -31,7 +31,7 @@ test("get chan.sankakucomplex.com post 29354061 with notes", async () => {
   expect(res.posts[0]).toHaveTag("note_translated");
   expect(res.posts[0]).toHaveNote("No.  \nI already told you...");
   expect(res.posts[0].pageUrl).toBe(pageUrl);
-  expect(res.posts[0].contentUrl).toBe(
-    "https://s.sankakucomplex.com/data/c8/37/c8376f02608548365d394a658e85cb3f.jpg?e=1637625020&m=A9xq-zht_FoVfoCiSRA-GA"
+  expect(res.posts[0].contentUrl).toContain(
+    "https://s.sankakucomplex.com/data/c8/37/c8376f02608548365d394a658e85cb3f.jpg"
   );
 });

src/engines/Shuushuu.test.ts

@@ -4,6 +4,12 @@ test("get e-shuushuu.net post 1063586", async () => {
   const pageUrl = "https://e-shuushuu.net/image/1063586/";
   const res = await scrapeUrl(pageUrl);
   expect(res.posts.length).toBe(1);
+  expect(res.posts[0]).toHaveTag("monster");
+  expect(res.posts[0]).toHaveTag("wings");
+  expect(res.posts[0]).toHaveTag("celebi", "character");
+  expect(res.posts[0]).toHaveTag("shiina_mirin", "artist");
+  expect(res.posts[0]).toHaveTag("pokémon", "copyright");
+  expect(res.posts[0]).toHaveResolution([1447, 1023]);
   expect(res.posts[0].pageUrl).toBe(pageUrl);
   expect(res.posts[0].contentUrl).toBe("https://e-shuushuu.net/images/2021-08-28-1063586.jpeg");
 });

src/engines/Shuushuu.ts

@@ -1,6 +1,6 @@
 import { ScrapeEngine, ScrapeResult, ScrapedPost, ScrapedTag } from "../ScrapeEngine";
 import { TagCategory } from "../BooruTypes";
-import { guessContentType } from "../Utility";
+import { guessContentType, parseResolutionString } from "../Utility";
 
 export default class Shuushuu implements ScrapeEngine {
   name = "shuushuu";
@@ -26,6 +26,19 @@ export default class Shuushuu implements ScrapeEngine {
     // Set content type
     post.contentType = guessContentType(post.contentUrl);
 
+    // Set resolution
+    const metaEls = Array.from(document.querySelector<HTMLDataListElement>(".meta > dl")?.children ?? []);
+    const dimsDtEl = metaEls.find((x) => (<any>x).innerText == "Dimensions:");
+    if (dimsDtEl) {
+      const dimsDtIdx = metaEls.indexOf(dimsDtEl);
+      let str = (<any>metaEls[dimsDtIdx + 1])?.innerText;
+      if (str) {
+        // When str is "1447x1023 (1.480 MPixel)"
+        str = str.split("(")[0]; // Remove the "(1.480 MPixel)" part
+        post.resolution = parseResolutionString(str);
+      }
+    }
+
     // Set rating
     post.rating = "safe";
 

src/engines/Zerochan.test.ts

@@ -4,8 +4,11 @@ test("get zerochan.net post 2934998", async () => {
   const pageUrl = "https://www.zerochan.net/2934998";
   const res = await scrapeUrl(pageUrl);
   expect(res.posts.length).toBe(1);
+  expect(res.posts[0]).toHaveTag("nintendo", "copyright");
+  expect(res.posts[0]).toHaveTag("akihorisu", "artist");
+  expect(res.posts[0]).toHaveTag("shirona_(pokémon)", "character");
+  expect(res.posts[0]).toHaveResolution([1186, 1009]);
   expect(res.posts[0].pageUrl).toBe(pageUrl);
-  expect(res.posts[0].tags.find((x) => x.name == "nintendo")).not.toBeUndefined();
   expect(res.posts[0].contentUrl).toBe("https://static.zerochan.net/Shirona.%28Pok%C3%A9mon%29.full.2934998.png");
 });
 
@@ -13,7 +16,8 @@ test("get zerochan.net gif post 3511809", async () => {
   const pageUrl = "https://www.zerochan.net/3511809";
   const res = await scrapeUrl(pageUrl);
   expect(res.posts.length).toBe(1);
-  expect(res.posts[0].pageUrl).toBe(pageUrl);
   expect(res.posts[0]).toHaveTag("rwby");
+  expect(res.posts[0]).toHaveResolution([500, 150]);
+  expect(res.posts[0].pageUrl).toBe(pageUrl);
   expect(res.posts[0].contentUrl).toBe("https://static.zerochan.net/Ruby.Rose.full.3511809.gif");
 });

src/engines/Zerochan.ts

@@ -1,6 +1,6 @@
 import { ScrapeEngine, ScrapeResult, ScrapedPost, ScrapedTag } from "../ScrapeEngine";
 import { TagCategory } from "../BooruTypes";
-import { guessContentType } from "../Utility";
+import { guessContentType, parseResolutionString } from "../Utility";
 
 export default class Zerochan implements ScrapeEngine {
   name = "zerochan";
@@ -30,6 +30,12 @@ export default class Zerochan implements ScrapeEngine {
     // Set content type
     post.contentType = guessContentType(post.contentUrl);
 
+    // Set resolution
+    const resSizeEl = document.querySelector("#large > p > br")?.parentNode as HTMLParagraphElement;
+    if (resSizeEl) {
+      post.resolution = parseResolutionString(resSizeEl.childNodes[0].textContent);
+    }
+
     // Set rating
     post.rating = "safe";
 
@@ -50,9 +56,11 @@ export default class Zerochan implements ScrapeEngine {
       switch (tagElement.className) {
         case "game":
         case "series":
+        case "studio":
           category = "copyright";
           break;
         case "character":
+        case "character group":
           category = "character";
           break;
         case "mangaka":

src/test/jest.setup.ts

@@ -3,10 +3,9 @@ import { TagCategory } from "../BooruTypes";
 import { ScrapedPost } from "../ScrapeEngine";
 
 // Expose JSDOM Element constructor
-// @ts-ignore
 global.Element = new JSDOM().window.Element;
 // 'Implement' innerText in JSDOM: https://github.com/jsdom/jsdom/issues/1245
-// @ts-ignore
+// This implementation is not 100% correct, but usually good enough so that the tests pass.
 Object.defineProperty(global.Element.prototype, "innerText", {
   get() {
     return this.textContent.replace(/(\r\n|\n|\r|^\s+|\s+$)/gm, "");