Merge pull request #81 from MillenniumEarl/simplify-post-parse

Simplify post parse

.prettierrc

@@ -2,5 +2,5 @@
   "semi": true,
   "trailingComma": "none",
   "singleQuote": false,
-  "printWidth": 90
+  "printWidth": 100
 }

.vscode/launch.json

@@ -5,7 +5,18 @@
             "name": "Test",
             "request": "launch",
             "command": "npm run test",
-            "cwd": "${workspaceFolder}"
+            "cwd": "${workspaceFolder}",
+        },
+        {
+            "type": "node-terminal",
+            "name": "Example",
+            "request": "launch",
+            "command": "npm run run-example",
+            "cwd": "${workspaceFolder}",
+            "skipFiles": [
+                "${workspaceFolder}/node_modules/**/*",
+                "<node_internals>/**/*"
+            ]
         },
     ]
 }

package-lock.json

@@ -22,7 +22,6 @@
         "@types/chai": "^4.2.15",
         "@types/chai-as-promised": "^7.1.3",
         "@types/inquirer": "^7.3.1",
-        "@types/lodash": "^4.14.168",
         "@types/luxon": "^1.25.2",
         "@types/mocha": "^8.2.1",
         "@types/node": "^14.14.27",
@@ -519,12 +518,6 @@
       "integrity": "sha512-cxWFQVseBm6O9Gbw1IWb8r6OS4OhSt3hPZLkFApLjM8TEXROBuQGLAH2i2gZpcXdLBIrpXuTDhH7Vbm1iXmNGA==",
       "dev": true
     },
-    "node_modules/@types/lodash": {
-      "version": "4.14.168",
-      "resolved": "https://registry.npmjs.org/@types/lodash/-/lodash-4.14.168.tgz",
-      "integrity": "sha512-oVfRvqHV/V6D1yifJbVRU3TMp8OT6o6BG+U9MkwuJ3U8/CsDHvalRpsxBqivn71ztOFZBTfJMvETbqHiaNSj7Q==",
-      "dev": true
-    },
     "node_modules/@types/luxon": {
       "version": "1.25.2",
       "resolved": "https://registry.npmjs.org/@types/luxon/-/luxon-1.25.2.tgz",
@@ -5482,12 +5475,6 @@
       "integrity": "sha512-cxWFQVseBm6O9Gbw1IWb8r6OS4OhSt3hPZLkFApLjM8TEXROBuQGLAH2i2gZpcXdLBIrpXuTDhH7Vbm1iXmNGA==",
       "dev": true
     },
-    "@types/lodash": {
-      "version": "4.14.168",
-      "resolved": "https://registry.npmjs.org/@types/lodash/-/lodash-4.14.168.tgz",
-      "integrity": "sha512-oVfRvqHV/V6D1yifJbVRU3TMp8OT6o6BG+U9MkwuJ3U8/CsDHvalRpsxBqivn71ztOFZBTfJMvETbqHiaNSj7Q==",
-      "dev": true
-    },
     "@types/luxon": {
       "version": "1.25.2",
       "resolved": "https://registry.npmjs.org/@types/luxon/-/luxon-1.25.2.tgz",

package.json

@@ -48,7 +48,6 @@
     "@types/chai": "^4.2.15",
     "@types/chai-as-promised": "^7.1.3",
     "@types/inquirer": "^7.3.1",
-    "@types/lodash": "^4.14.168",
     "@types/luxon": "^1.25.2",
     "@types/mocha": "^8.2.1",
     "@types/node": "^14.14.27",

src/example.ts

@@ -60,11 +60,7 @@ async function main() {
 
   // Log in the platform
   console.log("Authenticating...");
-  const result = await login(
+  const result = await login(process.env.F95_USERNAME, process.env.F95_PASSWORD, insert2faCode);
-    process.env.F95_USERNAME,
-    process.env.F95_PASSWORD,
-    insert2faCode
-  );
   console.log(`Authentication result: ${result.message}\n`);
 
   // Manage failed login
@@ -87,9 +83,7 @@ async function main() {
   latestQuery.includedTags = ["3d game"];
 
   const latestUpdates = await getLatestUpdates<Game>(latestQuery, 1);
-  console.log(
+  console.log(`"${latestUpdates.shift().name}" was the last "3d game" tagged game to be updated\n`);
-    `"${latestUpdates.shift().name}" was the last "3d game" tagged game to be updated\n`
-  );
 
   // Get game data
   for (const gamename of gameList) {

src/scripts/classes/handiwork/handiwork.ts

@@ -6,14 +6,7 @@
 "use strict";
 
 // Modules from files
-import {
+import { TAuthor, TRating, IHandiwork, TEngine, TCategory, TStatus } from "../../interfaces";
-  TAuthor,
-  TRating,
-  IHandiwork,
-  TEngine,
-  TCategory,
-  TStatus
-} from "../../interfaces";
 
 /**
  * It represents a generic work, be it a game, a comic, an animation or an asset.

src/scripts/classes/query/handiwork-search-query.ts

@@ -96,8 +96,7 @@ export default class HandiworkSearchQuery implements IQuery {
     // If the keywords are set or the number
     // of included tags is greather than 5,
     // we must perform a thread search
-    if (this.keywords || this.includedTags.length > MAX_TAGS_LATEST_SEARCH)
+    if (this.keywords || this.includedTags.length > MAX_TAGS_LATEST_SEARCH) return "thread";
-      return "thread";
 
     return DEFAULT_SEARCH_TYPE;
   }

src/scripts/classes/query/thread-search-query.ts

@@ -130,8 +130,7 @@ export default class ThreadSearchQuery implements IQuery {
     if (this.excludedTags) params["c[excludeTags]"] = this.excludedTags.join(",");
 
     // Set minimum reply number
-    if (this.minimumReplies > 0)
+    if (this.minimumReplies > 0) params["c[min_reply_count]"] = this.minimumReplies.toString();
-      params["c[min_reply_count]"] = this.minimumReplies.toString();
 
     // Add prefixes
     const parser = new PrefixParser();

src/scripts/classes/session.ts

@@ -210,8 +210,8 @@ export default class Session {
 
     // Search for expired cookies
     const jarValid =
-      this._cookieJar.getCookiesSync("https://f95zone.to").filter((el) => el.TTL() === 0)
+      this._cookieJar.getCookiesSync("https://f95zone.to").filter((el) => el.TTL() === 0).length ===
-        .length === 0;
+      0;
 
     return dateValid && hashValid && jarValid;
   }

src/scripts/constants/css-selector.ts

@@ -154,7 +154,15 @@ export const POST = {
    *
    * For use within a `THREAD.POSTS_IN_PAGE` selector.
    */
-  BOOKMARKED: '* ul.message-attribution-opposite >li > a[title="Bookmark"].is-bookmarked'
+  BOOKMARKED: '* ul.message-attribution-opposite >li > a[title="Bookmark"].is-bookmarked',
+  /**
+   * Button used to hide/show a spoiler element of a post.
+   */
+  SPOILER_BUTTON: "button.bbCodeSpoiler-button",
+  /**
+   * Contents of a spoiler element in a post.
+   */
+  SPOILER_CONTENT: "div.bbCodeSpoiler-content > div.bbCodeBlock--spoiler > div.bbCodeBlock-content"
 };
 
 export const MEMBER = {
@@ -205,8 +213,7 @@ export const MEMBER = {
    * If the text is `Unfollow` then the user is followed.
    * If the text is `Follow` then the user is not followed.
    */
-  FOLLOWED:
+  FOLLOWED: "div.memberHeader-buttons > div.buttonGroup:first-child > a[data-sk-follow] > span",
-    "div.memberHeader-buttons > div.buttonGroup:first-child > a[data-sk-follow] > span",
   /**
    * Button used to ignore/unignore the user.
    *

src/scripts/interfaces.ts

@@ -83,10 +83,7 @@ export type TCategory = "games" | "mods" | "comics" | "animations" | "assets";
 /**
  * Valid names of classes that implement the IQuery interface.
  */
-export type TQueryInterface =
+export type TQueryInterface = "LatestSearchQuery" | "ThreadSearchQuery" | "HandiworkSearchQuery";
-  | "LatestSearchQuery"
-  | "ThreadSearchQuery"
-  | "HandiworkSearchQuery";
 
 /**
  * Collection of values defined for each

src/scripts/network-helper.ts

@@ -86,9 +86,7 @@ export async function fetchHTML(
       error: null
     });
 
-    return isHTML
+    return isHTML ? success(response.value.data as string) : failure(unexpectedResponseError);
-      ? success(response.value.data as string)
-      : failure(unexpectedResponseError);
   } else return failure(response.value as GenericAxiosError);
 }
 
@@ -105,8 +103,7 @@ export async function authenticate(
   force: boolean = false
 ): Promise<LoginResult> {
   shared.logger.info(`Authenticating with user ${credentials.username}`);
-  if (!credentials.token)
+  if (!credentials.token) throw new InvalidF95Token(`Invalid token for auth: ${credentials.token}`);
-    throw new InvalidF95Token(`Invalid token for auth: ${credentials.token}`);
 
   // Secure the URL
   const secureURL = enforceHttpsUrl(urls.LOGIN);
@@ -216,9 +213,7 @@ export async function fetchGETResponse(
     const response = await axios.get(secureURL, commonConfig);
     return success(response);
   } catch (e) {
-    shared.logger.error(
+    shared.logger.error(`(GET) Error ${e.message} occurred while trying to fetch ${secureURL}`);
-      `(GET) Error ${e.message} occurred while trying to fetch ${secureURL}`
-    );
     const genericError = new GenericAxiosError({
       id: 1,
       message: `(GET) Error ${e.message} occurred while trying to fetch ${secureURL}`,
@@ -305,10 +300,7 @@ export function isStringAValidURL(url: string): boolean {
  * If `true`, the function will consider redirects a violation and return `false`.
  * Default: `false`
  */
-export async function urlExists(
+export async function urlExists(url: string, checkRedirect: boolean = false): Promise<boolean> {
-  url: string,
-  checkRedirect: boolean = false
-): Promise<boolean> {
   // Local variables
   let valid = false;
 
@@ -376,10 +368,7 @@ function manageLoginPOSTResponse(response: AxiosResponse<any>) {
   }
 
   // Get the error message (if any) and remove the new line chars
-  const errorMessage = $("body")
+  const errorMessage = $("body").find(GENERIC.LOGIN_MESSAGE_ERROR).text().replace(/\n/g, "");
-    .find(GENERIC.LOGIN_MESSAGE_ERROR)
-    .text()
-    .replace(/\n/g, "");
 
   // Return the result of the authentication
   const result = errorMessage.trim() === "";

src/scripts/scrape-data/handiwork-parse.ts

@@ -122,10 +122,7 @@ function stringToBoolean(s: string): boolean {
  *
  * Case-insensitive.
  */
-function getPostElementByName(
+function getPostElementByName(elements: IPostElement[], name: string): IPostElement | undefined {
-  elements: IPostElement[],
-  name: string
-): IPostElement | undefined {
   return elements.find((el) => el.name.toUpperCase() === name.toUpperCase());
 }
 
@@ -162,8 +159,7 @@ function fillWithPrefixes(hw: HandiWork, prefixes: string[]) {
 
     // Check what the prefix indicates
     if (stringInDict(prefix, shared.prefixes["engines"])) engine = prefix as TEngine;
-    else if (stringInDict(prefix, shared.prefixes["statuses"]))
+    else if (stringInDict(prefix, shared.prefixes["statuses"])) status = prefix as TStatus;
-      status = prefix as TStatus;
     else if (stringInDict(prefix, fakeModDict)) mod = true;
 
     // Anyway add the prefix to list
@@ -206,8 +202,7 @@ function fillWithPostData(hw: HandiWork, elements: IPostElement[]) {
 
   // Parse the censorship
   const censored =
-    getPostElementByName(elements, "censored") ||
+    getPostElementByName(elements, "censored") || getPostElementByName(elements, "censorship");
-    getPostElementByName(elements, "censorship");
   if (censored) hw.censored = stringToBoolean(censored.text);
 
   // Get the genres
@@ -249,8 +244,7 @@ function fillWithPostData(hw: HandiWork, elements: IPostElement[]) {
   //#region Get the changelog
   hw.changelog = [];
   const changelogElement =
-    getPostElementByName(elements, "changelog") ||
+    getPostElementByName(elements, "changelog") || getPostElementByName(elements, "change-log");
-    getPostElementByName(elements, "change-log");
   if (changelogElement) {
     const changelogSpoiler = changelogElement?.content.find((el) => {
       return el.type === "Spoiler" && el.content.length > 0;

src/scripts/scrape-data/json-ld.ts

@@ -59,9 +59,7 @@ function parseJSONLD(element: cheerio.Element): TJsonLD {
   const html = cheerio(element).html().trim();
 
   // Obtain the JSON-LD
-  const data = html
+  const data = html.replace('<script type="application/ld+json">', "").replace("</script>", "");
-    .replace('<script type="application/ld+json">', "")
-    .replace("</script>", "");
 
   // Convert the string to an object
   return JSON.parse(data);

src/scripts/scrape-data/post-parse.ts

@@ -5,6 +5,9 @@
 
 "use strict";
 
+// Import from files
+import { POST } from "../constants/css-selector";
+
 //#region Interfaces
 
 export interface IPostElement {
@@ -22,13 +25,11 @@ export interface ILink extends IPostElement {
 //#endregion Interfaces
 
 //#region Public methods
+
 /**
  * Given a post of a thread page it extracts the information contained in the body.
  */
-export function parseF95ThreadPost(
+export function parseF95ThreadPost($: cheerio.Root, post: cheerio.Cheerio): IPostElement[] {
-  $: cheerio.Root,
-  post: cheerio.Cheerio
-): IPostElement[] {
   // The data is divided between "tag" and "text" elements.
   // Simple data is composed of a "tag" element followed
   // by a "text" element, while more complex data (contained
@@ -40,34 +41,81 @@ export function parseF95ThreadPost(
   const elements = post
     .contents()
     .toArray()
-    .map((el) => parseCheerioNode($, el))
+    .map((el) => parseCheerioNode($, el)) // Parse the nodes
-    .filter((node) => node.name || node.text || node.content.length != 0);
+    .filter((el) => !isPostElementEmpty(el)) // Ignore the empty nodes
+    .map((el) => reducePostElement(el)); // Compress the nodes
 
   // ... then parse the elements to create the pairs of title/data
-  return parsePostElements(elements);
+  return associateElementsWithName(elements);
 }
+
 //#endregion Public methods
 
 //#region Private methods
 
+//#region Node type
+
+/**
+ * Check if the node passed as a parameter is a formatting one (i.e. `<b>`).
+ */
+function isFormattingNode(node: cheerio.Element): boolean {
+  const formattedTags = ["b", "i"];
+  return node.type === "tag" && formattedTags.includes(node.name);
+}
+
+/**
+ * Check if the node passed as a parameter is of text type.
+ */
+function isTextNode(node: cheerio.Element): boolean {
+  return node.type === "text";
+}
+
+/**
+ * Check if the node is a spoiler.
+ */
+function isSpoilerNode(node: cheerio.Cheerio): boolean {
+  return node.attr("class") === "bbCodeSpoiler";
+}
+
+/**
+ * Check if the node is a link or a image.
+ */
+function isLinkNode(node: cheerio.Element): boolean {
+  // Local variables
+  let valid = false;
+
+  // The node is a valid DOM element
+  if (node.type === "tag") {
+    const el = node as cheerio.TagElement;
+    valid = el.name === "a" || el.name === "img";
+  }
+
+  return valid;
+}
+
+/**
+ * Check if the node is a `noscript` tag.
+ */
+function isNoScriptNode(node: cheerio.Element): boolean {
+  return node.type === "tag" && node.name === "noscript";
+}
+
+//#endregion Node Type
+
+//#region Parse Cheerio node
+
 /**
  * Process a spoiler element by getting its text broken
  * down by any other spoiler elements present.
  */
-function parseCheerioSpoilerNode(
+function parseCheerioSpoilerNode($: cheerio.Root, node: cheerio.Cheerio): IPostElement {
-  $: cheerio.Root,
-  spoiler: cheerio.Cheerio
-): IPostElement {
   // A spoiler block is composed of a div with class "bbCodeSpoiler",
   // containing a div "bbCodeSpoiler-content" containing, in cascade,
   // a div with class "bbCodeBlock--spoiler" and a div with class "bbCodeBlock-content".
   // This last tag contains the required data.
 
   // Local variables
-  const BUTTON_CLASS = "button.bbCodeSpoiler-button";
+  const spoiler: IPostElement = {
-  const SPOILER_CONTENT_CLASS =
-    "div.bbCodeSpoiler-content > div.bbCodeBlock--spoiler > div.bbCodeBlock-content";
-  const content: IPostElement = {
     type: "Spoiler",
     name: "",
     text: "",
@@ -75,185 +123,219 @@ function parseCheerioSpoilerNode(
   };
 
   // Find the title of the spoiler (contained in the button)
-  const button = spoiler.find(BUTTON_CLASS).toArray().shift();
+  spoiler.name = node.find(POST.SPOILER_BUTTON).first().text().trim();
-  content.name = $(button).text().trim();
 
   // Parse the content of the spoiler
-  spoiler
+  spoiler.content = node
-    .find(SPOILER_CONTENT_CLASS)
+    .find(POST.SPOILER_CONTENT)
     .contents()
-    .map((idx, el) => {
+    .toArray()
-      // Convert the element
+    .map((el) => parseCheerioNode($, el));
-      const element = $(el);
-
-      // Parse nested spoiler
-      if (element.attr("class") === "bbCodeSpoiler") {
-        const spoiler = parseCheerioSpoilerNode($, element);
-        content.content.push(spoiler);
-      } else if (el.type === "text") {
-        // Append text
-        content.text += element.text();
-      }
-    });
 
   // Clean text
-  content.text = content.text.replace(/\s\s+/g, " ").trim();
+  spoiler.text = spoiler.text.replace(/\s\s+/g, " ").trim();
-  return content;
+  return spoiler;
 }
 
 /**
- * Check if the node passed as a parameter is of text type.
+ * Process a node that contains a link or image.
- * This also includes formatted nodes (i.e. `<b>`).
  */
-function isTextNode(node: cheerio.Element): boolean {
+function parseCheerioLinkNode(element: cheerio.Cheerio): ILink {
-  const formattedTags = ["b", "i"];
+  // Local variable
-  const isText = node.type === "text";
+  const link: ILink = {
-  const isFormatted = node.type === "tag" && formattedTags.includes(node.name);
+    type: "Link",
+    name: "",
+    text: "",
+    href: "",
+    content: []
+  };
 
-  return isText || isFormatted;
+  if (element.is("img")) {
+    link.type = "Image";
+    link.text = element.attr("alt");
+    link.href = element.attr("data-src");
+  } else if (element.is("a")) {
+    link.type = "Link";
+    link.text = element.text().replace(/\s\s+/g, " ").trim();
+    link.href = element.attr("href");
   }
 
+  return link;
+}
+
+/**
+ * Process a text only node.
+ */
+function parseCheerioTextNode(node: cheerio.Cheerio): IPostElement {
+  const content: IPostElement = {
+    type: "Text",
+    name: "",
+    text: getCheerioNonChildrenText(node),
+    content: []
+  };
+  return content;
+}
+
+//#endregion Parse Cheerio node
+
+//#region IPostElement utility
+
+/**
+ * Check if the node has non empty `name` and `text`.
+ */
+function isPostElementUnknown(node: IPostElement): boolean {
+  return node.name.trim() === "" && node.text.trim() === "";
+}
+
+/**
+ * Check if the node has a non empty property
+ * between `name`, `text` and `content`.
+ */
+function isPostElementEmpty(node: IPostElement): boolean {
+  return node.content.length === 0 && isPostElementUnknown(node);
+}
+
+/**
+ * Create a `IPostElement` without name, text or content.
+ */
+function createEmptyElement(): IPostElement {
+  return {
+    type: "Empty",
+    name: "",
+    text: "",
+    content: []
+  };
+}
+
+/**
+ * Check if the element contains the overview of a thread (post #1).
+ */
+function elementIsOverview(element: IPostElement): boolean {
+  // Search the text element that start with "overview"
+  const result = element.content
+    .filter((e) => e.type === "Text")
+    .find((e) => e.text.toUpperCase().startsWith("OVERVIEW"));
+  return result !== undefined;
+}
+
+/**
+ * If the element contains the overview of a thread, parse it.
+ */
+function getOverviewFromElement(element: IPostElement): string {
+  // Local variables
+  const alphanumericRegex = new RegExp("[a-zA-Z0-9]+");
+
+  // Get all the text values of the overview
+  const textes = element.content
+    .filter((e) => e.type === "Text")
+    .filter((e) => {
+      const cleanValue = e.text.toUpperCase().replace("OVERVIEW", "").trim();
+      const isAlphanumeric = alphanumericRegex.test(cleanValue);
+
+      return cleanValue !== "" && isAlphanumeric;
+    })
+    .map((e) => e.text);
+
+  // Joins the textes
+  return textes.join(" ");
+}
+
+//#endregion IPostElement utility
+
 /**
  * Gets the text of the node only, excluding child nodes.
  * Also includes formatted text elements (i.e. `<b>`).
  */
 function getCheerioNonChildrenText(node: cheerio.Cheerio): string {
+  // Local variable
+  let text = "";
+
+  // If the node has no children, return the node's text
+  if (node.contents().length === 1) {
+    // @todo Remove IF after cheerio RC6
+    text = node.text();
+  } else {
     // Find all the text nodes in the node
-  const text = node
+    text = node
       .first()
-    .contents()
+      .contents() // @todo Change to children() after cheerio RC6
-    .filter((idx, el) => {
+      .filter((idx, el) => isTextNode(el))
-      return isTextNode(el);
-    })
       .text();
+  }
 
   // Clean and return the text
   return text.replace(/\s\s+/g, " ").trim();
 }
 
-/**
- * Process a node and see if it contains a
- * link or image. If not, it returns `null`.
- */
-function parseCheerioLinkNode(element: cheerio.Cheerio): ILink | null {
-  //@ts-ignore
-  const name = element[0]?.name;
-  const link: ILink = {
-    name: "",
-    type: "Link",
-    text: "",
-    href: "",
-    content: []
-  };
-
-  if (name === "img") {
-    link.type = "Image";
-    link.text = element.attr("alt");
-    link.href = element.attr("data-src");
-  } else if (name === "a") {
-    link.type = "Link";
-    link.text = element.text().replace(/\s\s+/g, " ").trim();
-    link.href = element.attr("href");
-  }
-
-  return link.href ? link : null;
-}
-
 /**
  * Collapse an `IPostElement` element with a single subnode
  * in the `Content` field in case it has no information.
  */
-function reducePostElement(element: IPostElement): IPostElement {
+function reducePostElement(element: IPostElement, recursive = true): IPostElement {
-  if (element.content.length === 1) {
+  // Local variables
-    const content = element.content[0] as IPostElement;
+  const shallowCopy = Object.assign({}, element);
-    const nullValues =
-      (!element.name || !content.name) && (!element.text || !content.text);
-    const sameValues = element.name === content.name || element.text === content.text;
 
-    if (nullValues || sameValues) {
+  // Find the posts without name and text
-      element.name = element.name || content.name;
+  const unknownChildrens = shallowCopy.content.filter((e) => isPostElementUnknown(e));
-      element.text = element.text || content.text;
+  if (recursive) {
-      element.content.push(...content.content);
+    const recursiveUnknownChildrens = unknownChildrens.map((e) => reducePostElement(e));
-      element.type = content.type;
+    unknownChildrens.push(...recursiveUnknownChildrens);
-
-      // If the content is a link, add the HREF to the element
-      const contentILink = content as ILink;
-      const elementILink = element as ILink;
-      if (contentILink.href) elementILink.href = contentILink.href;
-    }
   }
 
-  return element;
+  // Eliminate non-useful child nodes
+  if (isPostElementUnknown(shallowCopy) && unknownChildrens.length > 0) {
+    // Find the valid elements to add to the node
+    const childContents = unknownChildrens
+      .filter((e) => !shallowCopy.content.includes(e))
+      .map((e) => (e.content.length > 0 ? e.content : e));
+
+    // Remove the empty elements
+    shallowCopy.content = shallowCopy.content.filter((e) => !unknownChildrens.includes(e));
+
+    // Merge the non-empty children of this node with
+    // the content of the empty children of this node
+    const newContent = [].concat(...childContents);
+    shallowCopy.content.push(...newContent);
+  }
+  // If the node has only one child, return it
+  else if (isPostElementUnknown(shallowCopy) && shallowCopy.content.length === 1) {
+    return shallowCopy.content[0];
+  }
+  return shallowCopy;
 }
 
 /**
  * Transform a `cheerio.Cheerio` node into an `IPostElement` element with its subnodes.
- * @param reduce Compress subsequent subnodes if they contain no information. Default: `true`.
  */
-function parseCheerioNode(
+function parseCheerioNode($: cheerio.Root, node: cheerio.Element): IPostElement {
-  $: cheerio.Root,
-  node: cheerio.Element,
-  reduce = true
-): IPostElement {
   // Local variables
-  const content: IPostElement = {
+  let post: IPostElement = createEmptyElement();
-    type: "Empty",
-    name: "",
-    text: "",
-    content: []
-  };
   const cheerioNode = $(node);
 
-  if (isTextNode(node)) {
+  // Parse the node
-    content.text = cheerioNode.text().replace(/\s\s+/g, " ").trim();
+  if (!isNoScriptNode(node)) {
-    content.type = "Text";
+    if (isTextNode(node) && !isFormattingNode(node)) post = parseCheerioTextNode(cheerioNode);
-  } else {
+    else if (isSpoilerNode(cheerioNode)) post = parseCheerioSpoilerNode($, cheerioNode);
-    // Get the number of children that the element own
+    else if (isLinkNode(node)) post = parseCheerioLinkNode(cheerioNode);
-    const nChildren = cheerioNode.children().length;
 
-    // Get the text of the element without childrens
+    // Parse the node's childrens
-    content.text = getCheerioNonChildrenText(cheerioNode);
+    const childPosts = cheerioNode
-
+      .contents() // @todo Change to children() after cheerio RC6
-    // Parse spoilers
+      .toArray()
-    if (cheerioNode.attr("class") === "bbCodeSpoiler") {
+      .filter((el) => el) // Ignore undefined elements
-      const spoiler = parseCheerioSpoilerNode($, cheerioNode);
+      .map((el) => parseCheerioNode($, el))
-
+      .filter((el) => !isPostElementEmpty(el));
-      // Add element if not null
+    post.content.push(...childPosts);
-      if (spoiler) {
-        content.content.push(spoiler);
-        content.type = "Spoiler";
-      }
-    }
-    // Parse links
-    else if (nChildren === 0 && cheerioNode.length != 0) {
-      const link = parseCheerioLinkNode(cheerioNode);
-
-      // Add element if not null
-      if (link) {
-        content.content.push(link);
-        content.type = "Link";
-      }
-    } else {
-      cheerioNode.children().map((idx, el) => {
-        // Parse the children of the element passed as parameter
-        const childElement = parseCheerioNode($, el);
-
-        // If the children is valid (not empty) push it
-        if ((childElement.text || childElement.content.length !== 0) && !isTextNode(el)) {
-          content.content.push(childElement);
-        }
-      });
-    }
   }
 
-  return reduce ? reducePostElement(content) : content;
+  return post;
 }
 
 /**
  * It simplifies the `IPostElement` elements by associating
  * the corresponding value to each characterizing element (i.e. author).
  */
-function parsePostElements(elements: IPostElement[]): IPostElement[] {
+function associateElementsWithName(elements: IPostElement[]): IPostElement[] {
   // Local variables
   const pairs: IPostElement[] = [];
   const specialCharsRegex = /^[-!$%^&*()_+|~=`{}[\]:";'<>?,./]/;
@@ -275,11 +357,11 @@ function parsePostElements(elements: IPostElement[]): IPostElement[] {
       lastPair.content.push(...elements[i].content);
     }
     // This is a special case
-    else if (elements[i].text.startsWith("Overview:\n")) {
+    else if (elementIsOverview(elements[i])) {
       // We add the overview to the pairs as a text element
       elements[i].type = "Text";
       elements[i].name = "Overview";
-      elements[i].text = elements[i].text.replace("Overview:\n", "");
+      elements[i].text = getOverviewFromElement(elements[i]);
       pairs.push(elements[i]);
     }
     // We have an element referred to the previous "title"

test/classes/mapping/thread.ts

@@ -34,8 +34,6 @@ export function suite(): void {
   it("Fetch post with invalid ID", async function fetchWithInvalidID() {
     Shared.setIsLogged(true);
     const thread = new Thread(-1);
-    await expect(thread.getPost(0)).to.be.rejectedWith(
+    await expect(thread.getPost(0)).to.be.rejectedWith("Index must be greater or equal than 1");
-      "Index must be greater or equal than 1"
-    );
   });
 }

test/classes/prefix-parser.ts

@@ -29,15 +29,7 @@ export function suite(): void {
 
     // Test values
     const testIDs = [103, 225, 44, 13, 2, 7, 22];
-    const testPrefixes = [
+    const testPrefixes = ["corruption", "pregnancy", "slave", "VN", "RPGM", "Ren'Py", "Abandoned"];
-      "corruption",
-      "pregnancy",
-      "slave",
-      "VN",
-      "RPGM",
-      "Ren'Py",
-      "Abandoned"
-    ];
 
     // Parse values
     const ids = parser.prefixesToIDs(testPrefixes);