chimi-scraper/dist/providers/games/itch.js

A TypeScript library for scraping game data from itch.io with a clean, scalable architecture

"use strict";
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    var desc = Object.getOwnPropertyDescriptor(m, k);
    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
      desc = { enumerable: true, get: function() { return m[k]; } };
    }
    Object.defineProperty(o, k2, desc);
}) : (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    o[k2] = m[k];
}));
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
    Object.defineProperty(o, "default", { enumerable: true, value: v });
}) : function(o, v) {
    o["default"] = v;
});
var __importStar = (this && this.__importStar) || (function () {
    var ownKeys = function(o) {
        ownKeys = Object.getOwnPropertyNames || function (o) {
            var ar = [];
            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
            return ar;
        };
        return ownKeys(o);
    };
    return function (mod) {
        if (mod && mod.__esModule) return mod;
        var result = {};
        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
        __setModuleDefault(result, mod);
        return result;
    };
})();
Object.defineProperty(exports, "__esModule", { value: true });
exports.ItchIO = void 0;
const cheerio = __importStar(require("cheerio"));
const game_parser_1 = require("../../models/game-parser");
const utils_1 = require("../../utils");
class ItchIO extends game_parser_1.GameParser {
    constructor() {
        super('https://itch.io', 'ItchIO', 'GAMES');
        this.httpClient = new utils_1.HttpClient(this.baseUrl);
    }
    async search(query, page = 1) {
        try {
            const searchUrl = `/search?q=${encodeURIComponent(query)}&page=${page}`;
            const html = await this.httpClient.get(searchUrl);
            const $ = cheerio.load(html);
            const results = [];
            $('.game_cell').each((_, element) => {
                const gameResult = this.parseGameCell($, element);
                if (gameResult) {
                    results.push(gameResult);
                }
            });
            const pagination = this.parsePagination($);
            return {
                currentPage: page,
                results,
                ...pagination
            };
        }
        catch (error) {
            throw new Error(`Failed to search games: ${error}`);
        }
    }
    async fetchNewAndPopular(page = 1) {
        return this.fetchGamesByCategory('new-and-popular', page);
    }
    async fetchTopSellers(page = 1) {
        return this.fetchGamesByCategory('top-sellers', page);
    }
    async fetchTopRated(page = 1) {
        return this.fetchGamesByCategory('top-rated', page);
    }
    async fetchNewest(page = 1) {
        return this.fetchGamesByCategory('newest', page);
    }
    async fetchGameInfo(gameUrl) {
        try {
            const url = gameUrl.startsWith('http') ? gameUrl : `${this.baseUrl}${gameUrl}`;
            const html = await this.httpClient.get(url);
            const $ = cheerio.load(html);
            const title = $('.game_title').first().text().trim() || $('h1').first().text().trim();
            const description = $('.formatted_description, .user_formatted').first().text().trim();
            // Extract basic info
            const developer = this.extractDeveloper($);
            const cover = this.extractCoverImage($);
            const platforms = this.extractPlatforms($);
            const genres = this.extractGenres($);
            const tags = this.extractTags($);
            // Extract pricing info
            const { price, isFree, isOnSale, originalPrice } = this.extractPricingInfo($);
            // Extract media
            const screenshots = this.extractScreenshots($);
            const videos = this.extractVideos($);
            // Extract ratings
            const { rating, ratingCount } = this.extractRatingInfo($);
            // Extract release date
            const releaseDate = $('.game_info_panel_widget abbr, .release_date abbr').attr('title');
            return {
                id: utils_1.ParserUtils.extractIdFromUrl(url),
                title,
                url,
                cover,
                description,
                developer,
                platforms,
                genres,
                tags,
                price,
                isFree,
                isOnSale,
                originalPrice,
                rating,
                ratingCount,
                releaseDate,
                // Additional metadata
                screenshots: screenshots.map(s => s.url),
                videos: videos.map(v => v.url)
            };
        }
        catch (error) {
            throw new Error(`Failed to fetch game info: ${error}`);
        }
    }
    async fetchGamesByCategory(category, page) {
        try {
            const categoryUrl = `/games/${category}${page > 1 ? `?page=${page}` : ''}`;
            const html = await this.httpClient.get(categoryUrl);
            const $ = cheerio.load(html);
            const results = [];
            $('.game_cell, .game_cell_data').each((_, element) => {
                const gameResult = this.parseGameCell($, element);
                if (gameResult) {
                    results.push(gameResult);
                }
            });
            const pagination = this.parsePagination($);
            return {
                currentPage: page,
                results,
                ...pagination
            };
        }
        catch (error) {
            throw new Error(`Failed to fetch ${category} games: ${error}`);
        }
    }
    parseGameCell($, element) {
        try {
            const $game = $(element);
            const title = $game.find('.game_title, .title').first().text().trim();
            const gameLink = $game.find('.game_link, a').first().attr('href');
            const url = gameLink ? utils_1.ParserUtils.toAbsoluteUrl(gameLink, this.baseUrl) : '';
            if (!title || !url) {
                return null;
            }
            const image = $game.find('.game_thumb img, img').first().attr('src');
            const priceText = $game.find('.price, .game_price').first().text().trim();
            const price = utils_1.ParserUtils.parsePrice(priceText);
            const isFree = !price;
            const platforms = utils_1.ParserUtils.parsePlatforms($, element);
            const developer = $game.find('.game_author, .author').first().text().trim();
            return {
                id: utils_1.ParserUtils.extractIdFromUrl(url),
                title,
                url,
                image,
                price,
                isFree,
                platforms,
                developer
            };
        }
        catch (error) {
            console.warn('Failed to parse game cell:', error);
            return null;
        }
    }
    parsePagination($) {
        const paginationText = $('.pager_label').text();
        const match = paginationText.match(/Page (\d+) of (\d+)/);
        if (match) {
            const currentPage = parseInt(match[1]);
            const totalPages = parseInt(match[2]);
            return {
                hasNextPage: currentPage < totalPages,
                totalPages
            };
        }
        // Check for next page link
        const hasNextPage = $('.next_page').length > 0;
        return { hasNextPage };
    }
    extractDeveloper($) {
        return $('.game_author a, .user_name').first().text().trim();
    }
    extractCoverImage($) {
        return $('.game_thumb img, .header_image img, .cover_image img').first().attr('src');
    }
    extractPlatforms($) {
        const platforms = [];
        if ($('.icon-windows, .fa-windows').length > 0)
            platforms.push('Windows');
        if ($('.icon-apple, .fa-apple').length > 0)
            platforms.push('macOS');
        if ($('.icon-linux, .fa-linux').length > 0)
            platforms.push('Linux');
        if ($('.icon-android, .fa-android').length > 0)
            platforms.push('Android');
        if ($('.icon-html5, .fa-html5').length > 0)
            platforms.push('Web');
        return platforms;
    }
    extractGenres($) {
        const genres = [];
        $('.game_genre, .genre_tag, .classification_tag').each((_, element) => {
            const genre = $(element).text().trim();
            if (genre)
                genres.push(genre);
        });
        return genres;
    }
    extractTags($) {
        const tags = [];
        $('.game_tag_link, .tag').each((_, element) => {
            const tag = $(element).text().trim();
            if (tag)
                tags.push(tag);
        });
        return tags;
    }
    extractPricingInfo($) {
        const priceElement = $('.buy_btn .price, .price').first();
        const priceText = priceElement.text().trim();
        let price = utils_1.ParserUtils.parsePrice(priceText);
        let isFree = !price;
        let isOnSale = false;
        let originalPrice;
        // Check for sale pricing
        const originalPriceElement = $('.original_price');
        if (originalPriceElement.length > 0) {
            const originalPriceText = originalPriceElement.text().trim();
            originalPrice = utils_1.ParserUtils.parsePrice(originalPriceText);
            if (originalPrice) {
                isOnSale = true;
            }
        }
        return { price, isFree, isOnSale, originalPrice };
    }
    extractScreenshots($) {
        const screenshots = [];
        $('.screenshot img, .screenshot_list img').each((_, element) => {
            const url = $(element).attr('src') || $(element).attr('data-src');
            if (url) {
                screenshots.push({ url });
            }
        });
        return screenshots;
    }
    extractVideos($) {
        const videos = [];
        // YouTube embeds
        $('iframe[src*="youtube.com"], iframe[src*="youtu.be"]').each((_, element) => {
            const url = $(element).attr('src');
            if (url) {
                videos.push({
                    url,
                    type: 'trailer'
                });
            }
        });
        // Vimeo embeds
        $('iframe[src*="vimeo.com"]').each((_, element) => {
            const url = $(element).attr('src');
            if (url) {
                videos.push({
                    url,
                    type: 'trailer'
                });
            }
        });
        return videos;
    }
    extractRatingInfo($) {
        const ratingText = $('.aggregate_rating, .rating_value').text();
        const rating = ratingText ? parseFloat(ratingText) : undefined;
        const ratingCountText = $('.rating_count').text();
        const ratingCount = utils_1.ParserUtils.parseNumber(ratingCountText);
        return { rating, ratingCount };
    }
}
exports.ItchIO = ItchIO;
//# sourceMappingURL=itch.js.map