Utility functions for Any Hackernews Link
当前为
此脚本不应直接安装。它是供其他脚本使用的外部库，要使用该库请加入元指令 // @require https://update.gf.qytechs.cn/scripts/524693/1525919/Any%20Hackernews%20Link%20Utils.js
// ==UserScript==
// @name Any Hackernews Link Utils
// @namespace http://tampermonkey.net/
// @version 0.1.0
// @description Utility functions for Any Hackernews Link
// @author RoCry
// @grant GM_xmlhttpRequest
// @connect hn.algolia.com
// @license MIT
// ==/UserScript==
/**
* Configuration
*/
const CONFIG = {
// Additional domains to ignore that couldn't be handled by @exclude
IGNORED_DOMAINS: [
'gmail.com',
'accounts.google.com',
'accounts.youtube.com',
'signin.',
'login.',
'auth.',
'oauth.',
],
// Patterns that indicate a search page
SEARCH_PATTERNS: [
'/search',
'/webhp',
'/results',
'?q=',
'?query=',
'?search=',
'?s='
],
// URL parameters to remove during normalization
TRACKING_PARAMS: [
'utm_source',
'utm_medium',
'utm_campaign',
'utm_term',
'utm_content',
'fbclid',
'gclid',
'_ga',
'ref',
'source'
],
// Minimum ratio of ASCII characters to consider content as English
MIN_ASCII_RATIO: 0.9,
// Number of characters to check for language detection
CHARS_TO_CHECK: 300
};
/**
* URL Utilities
*/
const URLUtils = {
/**
* Check if a URL should be ignored based on domain or search patterns
* @param {string} url - URL to check
* @returns {boolean} - True if URL should be ignored
*/
shouldIgnoreUrl(url) {
try {
const urlObj = new URL(url);
// Check remaining ignored domains
if (CONFIG.IGNORED_DOMAINS.some(domain => urlObj.hostname.includes(domain))) {
return true;
}
// Check if it's a search page
if (CONFIG.SEARCH_PATTERNS.some(pattern =>
urlObj.pathname.includes(pattern) || urlObj.search.includes(pattern))) {
return true;
}
return false;
} catch (e) {
console.error('Error checking URL:', e);
return false;
}
},
/**
* Normalize URL by removing tracking parameters and standardizing format
* @param {string} url - URL to normalize
* @returns {string} - Normalized URL
*/
normalizeUrl(url) {
try {
const urlObj = new URL(url);
// Remove tracking parameters
CONFIG.TRACKING_PARAMS.forEach(param => urlObj.searchParams.delete(param));
// Remove sepecial parameter for all hosts
// https://github.com/HackerNews/API?tab=readme-ov-file -> https://github.com/HackerNews/API
urlObj.searchParams.delete('tab');
// Handle GitHub repository paths
if (urlObj.hostname === 'github.com') {
// Split path into segments
const pathSegments = urlObj.pathname.split('/').filter(Boolean);
// Only process if we have at least username/repo
if (pathSegments.length >= 2) {
const [username, repo, ...rest] = pathSegments;
// If path contains tree/master, blob/master, or similar, remove them
if (rest.length > 0 && (rest[0] === 'tree' || rest[0] === 'blob')) {
urlObj.pathname = `/${username}/${repo}`;
}
}
}
// for arxiv
// https://arxiv.org/pdf/1706.03762 -> https://arxiv.org/abs/1706.03762
if (urlObj.hostname === 'arxiv.org') {
urlObj.pathname = urlObj.pathname.replace('/pdf/', '/abs/');
}
// Remove hash
urlObj.hash = '';
// Remove trailing slash for consistency
let normalizedUrl = urlObj.toString();
if (normalizedUrl.endsWith('/')) {
normalizedUrl = normalizedUrl.slice(0, -1);
}
return normalizedUrl;
} catch (e) {
console.error('Error normalizing URL:', e);
return url;
}
},
/**
* Compare two URLs for equality after normalization
* @param {string} url1 - First URL
* @param {string} url2 - Second URL
* @returns {boolean} - True if URLs match
*/
urlsMatch(url1, url2) {
try {
const u1 = new URL(this.normalizeUrl(url1));
const u2 = new URL(this.normalizeUrl(url2));
return u1.hostname.toLowerCase() === u2.hostname.toLowerCase() &&
u1.pathname.toLowerCase() === u2.pathname.toLowerCase() &&
u1.search === u2.search;
} catch (e) {
console.error('Error comparing URLs:', e);
return false;
}
}
};
/**
* Content Utilities
*/
const ContentUtils = {
/**
* Check if text is primarily English by checking ASCII ratio
* @param {string} text - Text to analyze
* @returns {boolean} - True if content is likely English
*/
isEnglishContent() {
try {
// Get text from title and first paragraph or relevant content
const title = document.title || '';
const firstParagraphs = Array.from(document.getElementsByTagName('p'))
.slice(0, 3)
.map(p => p.textContent)
.join(' ');
const textToAnalyze = (title + ' ' + firstParagraphs)
.slice(0, CONFIG.CHARS_TO_CHECK)
.replace(/\s+/g, ' ')
.trim();
if (!textToAnalyze) return true; // If no text found, assume English
// Count ASCII characters (excluding spaces and common punctuation)
const asciiChars = textToAnalyze.replace(/[\s\.,\-_'"!?()]/g, '')
.split('')
.filter(char => char.charCodeAt(0) <= 127).length;
const totalChars = textToAnalyze.replace(/[\s\.,\-_'"!?()]/g, '').length;
if (totalChars === 0) return true;
const asciiRatio = asciiChars / totalChars;
console.log('🈂️ ASCII Ratio:', asciiRatio.toFixed(2));
return asciiRatio >= CONFIG.MIN_ASCII_RATIO;
} catch (e) {
console.error('Error checking content language:', e);
return true; // Default to allowing English in case of error
}
}
};
/**
* HackerNews API Handler
*/
const HNApi = {
/**
* Search for a URL on HackerNews
* @param {string} normalizedUrl - URL to search for
* @param {Function} updateUI - Callback function to update UI with results
*/
checkHackerNews(normalizedUrl, updateUI) {
const apiUrl = `https://hn.algolia.com/api/v1/search?query=${encodeURIComponent(normalizedUrl)}&restrictSearchableAttributes=url`;
GM_xmlhttpRequest({
method: 'GET',
url: apiUrl,
onload: (response) => this.handleApiResponse(response, normalizedUrl, updateUI),
onerror: (error) => {
console.error('Error fetching from HN API:', error);
updateUI(null);
}
});
},
/**
* Handle the API response
* @param {Object} response - API response
* @param {string} normalizedUrl - Original normalized URL
* @param {Function} updateUI - Callback function to update UI with results
*/
handleApiResponse(response, normalizedUrl, updateUI) {
try {
const data = JSON.parse(response.responseText);
const matchingHits = data.hits.filter(hit => URLUtils.urlsMatch(hit.url, normalizedUrl));
if (matchingHits.length === 0) {
console.log('🔍 URL not found on Hacker News');
updateUI(null);
return;
}
const topHit = matchingHits.sort((a, b) => (b.points || 0) - (a.points || 0))[0];
const result = {
title: topHit.title,
points: topHit.points || 0,
comments: topHit.num_comments || 0,
link: `https://news.ycombinator.com/item?id=${topHit.objectID}`,
posted: new Date(topHit.created_at).toLocaleDateString()
};
console.log('📰 Found on Hacker News:', result);
updateUI(result);
} catch (e) {
console.error('Error parsing HN API response:', e);
updateUI(null);
}
}
};