// ==UserScript==
// @name NexusPHP 帖子评论收集
// @namespace http://tampermonkey.net/
// @version 0.2
// @description NexusPHP 帖子评论收集, 主要用于发药时过滤邀请人. 有问题请在 Github 反馈: https://github.com/IITII/nexusphp_forums
// @author IITII
// @require https://code.jquery.com/jquery-3.6.0.min.js
// @match https://**/forums.php*
// @icon https://avatars.githubusercontent.com/u/33705067?v=4
// @grant none
// @todo 做种条数, 本站 email, 用户被警告
// ==/UserScript==
// @require file:///D:/services/nexusphp_forums/nexusphp_comment.js
(async function () {
'use strict';
// setting
// 求药信息是否需要包含ID
const preIdIsRequired = true
// 求药ID 正则
const preIdReg = /注册(用户名|id)[：:]/ig
// 求药邮箱正则
const preEmailReg = /(邮箱|email)[：:]/ig
// 邮箱格式
const preEmailFormat = /[a-z0-9]@[a-z0-9]+\./ig
// 是否debug, debug下只获取当前页
const isDebug = false
// 是否获取用户做种大小
const getSeedingSize = false
// req break, lower break will reduce high load for server
const reqBreak = 1000
let href, curr_url, title, currPage, finalPage, rawComment = [], filterComArr = [], sep = '|', filterSep = ',', stopped = false
async function sleep(ms) {
console.log(`sleep ${ms}ms`);
return new Promise((resolve) => setTimeout(resolve, ms));
}
function sizeToGB(size, unit = 'mb', defUnit = 'gb') {
const singleUnit = 1000,
units = {
b: 1 / singleUnit,
kb: singleUnit,
mb: singleUnit * singleUnit,
gb: singleUnit * singleUnit * singleUnit,
tb: singleUnit * singleUnit * singleUnit * singleUnit,
pb: singleUnit * singleUnit * singleUnit * singleUnit * singleUnit,
}
unit = unit.trim()
unit = unit.toLowerCase()
unit = unit.replace('ib', 'b')
size = size.trim()
size = parseFloat(size)
let curUnit = units[unit]
if (!curUnit) {
throw new Error(`unknown unit: ${unit}`)
}
return +(size * curUnit / units[defUnit]).toFixed(2)
}
function sizeToGBWithBreak(str, withBreak = /\s+/) {
let [size, unit] = str.split(withBreak)
return sizeToGB(size, unit)
}
function sizeInfoFormat(sizeAndUnit) {
let arr = [sizeAndUnit.replace(/[ptgmk]?b/ig, ' '), sizeAndUnit.replace(/\d+\.?(\d+)?/ig, ' ')]
return arr.map(_ => _.trim())
}
/**
* 创建下载文件
* @param {String} fileName 文件名称
* @param {String} fileContent 文件内容
* @return {String}
*/
function createAndDownloadFile(fileName, fileContent) {
if (!fileName) {
return '文件名称为空';
}
if (typeof fileName !== 'string') {
return '文件名类型错误';
}
if (!fileContent) {
return '文件内容为空';
}
if (typeof fileContent !== 'string') {
fileContent = JSON.stringify(fileContent);
}
// 创建隐藏a标签
let aTag = document.createElement('a')
// 将文件内容转成blob对象
let blob = new Blob(["\ufeff" + fileContent], {type: 'text/csv,charset=UTF-8'})
// 设置下载文件名
aTag.download = fileName;
// 给a标签创建DOMString
aTag.href = URL.createObjectURL(blob)
// 模拟点击、下载
aTag.click()
// 释放DOMString
URL.revokeObjectURL(blob)
return ''
}
async function page(html = document) {
let jq = jQuery(html)
let userInfos, seedingSizes = [], userAddInfos, postBodys
userInfos = jq.find("table[id*='pid']").map(function () {
let $ = jQuery(this)
let floor, uid, userInfo, username, user_level, status, user_details_link, comment_time
floor = $.find("a:contains('#')").text()
// fix for pter 2FA
comment_time = $.find("span[title]:last").attr('title')
userInfo = $.find("span[class='nowrap']")
if (userInfo.text().match(/(無此帳戶|无此帐户)/)) {
status = 'banned'
username = 'banned'
user_level = 'Peasant_Name'
user_details_link = 'banned'
uid = -1
} else {
status = '正常'
username = userInfo.find("a[href*='userdetail']").text()
user_level = userInfo.find("a[href*='userdetail']").attr('class')
user_details_link = userInfo.find("a[href*='userdetail']")[0].href
uid = new URL(user_details_link).searchParams.get('id')
uid = +uid
}
return { floor, uid, username, user_level, status, user_details_link, comment_time }
}).toArray()
//
for (const info of userInfos) {
let size = 0
if (info.uid > 0) {
if (getSeedingSize) {
size = await seedingSize(info.uid)
await sleep(reqBreak)
} else {
size = '未获取'
}
}
seedingSizes.push(size)
}
userAddInfos = jq.find("td[class='rowfollow']:has(>img[alt='avatar'])").toArray()
.map(_ => _.innerText.replace(/([ptgmk]b)/ig, ' $1\n').replace(/(上传|做种积分)/ig, '\n$1').replace(/,/ig, ''))
postBodys = jq.find("div[id*='pid']").toArray().map(_ => _.innerText)
for (let i = 0; i < userInfos.length; i++) {
let { floor, uid, username, user_level, status, user_details_link, comment_time } = userInfos[i]
let seedingSize = seedingSizes[i], userAddInfo = userAddInfos[i], upload, download, shareRate,
postBody = postBodys[i], preId, preEmail
rawComment.push([floor, uid, username, user_level, status, user_details_link, comment_time, seedingSize,
userAddInfo, postBody].join(sep))
// if (i === 1) {
// debugger
// }
if (postBody.match(/引用/) || !postBody.match(preIdReg)|| !postBody.match(preEmailReg) || status !== '正常') {
console.log(`skip -> ${status} -> ${postBody}`)
continue
}
user_level = user_level.replace(/_Name/g, '')
upload = userAddInfo.split(/\n+/).find(_ => _.match(/上/))?.replace(/上[傳传][：:]/g, '')?.trim()
download = userAddInfo.split(/\n+/).find(_ => _.match(/下/))?.replace(/下[載载][：:]/g, '')?.trim()
shareRate = userAddInfo.split(/\n+/).find(_ => _.match(/分/))?.replace(/分享率[：:]/g, '')?.trim()
upload = sizeToGBWithBreak(upload)
download = sizeToGBWithBreak(download)
preId = postBody.split(/\n+/).find(_ => _.match(preIdReg))
if (!preId && preIdIsRequired) {
console.log(`skip -> ${postBody}`)
continue
}
preId = preId || ''
preId = preId.replace(preIdReg, ' ').trim().split(/\s+/).pop()
preEmail = postBody.split(/\n+/).filter(_ => _.match(preEmailReg)).find(_ => _.match(preEmailFormat))
.replace(preEmailReg, ' ').trim().split(/\s+/).pop()
filterComArr.push([floor, uid, username, user_level, status, seedingSize, comment_time,
upload, download, shareRate, preId, preEmail, user_details_link].join(filterSep))
}
}
function getSeedingUrl(uid, page) {
let url, idSel, sizeSel
idSel = "table[width='100%'] tr td[class='rowfollow']:nth-child(2) a"
sizeSel = "table[width='100%'] tr td[class='rowfollow']:nth-child(3)"
switch (window.location.hostname) {
case "pterclub.com":
url = `/getusertorrentlist.php?do_ajax=1&userid=${uid}&type=seeding&page=${page}`
sizeSel = "table[width='100%'] tr td[class='rowfollow']:nth-child(4)"
break;
case "xp.m-team.io":
url = `/getusertorrentlist.php?userid=${uid}&type=seeding&page=${i}`
break;
case "pt.2xfree.org":
default:
url = `/getusertorrentlistajax.php?userid=${uid}&type=seeding&page=${page}`
break;
}
return [url, idSel, sizeSel]
}
// let url = `/getusertorrentlistajax.php?do_ajax=1&userid=${uid}&type=seeding&page=${i}`
// 获取用户做种大小
async function seedingSize(uid, maxPage = 100000) {
let seedingCnt = 0
let seedingIds = [], seedingArr = [], seedingSize = 0, err = false, errText = ''
for (let i = 0; i < maxPage; i++) {
let [url, idSel, sizeSel] = getSeedingUrl(uid, i)
let rawHtml
try {
await jQuery.get(url).done(res => {
if (res.match(/上[傳传]/)) {
rawHtml = res
} else {
err = true
errText = res
console.log(`request to ${url} failed: ${res}`);
}
})
} catch (e) {
if (e.responseText.match(/上[傳传]/)) {
rawHtml = e.responseText
} else {
err = true
errText = e.responseText
console.log(`${e.statusText} -> ${e.responseText}`)
}
}
if (err) {
// return -1
return errText
}
let jq = jQuery(`<div id='jq'>${rawHtml}</div>`)
let totalSize = jq.find("div:contains('总大小')")
// let prePageSize = 0, pageSize = 100
if (totalSize.length === 0) {
let torrIds, torrents
torrIds = jq.find(idSel)
torrents = jq.find(sizeSel)
if (torrents.length === 0) {
// if (torrents.length === 0 || torrents.length < prePageSize) {
console.log(`maybe currPage: ${i} is the last`)
break
} else {
torrIds = torrIds.toArray().map(_ => _.href)
.map(_ => new URL(_).searchParams.get('id')).filter(_ => !!_)
torrents = torrents.toArray().map(_ => _.innerText)
let diff = []
for (let i = 0; i < torrIds.length; i++) {
const id = torrIds[i];
const torrent = torrents[i];
if (seedingIds.indexOf(id) > -1) {
console.log(`skip torrent id: ${id} ${torrent}`)
} else {
seedingIds.push(id)
diff.push(torrent)
}
}
if (diff.length == 0) {
console.log(`maybe curr seeding Page: ${i} is the last`)
break
} else {
seedingArr = seedingArr.concat(diff)
seedingCnt += diff.length
}
}
} else {
let seedingCnt = jq.find("div:contains('条记录'):first").text().split('|').shift().replace(/条记录[：:]/g, '').trim()
seedingCnt = parseInt(seedingCnt) || 0
let [size, unit] = jq.find("div:contains('总大小'):first").text().split('|').pop().replace(/总大小[：:]/g, '').trim().split(/\s+/)
totalSize = sizeToGB(size, unit)
seedingSize = totalSize
console.log(`${uid} seedingCnt -> ${seedingCnt}`);
return seedingSize
}
await sleep(reqBreak)
}
seedingSize = seedingArr.map(_ => sizeInfoFormat(_))
.map(([size, unit]) => sizeToGB(size, unit)).reduce((a, b) => a + b, 0)
return +seedingSize.toFixed(2)
}
async function main() {
rawComment = []
filterComArr = []
rawComment.push('楼层,uid,用户名,用户详情页,用户等级,做种大小GB,评论时间,额外信息,评论内容'.split(',').join(sep))
// 用户状态: 正常, 被ban, 警告
filterComArr.push('楼层,uid,用户名,用户等级,用户状态,做种大小GB,评论时间,上传量GB,下载量GB,分享率,预注册 ID,预注册邮箱,用户详情页'.split(',').join(filterSep))
href = window.location.href
curr_url = new URL(href)
title = jQuery(".embedded:contains('->')").text()
currPage = window.currentpage || +curr_url.searchParams.get('page') || 0
finalPage = jQuery("a[href*='?action=viewtopic']a[href*='page=']").toArray().map(_ => _.href)
finalPage = finalPage.filter(_ => !_.includes('&page=p'))
finalPage = [...new Set(finalPage)]
finalPage = finalPage.map(_ => +new URL(_).searchParams.get('page') || 0)
finalPage = finalPage.sort().pop() || 0
finalPage = isDebug ? Math.min(0, finalPage) : finalPage
let i = currPage
do {
console.log(`currPage: ${i} -> final: ${finalPage}`);
let html
if (i === currPage) {
html = document
} else {
curr_url.searchParams.set('page', i)
html = await jQuery.get(curr_url.toString())
}
await page(html)
await sleep(reqBreak)
i += 1
} while (!stopped && i <= finalPage)
if (isDebug) {
console.log(rawComment)
console.log(rawComment.join('\n'))
console.log(filterComArr.join('\n'))
} else {
createAndDownloadFile("raw.csv", rawComment.join('\n'))
createAndDownloadFile("filtered.csv", filterComArr.map(_ => _.split(sep).join(filterSep)).join('\n'))
}
}
const button = document.createElement("button");
button.textContent = "收集评论";
button.setAttribute("type", "button");
button.setAttribute("style", "position: fixed;display: block;top: 1px;width: 120px;height: 30px;");
document.body.appendChild(button);
button.addEventListener("click", async function () {
button.disabled = true
if (window.confirm('日志输出到 F12，请打开查看，执行完成后按钮可用, 刷新页面即可取消执行')) {
await main()
} else {
alert('用户主动取消')
}
button.disabled = false
});
})();
