注意：在您發布之後，您可能需要清除瀏覽器快取才能看到變更。
- Firefox或Safari：按住Shift時點擊重新整理，或按Ctrl-F5或Ctrl-R（Mac則為⌘-R）
- Google Chrome：按Ctrl-Shift-R（Mac則為⌘-Shift-R）
- Internet Explorer或Edge：按住Ctrl時點擊重新整理，或按Ctrl-F5
- Opera：按Ctrl-F5。
// origin es+ version of [[MediaWiki:PttHighlight.js]]
void function () {
const pttArticleId = {
decodeB64(b64) {
let bitString = ''
for (let i=0; i<b64.length; i++) {
const charCode = b64.codePointAt(i)
let bcode
if (charCode == 45) bcode = 62 // -
else if (charCode == 95) bcode = 63 // _
else if (charCode <= 57) bcode = charCode - 48 // 0-9
else if (charCode <= 90) bcode = charCode - 65 + 10 // A-Z
else if (charCode <= 122) bcode = charCode - 97 + 36 // a-z
else throw new Error('not ptt b64 string: ' + b64)
bitString += this.formatPadBit(bcode)
}
return bitString
},
formatPadBit(value) {
let string = value.toString(2)
while (string.length < 6) string = '0' + string
return string
},
toFileName(aid) {
const bitString = this.decodeB64(aid)
const time = parseInt(bitString.slice(0, -12), 2)
const random = parseInt(bitString.slice(-12), 2)
return `M.${time.toString(10)}.A.${random.toString(16)}`.toUpperCase()
}
}
function r(stringSegment) {
const regexpList = arguments
let resultString = stringSegment[0]
for (let i=1; i<stringSegment.length; i++) {
const regexp = regexpList[i]
if (typeof regexp == 'string') resultString += regexp
else resultString += regexp.source
resultString += stringSegment[i]
}
return new RegExp(resultString)
}
r.board = /\w[-_\w]*/
r.aid = /#([-_\w]{8})/
class Parser {
processRegexp(text, regexp, command) {
if (!regexp.global) {
regexp = new RegExp(regexp.source, regexp.flags + 'g')
}
const result = []
let index = 0
let scan
while (scan = regexp.exec(text)) {
const beforeMatch = text.slice(index, scan.index)
result.push(beforeMatch)
result.push(command(Array.from(scan)))
index = regexp.lastIndex
}
result.push(text.slice(index))
return result
}
token(text, type) {
const node = document.createElement('span')
node.textContent = text
node.classList.add(type)
return node
}
line(line, type) {
const lineNode = document.createElement('div')
lineNode.classList.add('line')
line.shift() // match()[0] is useless
for (const token of line) {
if (typeof token == 'string') {
const textNode = document.createTextNode(token)
lineNode.appendChild(textNode)
}
else if (token instanceof Node) {
lineNode.appendChild(token)
}
else {
console.error('unknown token: ', token)
}
}
if (type) lineNode.classList.add(type)
return lineNode
}
static rule(regexp, command) {
const copy = new this()
copy.regexp = regexp
copy.command = command
return copy
}
}
Parser.aidParser = {
__proto__: Parser.prototype,
processAid(text) {
return this.processRegexp(text, this.aidRegexp, this.aidToNode)
},
aidRegexp: r`${r.aid}(?:\\s*\\((${r.board})\\))?`,
aidToNode(scan) {
const aid = scan[1]
const board = scan[2] || 'not-found'
const anchor = document.createElement('a')
const fileName = pttArticleId.toFileName(aid)
anchor.href = `https://www.ptt.cc/bbs/${board}/${fileName}.html`
anchor.textContent = scan[0]
anchor.classList.add('ptt-aid')
return anchor
}
}
Parser.statusParser = {
__proto__: Parser.prototype,
keyRegexp: /\([^()]+\)/,
keyCommand(scan) {
return this.token(scan[0], 'key')
},
processKey(text) {
return this.processRegexp(
text, this.keyRegexp, this.keyCommand.bind(this)
)
}
}
const pttHighlight = {
r, Parser, pttArticleId,
matchCommand: [
Parser.rule(
// /^(\s*作者 )( *)(\w[\w\d]*)( *\()(.*)(\) *)( 看板 ?)( ?)([-\d\w_]+)( *)$/,
/^(\s*作者 )(.*)( 看板 )(.*)$/,
function command(match) {
match[1] = this.token(match[1], 'head')
match[3] = this.token(match[3], 'head')
return this.line(match, 'head')
}
),
Parser.rule(
/^(\s*(?:標題|時間|轉信|作者|看板) )(.*)$/,
function command(match) {
match[1] = this.token(match[1], 'head')
return this.line(match, 'head')
}
),
{
__proto__: Parser.aidParser,
// → illumi:要去 op 板啊^^ 推 140.123.225.127 12/19
// → TomRiddle: 五樓厲害了 61.216.122.103 06/21 18:18
// 放棄詭異的格式 /^\s*([推噓→])( +)(\w[\w\d]*)(:.*?)((?:\s[推噓→]\s{0,8})?(?:\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\ )?\d{2}\/\d{2}(?: \d{2}:\d{2})?)?( *)$/,
//regexp: /^\s*([推噓→])( +)(\w[\w\d]*)(:.*?)((?:\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\ )?\d{2}\/\d{2}(?: \d{2}:\d{2})?)?( *)$/,
regexp: /^\s*([推噓→])( +)([^:\n]+)(:.*?)((?:\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\ )?\d{2}\/\d{2}(?: \d{2}:\d{2})?)?( *)$/,
command(match) {
switch (match[1]) {
case '推':
match[1] = this.token(match[1], 'push')
break
case '噓':
match[1] = this.token(match[1], 'boo')
break
case '→':
match[1] = this.token(match[1], 'arrow')
break
}
match[3] = this.token(match[3], 'username')
if (match[5]) match[5] = this.token(match[5], 'date')
match.splice(4, 1, ...this.processAid(match[4]))
return this.line(match, 'comment')
}
},
{
__proto__: Parser.statusParser,
// 瀏覽 第 1/7 頁 ( 11%) 目前顯示: 第 01~22 行 (y)回應(X%)推文(h)說明(← )離開
regexp: /^(\s*瀏覽 第 \d+(?:\/\d+)? 頁 \(\s*\d+%\)\s)(.*?)(\(.*)$/,
markPageRange(match) {
let range = 'page-middle'
if (/第\s*01~/.test(match[2].textContent)) range = 'page-first'
else if (/100%/.test(match[1].textContent)) range = 'page-last'
match[1].classList.add(range)
},
command(match) {
match[1] = this.token(match[1], 'percent')
match[2] = this.token(match[2], 'line-number')
this.markPageRange(match)
match.splice(3,1, ...this.processKey(match[3]))
return this.line(match, 'status')
}
},
Parser.rule(
/^─+$/,
function (match) {
match.push(match[0])
return this.line(match, 'hr')
}
),
{
__proto__: Parser.aidParser,
regexp: /^※.*$/,
aidRegexp: r`(?:(${r.board})\\s*看板\\s*)?${r.aid}`,
aidToNode(scan) {
let text, aid, board
[text, board, aid] = scan
return super.aidToNode([text, aid, board])
},
command(match) {
const parseAid = this.processAid(match[0])
match.push(...parseAid)
return this.line(match, 'system')
}
},
Parser.rule(
/^(\s*▄+\s*)(請按任意鍵繼續)(\s*▄+\s*)$/,
function command(match) {
match[1] = this.token(match[1], 'back-wall')
match[3] = this.token(match[3], 'back-wall')
return this.line(match, 'anykey')
}
),
Parser.rule(
/^(\s*◆.*)(\s\[按任意鍵繼續\]\s*)$/,
function match(match) {
match[1] = this.token(match[1], 'action')
match[2] = this.token(match[2], 'prompt')
return this.line(match, 'anykey')
}
),
Parser.rule(
/^(【.*?】)(\s*.*\s*)((?:看板|系列|文摘)《.*》\s*)$/,
function command(match) {
match[2] = this.token(match[2], 'center')
return this.line(match, 'info-board')
}
),
Parser.rule(
/^\s+編號\s+.\s*.\s*.*\s+人氣.*$/,
function command(match) {
match.push(match[0])
return this.line(match, 'field-head')
}
),
{
__proto__: Parser.statusParser,
regexp: /^(\s*(?:文章選讀|選擇看板)\s?)(.*)$/,
command(match) {
match[1] = this.token(match[1], 'status-name')
match.splice(2,1, ...this.processKey(match[2]))
return this.line(match, 'status')
}
},
{
__proto__: Parser.prototype,
regexp: /^([>●]?)(\s*)(\d+\s|★\s*)(.)( \d|\d\d|爆|X.| )(..\/.*$)$/,
command(match) {
if (match[3].charAt(0) == '★') {
match[3] = this.token(match[3], 'announcement')
}
match[5] = this.markPushCount(match[5])
return this.line(match, 'list-entry')
},
markPushCount(symbol) {
const count = Number(symbol)
let type = 'low'
if (Number.isNaN(count)) {
if (symbol == '爆') type = 'high'
else if (symbol.charAt(0) == 'X') type = 'bad'
}
else {
if (count < 10) type = 'ones'
else if (count < 100) type = 'tens'
}
return this.token(symbol, type)
}
},
{
__proto__: Parser.aidParser,
regexp: /^[:>](?:|\s.*)$/,
command(match) {
const parseAid = this.processAid(match[0])
match.push(...parseAid)
return this.line(match, 'quote')
}
},
{
__proto__: Parser.aidParser,
regexp: /^.*$/,
command(match) {
const parseAid = this.processAid(match[0])
match.push(...parseAid)
return this.line(match)
}
}
],
textToNode(text) {
const node = document.createElement('pre')
for (let line of text.split(/\n/g)) {
if (line.slice(-1) == '\r') line = line.slice(0,-1)
for (const match of this.matchCommand) {
const scan = line.match(match.regexp)
if (scan) {
const lineNode = match.command(Array.from(scan))
node.appendChild(lineNode)
break
}
}
}
return node
},
processTemplate(root = document) {
return this.processRootSelector(root, '.ptt-format-text.raw')
},
processRootSelector(root = document, selector = '.ptt-format-text.raw') {
const rawList = root.querySelectorAll(selector)
for (const raw of rawList) {
const node = this.textToNode(raw.textContent)
node.className = raw.className
node.classList.remove('raw')
raw.parentNode.classList.remove('raw')
if (this.debug) raw.after(node)
else raw.replaceWith(node)
}
return rawList
},
processMarkdownLang(root = document) {
const markdownNodeList = root.querySelectorAll('.lang-ptt')
markdownNodeList.forEach(node => {
node.classList.add('ptt-format-text', 'raw')
})
return this.processRootSelector(root)
}
}
if (typeof exports == 'object') {
exports.pttHighlight = pttHighlight
}
else if (typeof window == 'object') {
window.pttHighlight = pttHighlight
}
else if (typeof this == 'object') {
this.pttHighlight = pttHighlight
}
pttHighlight.processTemplate()
}()