Optimized website

Server-side rendering for syntax highlighting and DOM tree optimizations.
This commit is contained in:
Angelos Chalaris
2018-01-04 00:01:11 +02:00
parent d0dd4182b3
commit 13c27db7dc
9 changed files with 1316 additions and 1328 deletions

View File

@ -8,8 +8,21 @@ const fs = require('fs-extra'),
chalk = require('chalk'),
md = require('markdown-it')(),
minify = require('html-minifier').minify;
var Prism = require('prismjs');
// Load helper functions (these are from existing snippets in 30 seconds of code!)
const isTravisCI = () => 'TRAVIS' in process.env && 'CI' in process.env;
const unescapeHTML = str =>
str.replace(
/&|<|>|'|"/g,
tag =>
({
'&': '&',
'&lt;': '<',
'&gt;': '>',
'&#39;': "'",
'&quot;': '"'
}[tag] || tag)
);
if(isTravisCI() && /^Travis build: \d+/g.test(process.env['TRAVIS_COMMIT_MESSAGE'])) {
console.log(`${chalk.green('NOBUILD')} index build terminated, parent commit is a Travis build!`);
process.exit(0);
@ -149,8 +162,10 @@ try {
md
.render(`\n${snippets[taggedSnippet[0] + '.md']}`)
.replace(/<h3/g, `<h3 id="${taggedSnippet[0].toLowerCase()}" class="section double-padded"`)
.replace(/<\/h3>/g, '</h3><div class="section double-padded">') +
'</div></div><br/>';
.replace(/<pre><code class="language-js">([^\0]*?)<\/code><\/pre>/gm, (match, p1) => `<pre class="language-js">${Prism.highlight(unescapeHTML(p1), Prism.languages.javascript)}</pre>`)
.replace(/<\/pre>\s+<pre/g, '</pre><label class="collapse">Show examples</label><pre') +
'<button class="primary clipboard-copy"><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-clipboard"><path d="M16 4h2a2 2 0 0 1 2 2v14a2 2 0 0 1-2 2H6a2 2 0 0 1-2-2V6a2 2 0 0 1 2-2h2"></path><rect x="8" y="2" width="8" height="4" rx="1" ry="1"></rect></svg>&nbsp;Copy to clipboard</button>' +
'</div></div>';
} else {
output += md
.render(`## ${capitalize(tag, true)}\n`)
@ -162,14 +177,49 @@ try {
.render(`\n${snippets[taggedSnippet[0] + '.md']}`)
.replace(/<h3/g, `<h3 id="${taggedSnippet[0].toLowerCase()}" class="section double-padded"`)
.replace(/<\/h3>/g, '</h3><div class="section double-padded">')
.replace(/<\/pre>\s+<pre>/g, '</pre><label class="collapse">Show examples</label><pre>') +
.replace(/<pre><code class="language-js">([^\0]*?)<\/code><\/pre>/gm, (match, p1) => `<pre class="language-js">${Prism.highlight(unescapeHTML(p1), Prism.languages.javascript)}</pre>`)
.replace(/<\/pre>\s+<pre/g, '</pre><label class="collapse">Show examples</label><pre') +
'<button class="primary clipboard-copy"><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-clipboard"><path d="M16 4h2a2 2 0 0 1 2 2v14a2 2 0 0 1-2 2H6a2 2 0 0 1-2-2V6a2 2 0 0 1 2-2h2"></path><rect x="8" y="2" width="8" height="4" rx="1" ry="1"></rect></svg>&nbsp;Copy to clipboard</button>' +
'</div></div><br/>';
'</div></div>';
}
}
output += uncategorizedOutput;
// Add the ending static part
output += `\n${endPart + '\n'}`;
// Optimize punctuation nodes
let count = 0;
do {
const punctuationRegex = /<span class="token punctuation">([^\0<]*?)<\/span>([\n\r\s]*)<span class="token punctuation">([^\0]*?)<\/span>/gm;
output = output.replace(punctuationRegex,
(match, p1, p2, p3) => `<span class="token punctuation">${p1}${p2}${p3}</span>`
);
count = 0;
while (punctuationRegex.exec(output) !== null) {
++count;
}
} while (count > 0);
// Optimize operator nodes
do {
const operatorRegex = /<span class="token operator">([^\0<]*?)<\/span>([\n\r\s]*)<span class="token operator">([^\0]*?)<\/span>/gm;
output = output.replace(operatorRegex,
(match, p1, p2, p3) => `<span class="token operator">${p1}${p2}${p3}</span>`
);
count = 0;
while (operatorRegex.exec(output) !== null) {
++count;
}
} while (count > 0);
// Optimize keyword nodes
do {
const keyWordRegex = /<span class="token keyword">([^\0<]*?)<\/span>([\n\r\s]*)<span class="token keyword">([^\0]*?)<\/span>/gm;
output = output.replace(keyWordRegex,
(match, p1, p2, p3) => `<span class="token keyword">${p1}${p2}${p3}</span>`
);
count = 0;
while (keyWordRegex.exec(output) !== null) {
++count;
}
} while (count > 0);
// Minify output
output = minify(output, {
collapseBooleanAttributes: true,