Created
March 26, 2024 10:20
-
-
Save shepherdwind/f23d46046e18c16192d16a6769113d5c to your computer and use it in GitHub Desktop.
download image from yuque exported markdown file
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import { remark } from 'remark'; | |
| import path from 'path'; | |
| import fs from 'fs/promises'; | |
| import { selectAll } from 'unist-util-select'; | |
| import { glob } from 'glob'; | |
| import { createWriteStream } from 'fs'; | |
| import { pipeline } from 'stream/promises'; | |
| import { request } from 'undici'; | |
| import { pinyin } from 'pinyin-pro'; | |
| export async function run() { | |
| const files = await glob('**/*.md', { | |
| ignore: 'node_modules/**' | |
| }); | |
| for (let file of files) { | |
| const doc = await fs.readFile(file); | |
| console.log(`start handle ${file}`); | |
| const content = await remark() | |
| .data('settings', { bullet: '-', listItemIndent: 'one' }) | |
| .use([ | |
| [ replaceHTML ], | |
| // [ relativeLink, { doc, mapping }], | |
| [ downloadAsset, { filePath: file } ], | |
| ]) | |
| .process(doc.toString()); | |
| await writeFile(file, content.toString()); | |
| console.log(`finish write file ${file}`); | |
| } | |
| } | |
| function downloadAsset(opts) { | |
| return async tree => { | |
| const docFilePath = opts.filePath; | |
| const base = pinyin(path.basename(docFilePath, '.md'), { | |
| toneType: 'none', | |
| type: 'array', | |
| nonZh: 'consecutive', | |
| }).map(o => o.trim()).join('_'); | |
| const assetsDir = path.join(docFilePath.split('/')[0], 'assets', base); | |
| const assetNodes = selectAll('image[url^=http]', tree); | |
| for (const node of assetNodes) { | |
| const assetName = new URL(node.url).pathname.split('/').pop(); | |
| const filePath = path.join(assetsDir, assetName); | |
| await download(node.url, path.join('./', filePath), { headers: { 'User-Agent': 'yuque-exporter' } }); | |
| node.url = path.relative(path.dirname(docFilePath), filePath); | |
| } | |
| }; | |
| } | |
| function replaceHTML() { | |
| return tree => { | |
| const htmlNodes = selectAll('html', tree); | |
| for (const node of htmlNodes) { | |
| if (node.value === '<br />' || node.value === '<br/>') { | |
| node.type = 'text'; | |
| node.value = '\n'; | |
| } | |
| } | |
| }; | |
| } | |
| async function download(url, filePath, opts = {}) { | |
| const { headers, ...otherOpts } = opts; | |
| const { body, statusCode } = await request(url, { | |
| headers: { | |
| 'User-Agent': 'yuque-exporter', | |
| ...headers, | |
| }, | |
| maxRedirections: 10, | |
| ...otherOpts, | |
| }); | |
| if (statusCode !== 200) { | |
| return url; | |
| } | |
| await mkdir(path.dirname(filePath)); | |
| await pipeline(body, createWriteStream(filePath)); | |
| return filePath; | |
| } | |
| async function mkdir(p) { | |
| await fs.mkdir(p, { recursive: true }); | |
| } | |
| async function writeFile(filePath, content) { | |
| await fs.mkdir(path.dirname(filePath), { recursive: true }); | |
| content = typeof content === 'string' || Buffer.isBuffer(content) ? content : JSON.stringify(content, null, 2); | |
| await fs.writeFile(filePath, content, 'utf-8'); | |
| } | |
| run(); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
use
The minimum required version of node is 18.
The majority of the code was copied from https://github.com/atian25/yuque-exporter/blob/master/src/lib/doc.ts.