Perf: read file woker (#1337)

* perf: read file worker

* fix: Http node url input

* fix: htm2md

* fix: html2md

* fix: ts

* perf: Problem classification increases the matching order

* feat: tool response answer
This commit is contained in:
Archer
2024-04-30 18:12:20 +08:00
committed by GitHub
parent 1529c1e991
commit b5f0ac3e1d
35 changed files with 413 additions and 398 deletions

View File

@@ -0,0 +1,40 @@
import TurndownService from 'turndown';
const domino = require('domino-ext');
const turndownPluginGfm = require('joplin-turndown-plugin-gfm');
export const html2md = (html: string): string => {
const turndownService = new TurndownService({
headingStyle: 'atx',
bulletListMarker: '-',
codeBlockStyle: 'fenced',
fence: '```',
emDelimiter: '_',
strongDelimiter: '**',
linkStyle: 'inlined',
linkReferenceStyle: 'full'
});
try {
const window = domino.createWindow(html);
const document = window.document;
turndownService.remove(['i', 'script', 'iframe']);
turndownService.addRule('codeBlock', {
filter: 'pre',
replacement(_, node) {
const content = node.textContent?.trim() || '';
// @ts-ignore
const codeName = node?._attrsByQName?.class?.data?.trim() || '';
return `\n\`\`\`${codeName}\n${content}\n\`\`\`\n`;
}
});
turndownService.use(turndownPluginGfm.gfm);
return turndownService.turndown(document);
} catch (error) {
console.log('html 2 markdown error', error);
return '';
}
};