NER nexpr

これは？

ライブドアブログのダンプデータからワードを抽出するためのページ

JavaScript とか Node.js とか Web Components とか C# とか .NET とか...
主にプログラミング言語やライブラリやツールや機能などの名前を取り出すのが目的

HTML データから不要なタグを除外した上で改行を維持して文字列として取得したいのでブラウザを使う

半角スペースや記号を名前に含むものがあるのでほとんどの記号を含む半角文字列を正規表現で取り出して除外パターンと一致しないのを残す

	[root@af929bf74005 /]# time dnf install git -y
	Fedora 38 - x86_64 9.5 MB/s \| 83 MB 00:08
	Fedora 38 openh264 (From Cisco) - x86_64 1.1 kB/s \| 2.5 kB 00:02
	Fedora Modular 38 - x86_64 3.2 MB/s \| 2.8 MB 00:00
	Fedora 38 - x86_64 - Updates 3.8 MB/s \| 36 MB 00:09
	Fedora Modular 38 - x86_64 - Updates 1.5 MB/s \| 2.1 MB 00:01
	Dependencies resolved.
	========================================================================================================================
	Package Architecture Version Repository Size
	========================================================================================================================

	export const sortDomain = (domains) => {
	return domains.map(x => x.split(".")).sort((a, b) => {
	const max = Math.max(a.length, b.length)
	const recur = (i = 0) => {
	if (i >= max) return 0
	return (a.at(-i - 1) ?? "").localeCompare(b.at(-i - 1) ?? "") \|\| recur(i + 1)
	}
	return recur()
	}).map(x => x.join("."))
	}

	<!DOCTYPE html>

	<meta charset="utf-8" />

	<script type="module">
	import { html, render } from "https://cdn.jsdelivr.net/npm/[email protected]/node.js"

	const page = async () => {
	const valueFormat = (value) => {
	try {

	import { LitElement, css, html } from "lit"
	import { live } from "lit/directives/live.js"

	export class TimeInput extends LitElement {
	static styles = css`
	.container {
	display: flex;
	gap: 10px;
	}
	input {

	<!doctype html>
	<meta charset="utf-8" />

	<script type="module">
	import { createApp, ref, h } from "https://unpkg.com/[email protected]/dist/vue.esm-browser.js"
	import htm from "https://cdn.jsdelivr.net/npm/[email protected]/mini/index.module.js"
	const html = htm.bind(
	(type, props, ...children) =>
	h(type, props, typeof type === "string" ? children : () => children)
	)

	import { html } from "lit"
	import { StylitElement } from "./StylitElement.js"

	class ExampleElement extends StylitElement {
	static properties = {
	n: {}
	}
	constructor() {
	super()
	this.n = 0

	import {html, LitElement} from "lit"
	import { directive, Directive } from "lit/directive.js"

	const customElementDirective = (element, Directive) => {
	return directive(
	class extends Directive {
	host = element
	}
	)
	}

	const add = (a, b) => a + b
	export default add