从 WordPress 到 Astro迁移实战全记录

1
import fs from "node:fs";
2
import path from "node:path";
3
import { parseWxr } from "./lib/wxr-parse.mjs";
4
import { htmlToMarkdown, rewriteImageUrls } from "./lib/html-to-md.mjs";
5

6
const ROOT = process.cwd();
7
const XML = path.join(ROOT, "muxui.WordPress.2026-07-23.xml");
8
const POSTS_DIR = path.join(ROOT, "src/content/posts");
9
const SITE_CONFIG = path.join(ROOT, "src/config/siteConfig.ts");
10

11
function decodeSlug(raw) {
12
  if (!raw) return "";
13
  try {
14
    return decodeURIComponent(raw);
15
  } catch {
16
    return raw;
17
  }
18
}
19

20
function safeSlug(title, id) {
21
  const base = title
22
    .toLowerCase()
23
    .replace(/[^\p{L}\p{N}]+/gu, "-")
24
    .replace(/^-+|-+$/g, "")
25
    .slice(0, 80);
26
  return base || `post-${id}`;
27
}
28

29
function pickCategory(cats) {
30
  const specific = cats.filter((c) => c !== "文章");
31
  return specific[0] || cats[0] || "";
32
}
33

34
function yamlEscape(s) {
35
  return String(s).replace(/\\/g, "\\\\").replace(/"/g, '\\"');
36
}
37

38
function toIsoDate(wpDate) {
39
  return wpDate.slice(0, 10);
40
}
41

42
function excerptFrom(md, excerpt) {
43
  if (excerpt?.trim()) return excerpt.trim().slice(0, 160);
44
  const plain = md
45
    .replace(/!\[[^\]]*\]\([^)]*\)/g, "")
46
    .replace(/\[[^\]]*\]\([^)]*\)/g, "$1")
47
    .replace(/[#>*`\[\]()\-_!]/g, " ")
48
    .replace(/https?:\/\/\S+/g, "")
49
    .replace(/\s+/g, " ")
50
    .trim();
51
  return plain.slice(0, 120);
52
}
53

54
function clearPostsDir(dir) {
55
  if (!fs.existsSync(dir)) {
56
    fs.mkdirSync(dir, { recursive: true });
57
    return;
58
  }
59
  for (const name of fs.readdirSync(dir)) {
60
    fs.rmSync(path.join(dir, name), { recursive: true, force: true });
61
  }
62
}
63

64
function buildFrontmatter(post, image, body, slug) {
65
  const tags = post.tags.map((t) => `"${yamlEscape(t)}"`).join(", ");
66
  const desc = excerptFrom(body, post.excerpt);
67
  return `---
68
title: "${yamlEscape(post.title)}"
69
published: ${toIsoDate(post.date)}
70
updated: ${toIsoDate(post.modified)}
71
description: "${yamlEscape(desc)}"
72
image: "${yamlEscape(image)}"
73
tags: [${tags}]
74
category: "${yamlEscape(pickCategory(post.cats))}"
75
draft: false
76
comment: true
77
slug: "${yamlEscape(slug)}"
78
---
79

80
`;
81
}
82

83
function patchSiteConfig(filePath) {
84
  let src = fs.readFileSync(filePath, "utf8");
85
  src = src.replace(/title: "Firefly"/, 'title: "Muxui"');
86
  src = src.replace(/subtitle: "Demo site"/, 'subtitle: "记录美好生活！"');
87
  src = src.replace(
88
    /site_url: "https:\/\/firefly\.cuteleaf\.cn"/,
89
    'site_url: "https://muxui.com"',
90
  );
91
  src = src.replace(
92
    /description:\s*\n\s*"[\s\S]*?",/,
93
    'description:\n\t\t"记录美好生活！",',
94
  );
95
  src = src.replace(
96
    /keywords: \[[\s\S]*?\],/,
97
    `keywords: [
98
\t\t"Muxui",
99
\t\t"博客",
100
\t\t"技术博客",
101
\t\t"muxui.com",
102
\t],`,
103
  );
104
  src = src.replace(/(navbar: \{[\s\S]*?title: )"Firefly"/, '$1"Muxui"');
105
  fs.writeFileSync(filePath, src);
106
}
107

108
function main() {
109
  if (!fs.existsSync(XML)) {
110
    console.error("Missing WXR:", XML);
111
    process.exit(1);
112
  }
113

114
  const { posts, attachmentsById } = parseWxr(XML);
115
  console.log(`Parsed ${posts.length} published posts`);
116

117
  clearPostsDir(POSTS_DIR);
118

119
  const used = new Set();
120
  for (const post of posts) {
121
    const body = htmlToMarkdown(post.content);
122
    let image = "";
123
    if (post.thumbnailId && attachmentsById.has(post.thumbnailId)) {
124
      image = rewriteImageUrls(attachmentsById.get(post.thumbnailId));
125
    }
126
    let slug = decodeSlug(post.slug) || safeSlug(post.title, post.id);
127
    // Windows-safe filenames: avoid reserved chars
128
    slug = slug.replace(/[<>:"/\\|?*]/g, "-").replace(/\.+$/g, "");
129
    if (used.has(slug)) slug = `${slug}-${post.id}`;
130
    used.add(slug);
131
    const fm = buildFrontmatter(post, image, body, slug);
132
    const out = path.join(POSTS_DIR, `${slug}.md`);
133
    fs.writeFileSync(out, fm + body + "\n", "utf8");
134
    console.log("Wrote", path.relative(ROOT, out));
135
  }
136

137
  patchSiteConfig(SITE_CONFIG);
138
  console.log("Patched", path.relative(ROOT, SITE_CONFIG));
139
  console.log("Done.");
140
}
141

142
main();

1
import fs from "node:fs";
2
import { XMLParser } from "fast-xml-parser";
3

4
const parser = new XMLParser({
5
  ignoreAttributes: false,
6
  attributeNamePrefix: "@_",
7
  cdataPropName: "__cdata",
8
  isArray: (name) =>
9
    ["item", "category", "wp:postmeta", "wp:comment"].includes(name),
10
});
11

12
function textOf(node) {
13
  if (node == null) return "";
14
  if (typeof node === "string" || typeof node === "number") return String(node);
15
  if (typeof node === "object" && "__cdata" in node)
16
    return String(node.__cdata ?? "");
17
  return String(node);
18
}
19

20
function metaMap(postmeta) {
21
  const map = {};
22
  for (const m of postmeta || []) {
23
    map[textOf(m["wp:meta_key"])] = textOf(m["wp:meta_value"]);
24
  }
25
  return map;
26
}
27

28
function categoriesOf(item) {
29
  const cats = [];
30
  const tags = [];
31
  for (const c of item.category || []) {
32
    const name = textOf(c);
33
    const domain = c["@_domain"];
34
    if (domain === "post_tag") tags.push(name);
35
    else if (domain === "category") cats.push(name);
36
  }
37
  return { cats, tags };
38
}
39

40
export function parseWxr(xmlPath) {
41
  const xml = fs.readFileSync(xmlPath, "utf8");
42
  const doc = parser.parse(xml);
43
  const channel = doc.rss.channel;
44
  const attachmentsById = new Map();
45
  const posts = [];
46

47
  for (const item of channel.item || []) {
48
    const type = textOf(item["wp:post_type"]);
49
    const id = textOf(item["wp:post_id"]);
50
    if (type === "attachment") {
51
      const url = textOf(item["wp:attachment_url"]) || textOf(item.guid) || "";
52
      attachmentsById.set(id, url);
53
      continue;
54
    }
55
    if (type !== "post") continue;
56
    if (textOf(item["wp:status"]) !== "publish") continue;
57

58
    const { cats, tags } = categoriesOf(item);
59
    const meta = metaMap(item["wp:postmeta"]);
60
    posts.push({
61
      id,
62
      title: textOf(item.title),
63
      slug: textOf(item["wp:post_name"]),
64
      date: textOf(item["wp:post_date"]),
65
      modified: textOf(item["wp:post_modified"]),
66
      excerpt: textOf(item["excerpt:encoded"]),
67
      content: textOf(item["content:encoded"]),
68
      cats,
69
      tags,
70
      thumbnailId: meta._thumbnail_id || "",
71
    });
72
  }
73

74
  return {
75
    site: {
76
      title: textOf(channel.title),
77
      link: textOf(channel.link),
78
      description: textOf(channel.description),
79
    },
80
    attachmentsById,
81
    posts,
82
  };
83
}

1
import TurndownService from "turndown";
2

3
const TARGET_PREFIX = "https://muxui.com/uploads/";
4

5
export function rewriteImageUrls(text) {
6
  return text
7
    .replace(/https?:\/\/img\.roven\.cc\/images\//g, TARGET_PREFIX)
8
    .replace(/https?:\/\/muxui\.com\/wp-content\/uploads\//g, TARGET_PREFIX)
9
    .replace(/(["'(=\s]|^)\/wp-content\/uploads\//g, `$1${TARGET_PREFIX}`);
10
}
11

12
function stripShortcodes(html) {
13
  return html
14
    .replace(/\[\/?b2[^\]]*\]/gi, "")
15
    .replace(/\[hide[^\]]*\]([\s\S]*?)\[\/hide\]/gi, "$1")
16
    .replace(/\[.*?\]/g, (m) => {
17
      const href = m.match(/https?:\/\/[^\s\]]+/);
18
      return href ? `<p><a href="${href[0]}">${href[0]}</a></p>` : "";
19
    });
20
}
21

22
export function htmlToMarkdown(html) {
23
  const cleaned = stripShortcodes(rewriteImageUrls(html));
24
  const td = new TurndownService({
25
    headingStyle: "atx",
26
    codeBlockStyle: "fenced",
27
    bulletListMarker: "-",
28
  });
29
  td.keep(["iframe", "video", "table"]);
30
  return td.turndown(cleaned).trim();
31
}