fix: extract excerpt before markdown conversion (#64)

hexojs · Jul 4, 2020 · 68ea753 · 68ea753
1 parent 4df53be
commit 68ea753
Show file tree

Hide file tree

Showing 2 changed files with 52 additions and 4 deletions.
diff --git a/lib/migrator.js b/lib/migrator.js
@@ -21,11 +21,15 @@ module.exports = async function(args) {
   let errNum = 0;
   let skipNum = 0;
   let input, feed;
-  const rExcerpt = /<a id="more"><\/a>/i;
+  const rExcerpt = /<!--+\s*more\s*--+>/i;
   const postExcerpt = '\n<!-- more -->\n';
   const posts = [];
   let currentPosts = [];
 
+  const md = str => {
+    return tomd.turndown(str);
+  };
+
   try {
     if (!source) {
       const help = [
@@ -59,19 +63,27 @@ module.exports = async function(args) {
       let { title, content, description } = item;
 
       const layout = status === 'draft' ? 'draft' : 'post';
-      content = tomd.turndown(content).replace(/\r\n/g, '\n');
 
       if (type !== 'page') {
         // Apply 'limit' option to post only
         if (postLimit >= limit) continue;
         postLimit++;
 
         if (rExcerpt.test(content)) {
-          content.replace(rExcerpt, postExcerpt);
+          content.replace(rExcerpt, (match, index) => {
+            const excerpt = md(content.substring(0, index).trim());
+            const more = md(content.substring(index + match.length).trim());
+
+            content = excerpt + postExcerpt + more;
+          });
         } else if (description) {
-          description = tomd.turndown(description).replace(/\r\n/g, '\n');
+          description = md(description);
           content = description + postExcerpt + content;
+        } else {
+          content = md(content);
         }
+      } else {
+        content = md(content);
       }
 
       if (!title) {

diff --git a/test/index.js b/test/index.js
@@ -101,6 +101,42 @@ describe('migrator', function() {
     posts.includes(unescape(percentEncoded.slug) + '.md').should.eql(true);
   });
 
+  it('excerpt', async () => {
+    const content = 'foo<!-- more -->bar';
+    const xml = `<rss><channel><title>test</title>
+    <item><title>baz</title><content:encoded><![CDATA[${content}]]></content:encoded></item>
+    </channel></rss>`;
+    const path = join(__dirname, 'excerpt.xml');
+    await writeFile(path, xml);
+    await m({ _: [path] });
+
+    const rendered = await readFile(join(hexo.source_dir, '_posts', 'baz.md'));
+    const rFrontMatter = /^([\s\S]+?)\n(-{3,}|;{3,})(?:$|\n([\s\S]*)$)/;
+    const output = rendered.match(rFrontMatter)[3].replace(/\r?\n|\r/g, '');
+
+    output.should.eql(content);
+
+    await unlink(path);
+  });
+
+  it('excerpt - wp:more', async () => {
+    const content = 'foo<!-- wp:more --><!-- more --><!-- wp:more -->bar';
+    const xml = `<rss><channel><title>test</title>
+    <item><title>baz</title><content:encoded><![CDATA[${content}]]></content:encoded></item>
+    </channel></rss>`;
+    const path = join(__dirname, 'excerpt.xml');
+    await writeFile(path, xml);
+    await m({ _: [path] });
+
+    const rendered = await readFile(join(hexo.source_dir, '_posts', 'baz.md'));
+    const rFrontMatter = /^([\s\S]+?)\n(-{3,}|;{3,})(?:$|\n([\s\S]*)$)/;
+    const output = rendered.match(rFrontMatter)[3].replace(/\r?\n|\r/g, '');
+
+    output.should.eql(content.replace(/<!-- wp:more -->/g, ''));
+
+    await unlink(path);
+  });
+
   it('no argument', async () => {
     try {
       await m({ _: [''] });