Rephrase wordpressdotcom importer alternative caveat.

The wordpressdotcom importer does import pages, posts, images, etc so previous statement was untrue.

Author: parkr

docs/_importers/wordpressdotcom.md

@@ -17,10 +17,9 @@ Their default values are what you see above.
 
 ### Further WordPress migration alternatives
 
-While the above method works, it does not import much of the metadata that is
-usually stored in WordPress posts and pages. If you need to export things like
-pages, tags, custom fields, image attachments and so on, the following resources
-might be useful to you:
+While the above method works, it doesn't import absolutely every piece of
+metadata. If you need to import custom fields from your pages and posts,
+the following resources might be useful to you:
 
 - [Exitwp](https://github.com/thomasf/exitwp) is a configurable tool written in
   Python for migrating one or more WordPress blogs into Jekyll (Markdown) format

lib/jekyll-import/importers/wordpressdotcom.rb

@@ -26,7 +26,7 @@ def self.download_images(title, post_doc, assets_folder)
         images = post_doc.css("img")
         return if images.empty?
 
-        Jekyll.logger.info "Downloading images for ", title
+        Jekyll.logger.info "Downloading:", "images for #{title}"
         images.each do |i|
           uri = URI::DEFAULT_PARSER.escape(i["src"])
 
@@ -54,8 +54,8 @@ def self.download_images(title, post_doc, assets_folder)
 
       class Item
         def initialize(node)
-          @node = node
           raise "Node is nil" if node.nil?
+          @node = node
         end
 
         def text_for(path)
@@ -220,7 +220,7 @@ def self.process(options)
         end
 
         import_count.each do |key, value|
-          Jekyll.logger.info "Imported #{value} #{key}s"
+          Jekyll.logger.info "Imported", "#{value} #{key}s"
         end
       end
 

test/test_wordpressdotcom_importer.rb

@@ -1,4 +1,5 @@
 require "helper"
+require "tmpdir"
 
 Importers::WordpressDotCom.require_deps
 
@@ -13,9 +14,11 @@ class TestWordpressDotComMigrator < Test::Unit::TestCase
     assert File.exist?(xml_path), "Expect xml file to exist"
     tmpdir = Dir.mktmpdir
     Dir.chdir(tmpdir) do
+      Jekyll.logger = ::Logger.new(File.open("output.log", "w"))
       JekyllImport::Importers::WordpressDotCom.process({"source" => xml_path})
+      Jekyll.logger = Jekyll::Stevenson.new
     end
-    
+
     # The old export produced the following files:
     # .
     # ├── _attachments
@@ -45,16 +48,21 @@ class TestWordpressDotComMigrator < Test::Unit::TestCase
     #             └── a-bright-night-sky-full-of-stars-and-the-milky.png
     assert_path_exist File.expand_path("assets/2025/01/a-bright-night-sky-full-of-stars-and-the-milky.png", tmpdir)
 
+    jekyllbot_author_data = {
+      "login"        => "jekyllbot",
+      "email"        => "[email protected]",
+      "display_name" => "jekyllbot",
+      "first_name"   => "Jekyllbot",
+      "last_name"    => "Hyde",
+    }
+
     assert_path_exist File.expand_path("_pages/2025-01-19-about.html", tmpdir)
     page_content = File.read(File.expand_path("_pages/2025-01-19-about.html", tmpdir))
-    assert_match %r{published: true}, page_content
-    assert_match %r{permalink: "\/about\/"}, page_content
-    assert_match %r{^author:\s*\n}, page_content
-    assert_match %r{\s\slogin: jekyllbot}, page_content
-    assert_match %r{\s\semail: [email protected]}, page_content
-    assert_match %r{\s\sdisplay_name: jekyllbot}, page_content
-    assert_match %r{\s\sfirst_name: Jekyllbot}, page_content
-    assert_match %r{\s\slast_name: Hyde}, page_content
+    page_front_matter = page_content.match(/^(---\n.*?---\n)/m)[0]
+    page_data = YAML.safe_load(page_front_matter)
+    assert_equal true, page_data["published"]
+    assert_equal "/about/", page_data["permalink"]
+    assert_equal jekyllbot_author_data, page_data["author"]
 
     assert_path_exist File.expand_path("_posts/2025-01-18-blog-post-1.html", tmpdir)
     post_content = File.read(File.expand_path("_posts/2025-01-18-blog-post-1.html", tmpdir))
@@ -64,7 +72,7 @@ class TestWordpressDotComMigrator < Test::Unit::TestCase
     assert_equal "Blog Post 1", post_data["title"]
     assert_equal ["Foo"], post_data["categories"]
     assert_equal ["code", "ruby", "ship", "stars"], post_data["tags"]
-    assert_equal "jekyllbot", post_data["author"]["login"]
+    assert_equal jekyllbot_author_data, post_data["author"]
     assert_equal "/2025/01/18/blog-post-1/", post_data["permalink"]
     assert_equal "101042542514", post_data["meta"]["_publicize_job_id"]
     assert_includes post_content, "<p>This is a blog post.</p>"