add an atom feed

build.py

@@ -10,7 +10,10 @@
 
 import argparse
 import glob
+import html
+from io import StringIO
 import markdown
+import operator
 import os
 import re
 import shutil
@@ -24,6 +27,8 @@ deploy_directory = '~/src/www-home'
 md_extensions = [
     'fenced_code', 'codehilite', 'nl2br', 'toc', 'smarty', 'tables', 'linkify']
 
+blog_entries = []
+
 
 def print_file(in_file, out_file):
   print('%-62s -> %s' % (in_file, out_file))
@@ -42,6 +47,13 @@ def copy_static_files():
       shutil.copy2(source, dest)
 
 
+def find_update_date(text):
+  match = re.search(r'^Published:? (\d{4}-\d{2}-\d{2})', text, re.MULTILINE)
+  if not match:
+    return None
+  return match.group(1)
+
+
 def process_markdown_files():
   template = open('template.html').read()
   for (dirpath, _, filenames) in os.walk(input_directory):
@@ -50,6 +62,8 @@ def process_markdown_files():
       if not markdown_filename.endswith('.md'):
         continue
 
+      blog_entry = {}
+
       markdown_file = open(markdown_filename)
       text = markdown_file.read()
       markdown_file.close()
@@ -62,6 +76,9 @@ def process_markdown_files():
         title = match.group(1).lstrip('# ')
       else:
         title = text
+
+      blog_entry['title'] = html.escape(title)
+
       title += ' | Colin McMillen'
       if markdown_filename == os.path.join(input_directory, 'index.md'):
         title = 'Colin McMillen'
@@ -75,9 +92,16 @@ def process_markdown_files():
       if page_url.endswith('index.html'):  # strip off index.html
         page_url = page_url[:-len('index.html')]
 
-      html = markdown.markdown(
+      update_date = find_update_date(text)
+      if update_date:
+        blog_entry['url'] = 'https://www.mcmillen.dev/' + page_url
+        blog_entry['date'] = update_date
+        blog_entries.append(blog_entry)
+
+      html_content = markdown.markdown(
           text, extensions=md_extensions, output_format='html5')
-      output = template.format(title=title, content=html, page_url=page_url)
+      output = template.format(
+          title=title, content=html_content, page_url=page_url)
 
       os.makedirs(out_dirpath, exist_ok=True)
       print_file(markdown_filename, out_fullpath)
@@ -93,15 +117,71 @@ grep -v ^output/google |
 grep -v ^output/drafts |
 perl -pe 's|output|https://www.mcmillen.dev|'
 > output/sitemap.txt""".split('\n'))
+  print_file('', 'output/sitemap.txt')
   os.system(sitemap_command)
 
 
-def make_rss():  # TODO: implement.
+def make_atom_feed():
-  pass
+  atom_template = '''<?xml version="1.0" encoding="utf-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom">
+
+  <title>Colin McMillen's Blog</title>
+  <link href="https://www.mcmillen.dev"/>
+  <link rel="self" href="https://www.mcmillen.dev/atom.xml"/>
+  <updated>{last_update}</updated>
+  <author>
+    <name>Colin McMillen</name>
+  </author>
+  <id>https://www.mcmillen.dev/</id>
+
+  {entries}
+</feed>
+'''
+
+  entry_template = '''
+  <entry>
+    <title>{title}</title>
+    <id>{url}</id>
+    <link rel="alternate" href="{url}"/>
+    <updated>{updated}</updated>
+    <summary>{summary}</summary>
+  </entry>
+'''
+
+  blog_entries.sort(key=operator.itemgetter('date'))
+
+  entries_io = StringIO()
+  last_update = None
+  for entry in blog_entries:
+    # We lie and pretend that all entries were written at noon UTC.
+    update_date = entry['date'] + 'T12:00:00+00:00'
+    last_update = update_date
+    entries_io.write(entry_template.format(
+        url=entry['url'],
+        title=entry['title'],
+        updated=update_date,
+        summary='TODO: fill this out.'))
+
+  entries_text = entries_io.getvalue()
+
+  atom_feed = atom_template.format(
+      last_update=last_update,
+      entries=entries_io.getvalue())
+  entries_io.close()
+
+  atom_filename = os.path.join(output_directory, 'atom.xml')
+  print_file('', atom_filename)
+  atom_file = open(atom_filename, 'w')
+  atom_file.write(atom_feed)
+  atom_file.close()
+
+
+def copy_site():
+  os.system('cp -r output/* %s' % deploy_directory)
 
 
 def deploy_site():
-  os.system('cp -r output/* %s' % deploy_directory)
+  copy_site()
   os.chdir(os.path.expanduser(deploy_directory))
   os.system('git add .')
   os.system('git commit -m "automated update from build.py"')
@@ -116,9 +196,12 @@ def main():
   parser.add_argument(
       '--fast', action='store_true',
       help='only rebuild content files')
+  parser.add_argument(
+      '--copy', action='store_true',
+      help='copy output files to www-home git repo')
   parser.add_argument(
       '--deploy', action='store_true',
-      help='deploy the site by pushing to the www-home git repo')
+      help='deploy the site by pushing the www-home git repo to production')
   args = parser.parse_args()
 
   if args.clean:
@@ -128,7 +211,10 @@ def main():
     copy_static_files()
   process_markdown_files()
   make_sitemap()
-  make_rss()
+  make_atom_feed()
+
+  if args.copy and not args.deploy:
+    copy_site()
 
   if args.deploy:
     if args.fast:

content/index.md

@@ -32,6 +32,7 @@ For more details, check my [resumé](resume.html) & [publications](publications.
   }
 </script>
 
+
 ## Latest Updates
 
 You can sign up for occasional updates via my [email newsletter](https://tinyletter.com/mcmillen).

template.html

@@ -6,6 +6,7 @@
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <link rel="icon" type="image/png" href="/favicon.ico">
 <link rel="canonical" href="https://www.mcmillen.dev/{page_url}">
+<link rel="alternate" type="application/atom+xml" href="https://www.mcmillen.dev/atom.xml" title="Colin McMillen's Blog - Atom">
 
 <title>{title}</title>