Improve link checking.

scripts/blog.py

@@ -63,6 +63,10 @@ JINJA_ENV = jinja2.Environment(
     autoescape=jinja2.select_autoescape()
 )
 
+WATCH = True
+if os.getenv('WATCH_AND_REBUILD', '1') == '0':
+    WATCH = False
+
 def update_statics():
     global ARTICLE_TEMPLATE
     ARTICLE_TEMPLATE = JINJA_ENV.get_template(ARTICLE_TEMPLATE_NAME)
@@ -122,6 +126,7 @@ def slugify(title):
     slug = unidecode(title).lower()
     slug = SLUG_REMOVE_RE.sub('', slug)
     slug = SLUG_HYPHENATE_RE.sub('-', slug)
+    slug = slug.strip('-')
 
     return slug.strip()
 
@@ -167,7 +172,7 @@ def get_out_path(front_matter):
 
     out_path = os.path.join(str(front_matter['date'].year), front_matter['slug'])
     if front_matter.get('lang', LANG_PRIORITY[0]) != LANG_PRIORITY[0]:
-        out_path = os.path.join(str(front_matter['date'].year), front_matter['lang'], front_matter['slug'])
+        out_path = os.path.join(front_matter['lang'], str(front_matter['date'].year), front_matter['slug'])
     return out_path
 
 
@@ -291,15 +296,6 @@ def summarize(doc):
         for child in summary.children:
             result.append(child)
 
-    # Update summary links and hrefs
-    for v in result.find_all('video') + result.find_all('image'):
-        if 'src' in v.attrs and ':' not in v['src']:
-            v['src'] = '/blog/' + v['src'].lstrip('/')
-
-    for v in result.find_all('a'):
-        if 'href' in v.attrs and ':' not in v['href']:
-            v['href'] = '/blog/' + v['href'].lstrip('/')
-
     return result
 
 def render_index(docs, dest_top):
@@ -388,7 +384,7 @@ def render_categories(docs, dest_top):
         result = CATEGORY_LIST_TEMPLATE.render(
             posts=posts,
         )
-        path = os.path.join(dest_top, "tags", tag, "index.html")
+        path = os.path.join(dest_top, "tags", tag.replace('/', '_'), "index.html")
         os.makedirs(os.path.dirname(path), exist_ok=True)
         with open(path, 'wt') as f:
             f.write(result)
@@ -520,6 +516,10 @@ def main(source_top, dest_top):
     docs = regen_all(source_top, dest_top)
     logging.info("Initial load completed in {:.2f}s".format(time.time() - t0))
 
+    if not WATCH:
+        logging.info("Build completed in {:.2f}s".format(time.time() - t0))
+        return 0
+
     ## Updating
     for event in notifier.event_gen(yield_nones=False):
         (ev, types, directory, file) = event
@@ -573,11 +573,12 @@ def main(source_top, dest_top):
             docs[filepath] = (doc, front_matter, out_path)
             doc_full_path = os.path.join(dest_top, out_path)
             print("Updated: {}.html".format(doc_full_path))
-            os.makedirs(os.path.dirname(doc_full_path), exist_ok=True)
+            os.makedirs(os.path.dirname(doc_full_path + '/index.html'), exist_ok=True)
             # print("==", doc_full_path)
             with open(doc_full_path + '/index.html', 'wt') as f:
                 try:
                     render_article(doc, front_matter, f, out_path)
+                    render_archive(docs, dest_top)
                 except:
                     logging.error(traceback.format_exc())
                     logging.error("Rendering failed 😿")

scripts/test-links.py

@@ -3,6 +3,7 @@
 import logging
 import os
 import sys
+import urllib.parse
 
 from bs4 import BeautifulSoup as bs4
 
@@ -19,25 +20,32 @@ def main(files_top):
                 print("\r{} files".format(len(found_files)), end='', flush=True)
 
     print()
-    found_broken = False
+    found_broken = 0
     for fpath in tqdm(found_files):
         with open(fpath) as f:
             tree = bs4(f.read(), features='lxml', parser='html5')
-        for link in tree.find_all('a'):
-            if 'href' not in link.attrs:
-                continue
-            if ':' in link['href']:
-                continue
-            if link['href'].startswith('/'):
-                target = link['href']  # TODO: Find a better way to model the root
-            else:
-                target = os.path.join(os.path.dirname(fpath), link['href'])
-            if os.path.isdir(target):
-                pass
-            elif not os.path.exists(target):
-                print("[{}] -[ error ]-> {} | {}".format(fpath, target, link['href']))
+
+        for tag, attr in [('a', 'href'), ('img', 'src'), ('audio', 'src'), ('video', 'src')]:
+            for link in tree.find_all(tag):
+                if attr not in link.attrs:
+                    continue
+                link.attrs[attr] = link.attrs[attr].split('#')[0]
+                if not link.attrs[attr]:
+                    continue
+                if ':' in link[attr]:
+                    continue
+                if link[attr].startswith('/'):
+                    target = os.path.join(os.path.abspath(files_top), urllib.parse.unquote(link[attr].lstrip('/')))
+                else:
+                    target = os.path.join(os.path.dirname(fpath), urllib.parse.unquote(link[attr]))
+                if os.path.isdir(target):
+                    pass
+                elif not os.path.exists(target):
+                    print("[{}] -[ error ]-> {} | {}".format(fpath, target, link[attr]))
+                    found_broken += 1
 
     if found_broken:
+        print(f"Found {found_broken} broken links")
         exit(1)
     else:
         exit(0)

scripts/upload.sh

@@ -13,11 +13,15 @@ cd ../scripts
 rm -Rf ../_gen/notes
 WATCH_AND_REBUILD=0 python3 generate.py ~/.logs/brain ../_gen/notes
 
+rm -Rf ../_gen/blog
+WATCH_AND_REBUILD=0 python3 blog.py ~/cloud/nextcloud/blog/posts/ ../_gen/blog
+
 # Upload notes
 cd ../_gen
 rsync -HPaz static/ --delete-after --exclude='*.html' root@codigoparallevar.com:/mnt/vols/misc/codigoparallevar/static/
 rsync -HPaz notes/ --delete-after --exclude='xapian' --exclude='*.sqlite3' root@codigoparallevar.com:/mnt/vols/misc/codigoparallevar/notes/
 rsync -HPaz notes/db.sqlite3 root@codigoparallevar.com:/mnt/vols/misc/codigoparallevar-api/
+rsync -HPaz blog/ --delete-after --exclude='xapian' --exclude='*.sqlite3' root@codigoparallevar.com:/mnt/vols/misc/codigoparallevar/blog/
 
 # Restart API server
 ssh root@codigoparallevar.com docker restart notes-api-server

static/article.tmpl.html

@@ -38,7 +38,7 @@
             </time>
             <ul class="post-tags">
               {% for post_tag in post_tags %}
-              <li class="post-tag"><a href="{{ base_path }}/tags/{{ post_tag }}/"</a>{{ post_tag }}</a></li>
+              <li class="post-tag"><a href="{{ base_path }}/tags/{{ post_tag |urlencode|replace('/', '_') }}/"</a>{{ post_tag }}</a></li>
               {% endfor %}
             </ul>
           </div>

static/article_list.tmpl.html

@@ -42,7 +42,7 @@
             <div class="post-metadata">
               <ul class="post-tags">
                 {% for post_tag in post.post_tags %}
-                  <li class="post-tag"><a href="../tags/{{ post_tag }}/"</a>{{ post_tag }}</a></li>
+                  <li class="post-tag"><a href="../tags/{{ post_tag |urlencode|replace('/', '_') }}/">{{ post_tag }}</a></li>
                 {% endfor %}
               </ul>
             </div>

static/blog_index.tmpl.html

@@ -40,7 +40,7 @@
               </time>
               <ul class="post-tags">
                 {% for post_tag in post.post_tags %}
-                <li class="post-tag"><a href="tags/{{ post_tag }}/"</a>{{ post_tag }}</a></li>
+                <li class="post-tag"><a href="tags/{{ post_tag |urlencode|replace('/', '_') }}/"</a>{{ post_tag }}</a></li>
                 {% endfor %}
               </ul>
             </div>

static/category_list.tmpl.html

@@ -42,7 +42,7 @@
             <div class="post-metadata">
               <ul class="post-tags">
                 {% for post_tag in post.post_tags %}
-                  <li class="post-tag"><a href="../../tags/{{ post_tag }}/"</a>{{ post_tag }}</a></li>
+                  <li class="post-tag"><a href="../../tags/{{ post_tag |urlencode|replace('/', '_') }}/">{{ post_tag }}</a></li>
                 {% endfor %}
               </ul>
             </div>