tags!

notes2web.py

@@ -2,6 +2,7 @@
 
 
 from bs4 import BeautifulSoup as bs
+import frontmatter
 import magic
 import sys
 import pathlib
@@ -42,6 +43,9 @@ def get_dirs(folder):
 
     return r
 
+def update_required(src_filename, output_filename):
+    return not os.path.exists(output_filename) or os.path.getmtime(src_filename) > os.path.getmtime(output_filename)
+
 
 def get_args():
     """ Get command line arguments """
@@ -57,6 +61,8 @@ def get_args():
     parser.add_argument('-I', '--template-index-foot', type=pathlib.Path, default=pathlib.Path('/opt/notes2web/templates/indexfoot.html'))
     parser.add_argument('-s', '--stylesheet', type=pathlib.Path, default=pathlib.Path('/opt/notes2web/styles.css'))
     parser.add_argument('-e', '--extra-index-content', type=pathlib.Path, default=pathlib.Path('/opt/notes2web/templates/extra_index_content.html'))
+    parser.add_argument('-n', '--index-article-names', action='append', default=['index.md'])
+    parser.add_argument('-F', '--force', action="store_true", help="Generate new output html even if source file was modified before output html")
     return parser.parse_args()
 
 
@@ -87,22 +93,54 @@ def main(args):
 
     markdown_files, plaintext_files, other_files = get_files(args.notes)
 
+    print(f"{args.index_article_names=}")
+
+    dirs_with_index_article = []
+
     print(f"{markdown_files=}")
+    tag_dict = {}
     for filename in markdown_files:
         print(f"{filename=}")
-        html = pypandoc.convert_file(filename, 'html', extra_args=[f'--template={args.template}'])
+        print(f"{os.path.basename(filename)=}")
-        output_filename = os.path.splitext(re.sub(f"^{args.notes.name}", args.output_dir.name, filename))[0] + '.html'
-        os.makedirs(os.path.dirname(output_filename), exist_ok=True)
 
-        with open(output_filename, 'w+') as fp:
+        if os.path.basename(filename) in args.index_article_names:
-            fp.write(html)
+            output_filename = os.path.join(
+                    os.path.dirname(re.sub(f"^{args.notes.name}", args.output_dir.name, filename)),
+                    'index.html'
+                    )
+            dirs_with_index_article.append(os.path.dirname(re.sub(f"^{args.notes.name}", args.output_dir.name, filename)))
+        else:
+            output_filename = os.path.splitext(re.sub(f"^{args.notes.name}", args.output_dir.name, filename))[0] + '.html'
+
+        fm = frontmatter.load(filename)
+        if isinstance(fm.get('tags'), list):
+            for tag in fm.get('tags'):
+                if tag in tag_dict.keys():
+                    tag_dict[tag].append({
+                        'path': str(pathlib.Path(*pathlib.Path(output_filename).parts[1:])),
+                        'title': fm.get('title')
+                        })
+                else:
+                    tag_dict[tag] = [ {
+                        'path': str(pathlib.Path(*pathlib.Path(output_filename).parts[1:])),
+                        'title': fm.get('title')
+                        } ]
+        print(f"{output_filename=}")
+
+        if update_required(filename, output_filename) or args.force:
+            html = pypandoc.convert_file(filename, 'html', extra_args=[f'--template={args.template}'])
+            os.makedirs(os.path.dirname(output_filename), exist_ok=True)
+
+            with open(output_filename, 'w+') as fp:
+                fp.write(html)
 
     print(f"{plaintext_files=}")
     for filename in plaintext_files:
+        title = os.path.basename(re.sub(f"^{args.notes.name}", args.output_dir.name, filename))
         output_filename = re.sub(f"^{args.notes.name}", args.output_dir.name, filename) + '.html'
         os.makedirs(os.path.dirname(output_filename), exist_ok=True)
-        title = os.path.basename(output_filename)
         html = re.sub(r'\$title\$', title, TEXT_ARTICLE_TEMPLATE_HEAD)
+        html = re.sub(r'\$h1title\$', title, html)
         html = re.sub(r'\$raw\$', os.path.basename(filename), html)
         with open(filename) as fp:
             html += fp.read()
@@ -117,19 +155,38 @@ def main(args):
         os.makedirs(os.path.dirname(output_filename), exist_ok=True)
         shutil.copyfile(filename, output_filename)
 
+    tagdir = os.path.join(args.output_dir, '.tags')
+    os.makedirs(tagdir, exist_ok=True)
+
+    for tag in tag_dict.keys():
+        html = re.sub(r'\$title\$', f'{tag}', INDEX_TEMPLATE_HEAD)
+        html = re.sub(r'\$h1title\$', f'tag: {tag}', html)
+        html = re.sub(r'\$extra_content\$', '', html)
+
+        for entry in tag_dict[tag]:
+            html += f"<div class=\"article\"><a href=\"/{entry['path']}\">{entry['title']}</a></div>"
+        html += INDEX_TEMPLATE_FOOT
+
+        with open(os.path.join(tagdir, f'{tag}.html'), 'w+') as fp:
+            fp.write(html)
+
+
 
     dirs_to_index = [args.output_dir.name] + get_dirs(args.output_dir)
     print(f"{dirs_to_index=}")
     print(f"{os.path.commonpath(dirs_to_index)=}")
 
     for directory in dirs_to_index:
+        if directory in dirs_with_index_article:
+            continue
         paths = os.listdir(directory)
         print(f"{paths=}")
 
         indexentries = []
         
         for path in paths:
-            if path == 'index.html':
+            print(f"{path=}")
+            if path in [ 'index.html', '.git' ]:
                 continue
 
             fullpath = os.path.join(directory, path)
@@ -141,8 +198,11 @@ def main(args):
                     title = soup.find('title').get_text()
                 except AttributeError:
                     title = path
-            else:
+            elif os.path.isdir(fullpath):
                 title = path
+            else:
+                # don't add plaintext files to index, since they have a html wrapper
+                continue
 
             if title.strip() == '':
                 title = path
@@ -157,6 +217,7 @@ def main(args):
         indexentries.sort(key=lambda entry: entry['isdirectory'], reverse=True)
         
         html = re.sub(r'\$title\$', directory, INDEX_TEMPLATE_HEAD)
+        html = re.sub(r'\$h1title\$', directory, html)
         html = re.sub(r'\$extra_content\$',
                 EXTRA_INDEX_CONTENT if directory == os.path.commonpath(dirs_to_index) else '',
                 html
@@ -170,6 +231,7 @@ def main(args):
             fp.write(html)
 
     shutil.copyfile(args.stylesheet, os.path.join(args.output_dir.name, 'styles.css'))
+    print(tag_dict)
 
     return 0
 

readme.md

@@ -4,13 +4,26 @@ View your notes as a static html site.
 
 ![](./screenshot.png)
 
+
+## Why?
+
+I want to be able to view my notes in a more convenient way.
+I was already writing them in Pandoc markdown and could view them as PDFs but that wasn't quite
+doing it for me:
+
+- It was inconvenient to flick through multiple files of notes to find the right PDF
+- It was annoying to sync to my phone
+- PDFs do not scale so they were hard to read on smaller screens
+- Probably more reasons I can't think of right now
+
+
 ## Install
 
-0. Install [Pandoc](https://pandoc.org/index.html) and [yq](https://github.com/mikefarah/yq)
+0. Install [Pandoc](https://pandoc.org/index.html) and [Pip](https://github.com/pypa/pip)
 
    On arch:
    ```
-   # pacman -S pandoc yq
+   # pacman -S pandoc python-pip
    ```
 
 1. Run `make install` as root
@@ -18,7 +31,7 @@ View your notes as a static html site.
 ## Usage
 
 ```
-$ notes2web.py NOTES_DIRECTORY_1
+$ notes2web.py notes_directory
 ```
 
 Output of `notes2web.py --help`:
@@ -27,7 +40,7 @@ Output of `notes2web.py --help`:
 usage: notes2web.py [-h] [-o OUTPUT_DIR] [-t TEMPLATE] [-H TEMPLATE_TEXT_HEAD]
                     [-f TEMPLATE_TEXT_FOOT] [-i TEMPLATE_INDEX_HEAD]
                     [-I TEMPLATE_INDEX_FOOT] [-s STYLESHEET]
-                    [-e EXTRA_INDEX_CONTENT]
+                    [-e EXTRA_INDEX_CONTENT] [-n INDEX_ARTICLE_NAMES]
                     notes
 
 positional arguments:
@@ -43,6 +56,7 @@ optional arguments:
   -I TEMPLATE_INDEX_FOOT, --template-index-foot TEMPLATE_INDEX_FOOT
   -s STYLESHEET, --stylesheet STYLESHEET
   -e EXTRA_INDEX_CONTENT, --extra-index-content EXTRA_INDEX_CONTENT
+  -n INDEX_ARTICLE_NAMES, --index-article-names INDEX_ARTICLE_NAMES
 ```
 
 The command will generate a website in the `output-dir` directory (`./web` by default).

requirements.txt

@@ -1,3 +1,6 @@
 beautifulsoup4==4.9.3
 pypandoc==1.5
+python-frontmatter==1.0.0
+python-magic==0.4.24
+PyYAML==5.4.1
 soupsieve==2.2.1

styles.css

@@ -35,6 +35,7 @@ pre {
 	background-color: #d9d9d9 ;
 	color: #000;
 	padding: 1em;
+	overflow-x: scroll;
 }
 
 details {

templates/article.html

@@ -9,6 +9,12 @@
 </head>
 
 <body>
+	<p style="font-size: 0.7em"> tags:
+	$for(tags)$
+	<a href="/.tags/$it$.html">$it$</a>$sep$,
+	$endfor$
+	</p>
 	$body$
+
 	<p style="font-size: 0.7em;"> page generated by <a href="https://github.com/alvierahman90/notes2web">notes2web</a></p>
 </body>

templates/extra_index_content.html

@@ -1 +1,4 @@
-<p>These are my personal notes. Correctness is not guaranteed.</p>
+<p>
+These are my personal notes. Correctness is not guaranteed.
+Browse by tag <a href="/.tags">here</a>.
+</p>

templates/index.html

@@ -6,7 +6,7 @@
 <title>$title$</title>
 </head>
 <body>
-	<h1>$title$</h1>
+	<h1>$h1title$</h1>
 $body$
 
 <p style="font-size: 0.7em;"> page generated by <a href="https://github.com/alvierahman90/notes2web">notes2web</a></p>