From 72ed1b208d4fd148ea45e56f1d2b2378790f7d7d Mon Sep 17 00:00:00 2001
From: Kévin Le Gouguec <kevin.legouguec@gmail.com>
Date: Thu, 16 Apr 2020 23:35:30 +0200
Subject: Touch up web page titles

Still a few wrinkles to iron out, e.g. the toplevel page's <title>.
---
 repo/www/generate-index.py | 99 ++++++++++++++++++++++++++++++----------------
 1 file changed, 64 insertions(+), 35 deletions(-)

(limited to 'repo/www/generate-index.py')

diff --git a/repo/www/generate-index.py b/repo/www/generate-index.py
index acbeb52..62daf79 100755
--- a/repo/www/generate-index.py
+++ b/repo/www/generate-index.py
@@ -1,8 +1,9 @@
 #!/usr/bin/env python3
 
 from argparse import ArgumentParser
+from itertools import chain
 import json
-from os import path
+from pathlib import Path
 from subprocess import run
 
 from git import Repo
@@ -16,7 +17,10 @@ def parse_arguments():
         'site_tree', help='JSON file describing the page tree.'
     )
     parser.add_argument(
-        'target', help='Pathspec to generate an index for.'
+        'site_title', help='Title to add to <title>.'
+    )
+    parser.add_argument(
+        'target', help='Subfolder to generate an index for.'
     )
     parser.add_argument(
         'output', help='Path to the output file.'
@@ -30,55 +34,80 @@ def list_files(tree_file, folder):
     return directories[folder].subfolders, directories[folder].files
 
 
-def generate_index_page(title, directories, files, intro_text):
-    dir_list = '\n'.join(
-        f'- [{d}/]({d}/index.html)' for d in directories
-    )
-
-    file_list = '\n'.join(
-        f'- [{f}]({f}.html)' for f in files
+def has_title(markdown):
+    pandoc = run(
+        ('pandoc', '-t', 'json'), input=markdown,
+        check=True, text=True, capture_output=True
     )
+    ast = json.loads(pandoc.stdout)
+    return 'title' in ast['meta']
 
-    return f'''\
-% {title}
-
-{intro_text}
 
-{dir_list}
-{file_list}
-'''
+def format_readme_index(readme, title, items):
+    blocks = [
+        readme,
+        f'# {title}',
+        items
+    ]
 
-def convert_page(content, output):
-    run(
-        ('pandoc', '-s', '-o', output), input=content, check=True, text=True
-    )
+    if not has_title(readme):
+        blocks.insert(0, '% README')
 
+    return '\n\n'.join(blocks)
 
-def main(arguments):
-    target = arguments.target
 
-    folders, files = list_files(arguments.site_tree, target)
+def format_plain_index(title, items):
+    return '\n'.join((f'% {title}', items))
 
-    names = []
-    intro_text = ''
 
-    repo = Repo(search_parent_directories=True)
+def list_pages(files):
+    readme = None
+    pages = []
 
     for f in files:
-        name, _ = path.splitext(f)
-        if name == 'README':
-            with open(path.join(repo.working_dir, target, f)) as intro:
-                intro_text = intro.read()
+        page = Path(f).stem
+
+        if page == 'README':
+            readme = f
         else:
-            names.append(name)
+            pages.append(page)
+
+    return pages, readme
+
 
-    title = path.basename(target) if target else 'index'
+def format_index(target, directories, files):
+    pages, readme = list_pages(files)
 
-    convert_page(
-        generate_index_page(title, folders, names, intro_text),
-        arguments.output
+    dir_list = (
+        f'- [{d}/]({d}/index.html)' for d in directories
+    )
+    page_list = (
+        f'- [{p}]({p}.html)' for p in pages
+    )
+    items = '\n'.join(chain(dir_list, page_list))
+
+    title = f'Index for {target}' if target else 'Index'
+
+    if readme is None:
+        return format_plain_index(title, items)
+
+    repo_top = Repo(search_parent_directories=True).working_dir
+    intro = Path(repo_top, target, readme).read_text()
+    return format_readme_index(intro, title, items)
+
+
+def convert_page(content, output, site_title):
+    run(
+        ('pandoc', '-s', '-o', output, '-T', site_title),
+        input=content, check=True, text=True
     )
 
 
+def main(arguments):
+    folders, files = list_files(arguments.site_tree, arguments.target)
+    index_page = format_index(arguments.target, folders, files)
+    convert_page(index_page, arguments.output, arguments.site_title)
+
+
 if __name__ == '__main__':
     main(parse_arguments())
-- 
cgit v1.2.3