www/build.py

#!/usr/bin/env python

import argparse
import copy
import os
import pathlib
import shutil

import bs4
import markdown


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("src", help="source directory", nargs="?", default="./src")
    parser.add_argument("out", help="output directory", nargs="?", default="./out")
    args = parser.parse_args()

    src = pathlib.Path(args.src)
    out = pathlib.Path(args.out)

    md = markdown.Markdown(extensions=["attr_list", "footnotes", "md_in_html", "meta"])

    # Clean the output directory
    if out.exists():
        print("Removing  ", out)
        shutil.rmtree(out)

    # Load the template
    template = bs4.BeautifulSoup(
        (src / ".template.html").read_text(encoding="utf8"),
        features="html.parser")

    count = 0
    for dirpath, _, filenames in os.walk(src):
        dirpath = pathlib.Path(dirpath).relative_to(src)
        for filename in filenames:
            if filename[0] == ".":
                continue  # Skip dotfiles
            count += 1

            path = src / dirpath / filename
            dest = out / dirpath / filename
            os.makedirs(dest.parent, exist_ok=True)

            # Copy any file types without content processing
            if filename.rsplit(".")[-1] not in ("html", "md"):
                print("Copying   ", path)
                shutil.copyfile(path, dest)
                continue

            content = path.read_text(encoding="utf8")
            meta = {}

            # Preprocess markdown into html
            if dest.name.endswith(".md"):
                print("Converting", path)
                md.reset()
                dest = dest.with_suffix(".html")
                content = md.convert(content)
                meta = md.Meta

            if dest.name.endswith("html"):
                # Inject content into the template
                page_content = bs4.BeautifulSoup(content, features="html.parser")
                page = copy.copy(template)
                article = page.new_tag("article")
                article.append(page_content)
                page.article.replace_with(article)

                # Inject path into the nav
                for i in range(len(dirpath.parts)):
                    a = page.new_tag("a")
                    a["href"] = "/" + "/".join(dirpath.parts[:i+1]) + "/"
                    a.string = dirpath.parts[i]
                    page.nav.append(a)
                    page.nav.append(page.new_string("/"))

                # Apply metadata to the template
                if meta_title := meta.get("title"):
                    page.title.string = meta_title[0]
                    page.header.h1.string = meta_title[0]

                if meta_date := meta.get("date"):
                    p = page.new_tag("p")
                    p["class"] = "metadata"
                    p.string = "Date: " + meta_date[0]
                    page.header.append(p)

                if meta_author := meta.get("author"):
                    p = page.new_tag("p")
                    p["class"] = "metadata"
                    p.string = "Author: " + meta_author[0]
                    page.header.append(p)

                if meta_source := meta.get("source"):
                    for source_url in meta_source:
                        a = page.new_tag("a")
                        a["href"] = source_url
                        a.string = source_url
                        p = page.new_tag("p")
                        p["class"] = "metadata"
                        p.string = "URL: "
                        p.append(a)
                        page.header.append(p)

                # The fully templated page is the new content
                content = str(page)

            # Write the destination file
            print("Writing   ", dest)
            dest.write_text(content)

    print("Processed", count, "files")


if __name__ == "__main__":
    main()