new-codigoparallevar/scripts/blog.py

#!/usr/bin/env python3

MARKDOWN_EXTENSION = '.md'
EXTENSIONS = [
    MARKDOWN_EXTENSION,
]

MARKDOWN_EXTRA_FEATURES = [
    # See more in: https://python-markdown.github.io/extensions/
    'markdown.extensions.fenced_code',
    'markdown.extensions.codehilite',
    'markdown.extensions.extra',
]

import json
import logging
import sys
import os
import datetime
import jinja2
import shutil

import yaml
import markdown
import re
from unidecode import unidecode

NIKOLA_DATE_RE = re.compile(r'^([0-2]\d|30|31)\.(0\d|1[012])\.(\d{4}), (\d{1,2}):(\d{2})$')

COMPLETE_DATE_RE = re.compile(r'^(\d{4})-(0\d|1[012])-([0-2]\d|30|31) '
                            + r'(\d{2}):(\d{2})(:\d{2})( .+)?$')
SLUG_HYPHENATE_RE = re.compile(r'[\s\-]+')
SLUG_REMOVE_RE = re.compile(r'[^\s\-a-zA-Z0-9]*')

ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

STATIC_PATH = os.path.join(ROOT_DIR, 'static')
ARTICLE_TEMPLATE_NAME = 'article.tmpl.html'
STATIC_RESOURCES = (
    ('style.css', 'css/style.css'),
)

JINJA_ENV = jinja2.Environment(
    loader=jinja2.FileSystemLoader(STATIC_PATH),
    autoescape=jinja2.select_autoescape()
)
ARTICLE_TEMPLATE = JINJA_ENV.get_template(ARTICLE_TEMPLATE_NAME)

def parse_nikola_date(match):
    return datetime.datetime(year=int(match.group(3)),
                             month=int(match.group(2)),
                             day=int(match.group(1)),
                             hour=int(match.group(4)),
                             minute=int(match.group(5)))

def parse_complete_date(match):
    return datetime.datetime.strptime(match.group(0), '%Y-%m-%d %H:%M:%S %Z%z')

def slugify(title):
    """
    Made for compatibility with Nikola's slugify within CodigoParaLlevar blog.
    """
    slug = unidecode(title).lower()
    slug = SLUG_REMOVE_RE.sub('', slug)
    slug = SLUG_HYPHENATE_RE.sub('-', slug)

    return slug.strip()

def read_markdown(path):
    with open(path, 'rt') as f:
        data = f.read()
    if data.startswith('---'):
        start = data.index('\n')
        if '---\n' not in data[start:]:
            raise Exception('Front matter not finished on: {}'.format(path))
        front_matter_str, content = data[start:].split('---\n', 1)
        front_matter = yaml.load(front_matter_str, Loader=yaml.SafeLoader)
    else:
        raise Exception('Front matter is needed for proper rendering. Not found on: {}'.format(
            path
        ))
    doc = markdown.markdown(content, extensions=MARKDOWN_EXTRA_FEATURES)
    return doc, front_matter


def get_out_path(front_matter):
    if 'date' in front_matter:
        if m := NIKOLA_DATE_RE.match(front_matter['date']):
            front_matter['date'] = parse_nikola_date(m)
        elif m := COMPLETE_DATE_RE.match(front_matter['date']):
            front_matter['date'] = parse_complete_date(m)
        else:
            raise NotImplementedError('Unknown date format: {}'.format(
                front_matter['date']))
    else:
        raise Exception('No date found on: {}'.format(
            path
        ))

    if 'slug' not in front_matter:
        if 'title' not in front_matter:
            raise Exception('No title found on: {}'.format(
                path
            ))

        front_matter['slug'] = slugify(front_matter['title'])

    out_path = os.path.join(str(front_matter['date'].year), front_matter['slug'])
    return out_path


def load_all(top_dir_relative):
    top = os.path.abspath(top_dir_relative)

    docs = []

    for root, dirs, files in os.walk(top):
        for name in files:
            if all([not name.endswith(ext) for ext in EXTENSIONS]):
                # The logic is negative... but it works
                continue

            if name.endswith(MARKDOWN_EXTENSION):
                path = os.path.join(root, name)
                doc, front_matter = read_markdown(path)
                out_path = get_out_path(front_matter)
                docs.append((doc, front_matter, out_path))
            else:
                raise NotImplementedError('Unknown filetype: {}'.format(name))

    return docs

def render_article(doc, f):
    result = ARTICLE_TEMPLATE.render(content=doc)
    f.write(result)

def main(source_top, dest_top):
    docs = load_all(source_top)
    for (doc, front_matter, out_path) in docs:
        doc_full_path = os.path.join(dest_top, out_path)
        os.makedirs(os.path.dirname(doc_full_path), exist_ok=True)
        print("==", doc_full_path)
        with open(doc_full_path + '.html', 'wt') as f:
            render_article(doc, f)

    for src, dest in STATIC_RESOURCES:
        target_dest = os.path.join(dest_top, dest)
        os.makedirs(os.path.dirname(target_dest), exist_ok=True)
        shutil.copy(os.path.join(STATIC_PATH, src), target_dest)

if __name__ == "__main__":
    if len(sys.argv) != 3:
        print("Usage: {} SOURCE_TOP DEST_TOP".format(sys.argv[0]))
        exit(0)

    logging.basicConfig(level=logging.INFO, format="%(levelname)-8s %(message)s")
    main(sys.argv[1], sys.argv[2])