#!/usr/bin/env python3 from collections import namedtuple, Counter from datetime import date from functools import partial from glob import glob from itertools import chain from jinja2 import Environment, FileSystemLoader, Markup from markdown import markdown from operator import attrgetter from subprocess import call as run from sys import argv import meta import os CONTENT = './content' OUTPUT = './output' KWORDS = { 'Tomasz Kramkowski', 'the-tk', 'blog', 'projects', 'TK', 'Arch-TK' } ERRORS = { 403: "Forbidden", 404: "File Not Found", 500: "Internal Server Error", 502: "Bad Gateway", 503: "Service Unavailable", 504: "Gateway Timeout", } md = lambda s: Markup(markdown(s, extensions=['markdown.extensions.fenced_code', 'markdown.extensions.codehilite'])) File = namedtuple('File', 'content date slug ext') def parse(path, has_date=False): with open(path) as f: content = meta.parse(f.read()) slug, ext = os.path.splitext(os.path.split(path)[1]) if has_date: y, m, d, slug = slug.split('-', 3) dat = date(int(y), int(m), int(d)) else: dat = None return File(content, dat, slug, ext) Post = namedtuple('Post', 'title date tags pre content location') def readpost(path): pf = parse(path, has_date=True) location = os.path.join('post', '{:04}'.format(pf.date.year), '{:02}'.format(pf.date.month), '{:02}'.format(pf.date.day), pf.slug + '.html') tags = set(pf.content.meta.get('tags', [])) return Post(' '.join(pf.content.meta['title']), pf.date, tags, pf.content.sects['pre'], '\n'.join(pf.content.sects.values()), location) Project = namedtuple('Project', 'name description source ml aur content location') def readproject(path): pf = parse(path) source = pf.content.meta.get('source', [None])[0] ml = pf.content.meta.get('ml', [None])[0] aur = pf.content.meta.get('aur', [None])[0] content = '\n'.join(pf.content.sects.values()) location = os.path.join('project', pf.slug + '.html') return Project(' '.join(pf.content.meta['name']), pf.content.sects.get('desc', content), source, ml, aur, content, location); def output(dest, cont): dest = os.path.join(OUTPUT, dest) destdir = os.path.dirname(dest) if not os.path.exists(destdir): os.makedirs(destdir) with open(dest, 'w') as f: f.write(cont) def outputpost(env, post, tags): page = { 'title': post.title, 'description': md(post.pre), 'keywords': ', '.join(KWORDS | post.tags), 'date': post.date, 'content': md(post.content), 'tags': sorted(post.tags & set(tags)), } output(post.location, env.get_template('post.html').render(page=page)) def outputtag(env, posts, tag): page = { 'title': 'Posts tagged - {}'.format(tag), 'description': "List of posts tagged '{}'.".format(tag), 'keywords': ', '.join(KWORDS | {tag}), 'tag': tag, 'posts': [ p for p in posts if tag in p.tags ], } output('tag/{}.html'.format(tag), env.get_template('tag.html').render(page=page)) def outputproject(env, project): page = { 'title': project.name, 'description': project.description.replace('\n', ' '), 'keywords': ', '.join(KWORDS | {project.name}), 'content': md(project.content), 'source': project.source, 'ml': project.ml, 'aur': project.aur, } output(project.location, env.get_template('project.html').render(page=page)) def main(argv=['generate.py']): from getopt import getopt global OUTPUT opts = dict(getopt(argv[1:], "o:b:")[0]) OUTPUT = opts.get('-o', OUTPUT) baseurl = opts.get('-b', 'https://the-tk.com') run(['rm', '-rf', os.path.join(OUTPUT, '*')]) env = Environment(loader=FileSystemLoader('./templates'), autoescape=True) env.globals['site'] = {'baseurl': baseurl} env.filters['datefmt'] = lambda d, f='%Y-%m-%d': d.strftime(f) posts = glob(os.path.join(CONTENT, 'posts', '*')) posts = [p for p in posts if not os.path.split(p)[1].startswith('_')] posts = sorted(map(readpost, posts), key=attrgetter('date'), reverse=True) tags = Counter(chain(*(p.tags for p in posts))) tags = sorted(k for k, v in tags.items() if v > 1) page = { 'title': 'Blog', 'description': 'New posts will appear chronologically on this page.', 'keywords': ', '.join(KWORDS), 'tags': tags, 'posts': posts, } output('blog.html', env.get_template('blog.html').render(page=page)) projects = glob(os.path.join(CONTENT, 'projects', '*')) projects = [p for p in projects if not os.path.split(p)[1].startswith('_')] projects = sorted(map(readproject, projects), key=lambda e: e.name.lower()) page = { 'title': 'Projects', 'description': 'This is a listing of all of my notable projects.', 'keywords': ', '.join(KWORDS), 'projects': projects, } output('projects.html', env.get_template('projects.html').render(page=page)) page = { 'title': 'the-tk.com', 'description': 'This is my website. Here you can find some blog posts and descriptions for notable projects.', 'keywords': ', '.join(KWORDS), 'posts': posts[:3], } output('index.html', env.get_template('index.html').render(page=page)) for post in posts: outputpost(env, post, tags) for tag in tags: outputtag(env, posts, tag) for project in projects: outputproject(env, project) for code, error in ERRORS.items(): page = { 'title': 'Error: {} - {}'.format(code, error), 'code': code, 'error': error, } output('error/{}.html'.format(code), env.get_template('error.html').render(page=page)) run(['scss', os.path.join(CONTENT, 'style.scss'), os.path.join(OUTPUT, 'style.css')]) run(['cp', '-r', os.path.join(CONTENT, 'images'), OUTPUT]) if __name__ == '__main__': from sys import argv main(argv)