blog.ari.lt/scripts/blog.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""blog manager"""

from __future__ import annotations

import datetime
import hashlib
import json
import os
import re
import shutil
import string
import subprocess
import sys
import tempfile
import typing
import xml.etree.ElementTree as etree
from collections import Counter
from glob import iglob
from html import escape as html_escape
from threading import Thread
from timeit import default_timer as code_timer

import mistune
import mistune.core
import mistune.inline_parser
import mistune.plugins
import unidecode
import web_mini
from readtime import of_markdown as read_time_of_markdown  # type: ignore
from readtime.result import Result as MarkdownResult  # type: ignore

# from warnings import filterwarnings as filter_warnings


__version__: typing.Final[int] = 2
GEN: typing.Final[str] = f"ari-web blog generator version {__version__}"


OK: typing.Final[int] = 0
ER: typing.Final[int] = 1

CONFIG_FILE: typing.Final[str] = "blog.json"
DEFAULT_CONFIG: dict[str, typing.Any] = {
    "title": "blog",
    "header": "blog",
    "description": "my blog page",
    "posts-dir": "b",
    "assets-dir": "content",
    "rss-file": "rss.xml",
    "blog-keywords": [
        "blog",
        "blog page",
        "blog post",
        "personal",
        "website",
    ],
    "default-keywords": [
        "blog",
        "blog page",
        "blog post",
        "personal",
        "website",
    ],
    "website": "https://example.com",
    "blog": "https://blog.example.com",
    "source": "/git",
    "visitor-count": "/visit",
    "comment": "/c",
    "theme": {
        "primary": "#000",
        "secondary": "#fff",
        "type": "dark",
    },
    "manifest": {
        "icons": [
            {
                "src": "/favicon.ico",
                "sizes": "128x128",
                "type": "image/x-icon",
            },
        ],
    },
    "author": "John Doe",
    "email": "me@example.com",
    "locale": "en_GB",
    "recents": 14,
    "indent": 4,
    "markdown-plugins": [  # good defaults
        "speedup",
        "strikethrough",
        "insert",
        "superscript",
        "subscript",
        "footnotes",
        "abbr",
    ],
    "editor": ["vim", "--", "%s"],
    "context-words": [
        "the",
        "a",
        "about",
        "etc",
        "on",
        "at",
        "in",
        "by",
        "its",
        "i",
        "to",
        "my",
        "of",
        "between",
        "because",
        "of",
        "or",
        "how",
        "to",
        "begin",
        "is",
        "this",
        "person",
        "important",
        "homework",
        "and",
        "cause",
        "how",
        "what",
        "for",
        "with",
        "without",
        "using",
        "im",
    ],
    "wslug-limit": 10,
    "slug-limit": 96,
    "license": "GPL-3.0-or-later",
    "recent-title-trunc": 16,
    "server-host": "127.0.0.1",
    "server-port": 8080,
    "post-preview-size": 196,
    "read-wpm": 150,
    "top-words": 64,
    "top-tags": 64,
    "legal": '<b>Legal disclaimer</b>: The content of this blog post is my personal opinion, experience, and thoughts as an individual sharing her perspective on this topic. I am not a professional or expert in this area, and the information provided in this blog post is for general informational, educational, entertainment, and recreational purposes only. Accuracy is strived for, but no warranty is given in respect to the reliability or completeness of such content, express or implied. Your acceptance of any reliance on any part of such content from the blog is entirely at your own risk. Under no circumstances shall I, and any of my representatives be held liable for damages arising from access and use of the blog post or linked sites and sources. Furthermore, there might be some links from this blog post to other websites; I have no control over those and do not endorse their content. You are responsible for your actions and subsequent results in using this material. This disclaimer is provided as per the legal basis underlying <a href="https://ari.lt/legal">https://ari.lt/legal</a> and should be considered in further reference to privacy, data security, and liability. Your use of this blog post acknowledges your acceptance of the conditions, and exempts me from any liability that comes from any possible inaccuracy, incompleteness, unreliability, inconsistencies, (un)availability, or any other errors in the information provided herein.',
    "posts": {},
}

NCI: bool = "CI" not in os.environ
NOCLR: bool = "NOCLR" in os.environ

LOG_CLR: str = "\033[90m"
ERR_CLR: str = "\033[1m\033[31m"
NEW_CLR: str = "\033[1m\033[32m"
IMP_CLR: str = "\033[1m\033[35m"

HTML_BEGIN: typing.Final[
    str
] = """<!DOCTYPE html>
<html lang="{lang}">
<head>
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="icon" href="/favicon.ico" sizes="128x128" type="image/x-icon" />
<meta
    name="keywords"
    content="{keywords}"
/>
<meta
  name="robots"
  content="follow, index, max-snippet:-1, max-video-preview:-1, max-image-preview:large"
/>
<meta name="color-scheme" content="{theme_type}" />
<meta name="theme-color" content="{theme_primary}" />
<link rel="manifest" href="/manifest.json" />
<link rel="canonical" href="{blog}/{path}">
<style type="text/css">
:root{{color-scheme:{theme_type};--b:{theme_primary};--f:{theme_secondary}}}\
html{{background-color:var(--b);color:var(--f)}}{critical_css}
</style>
<link
  href="/{styles}"
  rel="preload"
  referrerpolicy="no-referrer"
  type="text/css"
  as="style"
  onload="this.onload=null;this.rel='stylesheet'"
/>
<noscript>
<link
  href="/{styles}"
  rel="stylesheet"
  referrerpolicy="no-referrer"
  type="text/css"
/>
</noscript>
<link
  href="/{rss}"
  referrerpolicy="no-referrer"
  title="{blog_title}"
  rel="alternate"
  type="application/rss+xml"
/>
<meta name="author" content="{author}" />
<meta name="generator" content="{gen}" />
<meta property="og:locale" content="{locale}" />
<meta name="license" content="{license}">
<link rel="sitemap" href="/sitemap.xml" type="application/xml">"""

POST_TEMPLATE: typing.Final[str] = (
    HTML_BEGIN
    + """
<style type="text/css">{post_critical_css}</style>
<title>{blog_title} -> {post_title}</title>
<meta name="description" content="{post_title} by {author} at {post_creation_time} GMT -- {post_description}" />
<meta property="article:read_time" content="{post_read_time}" />
<meta property="og:type" content="article" />
</head>

<body>
<header role="group">
 <h1 role="heading" aria-level="1">{post_title}</h1>

 <nav id="info-bar" role="menubar">
   <a role="menuitem" aria-label="skip" href="#main">skip</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem"><time>{post_creation_time}</time> GMT{post_edit_time}</span>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem" >visitor <img src="{visitor_count}" alt="visitor count" /></span>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem"><time>{post_read_time}</time> read</span>
   <br role="seperator" aria-hidden="true" />

   <a role="menuitem" href="/">home</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="/stats">stats</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{comment}">comment</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{website}">website</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{source}">src</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="/{rss}">rss</a>

   <hr aria-hidden="true" role="seperator" />
 </nav>
</header>
<main> <article id="main"><p id=legal>{legal}</p> {post_content}</article> </main>
<footer><p>{author} &lt;<a href="mailto:{email}">{email}</a>&gt; + {license}</p></footer>
</body>
</html>"""
)

INDEX_TEMPLATE: typing.Final[str] = (
    HTML_BEGIN
    + """
<title>{blog_title}</title>
<meta name="description" content="{blog_description}" />
<meta property="og:type" content="website" />
</head>

<body>
<header role="group">
 <h1 role="heading" aria-level="1">{blog_header}</h1>

 <nav id="info-bar" role="menubar">
   <a role="menuitem"
     aria-label="skip"
     href="#main">skip</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem">latest post : <a href="/{latest_post_path}">{latest_post_title_trunc}</a> at <time>{latest_post_creation_time}</time> GMT</span>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem">visitor <img src="{visitor_count}" alt="visitor count" /></span>
   <br role="seperator" aria-hidden="true" />

   <a role="menuitem" href="/stats">stats</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{comment}">comment</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{website}">website</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{source}">src</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="/{rss}">rss</a>

   <hr aria-hidden="true" role="seperator" />
 </nav>
</header>
<main> <article id="main"><ol reversed id=blist>{blog_list}</ol></article> </main>
<footer><p>{author} &lt;<a href="mailto:{email}">{email}</a>&gt; + {license}</p></footer>
</body>
</html>"""
)

STATS_TEMPLATE: typing.Final[str] = (
    HTML_BEGIN
    + """
<title>{blog_title} -> stats</title>
<meta name="description" content="stats of {blog_title}, {blog_description}" />
<meta property="og:type" content="website" />
</head>

<body>
<header role="group">
 <h1 role="heading" aria-level="1">stats of {blog_header}</h1>

 <nav id="info-bar" role="menubar">
   <a role="menuitem"
     aria-label="skip"
     href="#main">skip</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <span role="menuitem"
      >visitor <img src="{visitor_count}" alt="visitor count"
   /></span>
   <br role="seperator" aria-hidden="true" />

   <a role="menuitem" href="/">home</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{comment}">comment</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{website}">website</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="{source}">src</a>
   <span role="seperator" aria-hidden="true"> | </span>

   <a role="menuitem" href="/{rss}">rss</a>

   <hr aria-hidden="true" role="seperator" />
 </nav>
</header>
<main>
 <article id="main">

 <ul id=blist>
   <li>total count of blog posts : <code>{post_count}</code></li>
   <li>edited post count : <code>{edited_post_count}</code>, <code>{edited_post_count_p:.2f}%</code></li>
   <li>
   total read time : <time>{read_time}</time>
   <ul>
   <li>average read time : <time>{avg_read_time}</time></li>
   </ul>
   </li>

   <li>
      content

      <ul>
        <li>characters : <code>{char_count}</code></li>
        <ul>
        <li>average count of characters : <code>{avg_chars:.2f}</code></li>
        </ul>
      </ul>

      <ul>
        <li>words : <code>{word_count}</code></li>
        <ul>
        <li>average count of words : <code>{avg_words:.2f}</code></li>
        <li>average word length : <code>{avg_word_len:.2f}</code></li>
        <li>
          top {top_words} used words
          <ol>{word_most_used}</ol>
        </li>
        </ul>
      </ul>

      <ul>
        <li>tags : <code>{tag_count}</code></li>
        <ul>
        <li>average count of tags : <code>{avg_tags}</code></li>
        <li>
          top {top_tags} used tags
          <ol>{tags_most_used}</ol>
        </li>
        <li>default tags <ol>{default_tags}</ol></li>
        </ul>
      </ul>
   </li>

   <li>
    time ( GMT )
    <ul>
      <li>average posts by year : {posts_by_yr_avg} <ol>{posts_by_yr}</ol></li>
      <li>average posts by month : {posts_by_month_avg} <ol>{posts_by_month}</ol></li>
      <li>average posts by day : {posts_by_day_avg} <ol>{posts_by_day}</ol></li>
      <li>average posts by hour : {posts_by_hr_avg} <ol>{posts_by_hr}</ol></li>
    </ul>
   </li>
 </ul>
 </article>
</main> <footer><p>{author} &lt;<a href="mailto:{email}">{email}</a>&gt; + {license}</p></footer> </body>
</html>"""
)

if NCI:
    import http.server

    import pyfzf  # type: ignore
else:
    pyfzf: typing.Any = None
    http: typing.Any = None


class Commands:
    def __init__(self) -> None:
        self.commands: dict[str, typing.Callable[[dict[str, typing.Any]], int]] = {}

    def new(
        self, fn: typing.Callable[[dict[str, typing.Any]], int]
    ) -> typing.Callable[[dict[str, typing.Any]], int]:
        self.commands[fn.__name__] = fn
        return fn

    def __getitem__(self, name: str) -> typing.Callable[[dict[str, typing.Any]], int]:
        return self.commands[name]


cmds: Commands = Commands()
ecmds: Commands = Commands()


def ctimer() -> float:
    return code_timer() if NCI else 0


def log(msg: str, clr: str = LOG_CLR) -> int:
    if NCI:
        print(
            (
                f"{datetime.datetime.now()} | {msg}"
                if NOCLR
                else f"{clr}{datetime.datetime.now()} | {msg}\033[0m"
            ),
            file=sys.stderr,
        )

    return OK


def llog(msg: str) -> int:
    return log(msg, "\033[0m")


def err(msg: str) -> int:
    log(msg, ERR_CLR)
    return ER


def lnew(msg: str) -> int:
    return log(msg, NEW_CLR)


def imp(msg: str) -> int:
    return log(msg, IMP_CLR)


def slugify(
    title: str,
    context_words: typing.Sequence[str] | None = None,
    wslug_limit: int = DEFAULT_CONFIG["wslug-limit"],
    slug_limit: int = DEFAULT_CONFIG["slug-limit"],
) -> str:
    return (
        "-".join(
            [
                w
                for w in "".join(
                    c
                    for c in unidecode.unidecode(title).lower()
                    if c not in string.punctuation
                ).split()
                if w not in (context_words or [])
            ][:wslug_limit]
        )[:slug_limit].strip("-")
        or "post"
    )


def rf_format_time(ts: float) -> typing.Tuple[datetime.datetime, str]:
    d: datetime.datetime = datetime.datetime.utcfromtimestamp(ts)
    return d, d.strftime("%Y-%m-%d %H:%M:%S")


def rformat_time(ts: float) -> str:
    return datetime.datetime.utcfromtimestamp(ts).strftime("%Y-%m-%d %H:%M:%S")


def format_time(ts: float) -> str:
    return f"{rformat_time(ts)} GMT"


def select_multi(options: typing.Sequence[str]) -> list[str]:
    if not options:
        return []

    return pyfzf.FzfPrompt().prompt(
        choices=options,
        fzf_options="-m",
    )


def select_posts(posts: dict[str, dict[str, typing.Any]]) -> tuple[str, ...]:
    return tuple(
        map(
            lambda opt: opt.split("|", maxsplit=1)[0].strip(),
            select_multi(
                tuple(
                    f"{slug} | {post['title']} | {post['description']}"
                    for slug, post in posts.items()
                ),
            ),
        )
    )


if NCI:
    try:
        import readline
    except Exception:
        readline: typing.Any = None

    def iinput(prompt: str, default_text: str = "", force: bool = True) -> str:
        default_text = default_text.strip()

        if readline is not None and default_text:

            def hook() -> None:
                readline.insert_text(default_text)
                readline.redisplay()

            readline.set_pre_input_hook(hook)

        while not (user_input := input(f"\033[1m{prompt}\033[0m ").strip()) and force:
            pass

        if readline is not None:
            readline.set_pre_input_hook()

        return user_input

else:

    def iinput(prompt: str, default_text: str = "", force: bool = True) -> str:
        raise ValueError(
            f"cannot read user input in CI mode, prompt : {prompt!r}; default text : {default_text!r}; force : {force!r}"
        )


def yn(prompt: str, default: str = "y") -> bool:
    return (iinput(f"{prompt} ? [y/n]", default) + default)[0].lower() == "y"


def get_tmpfile(name: str) -> str:
    return f"{tempfile.gettempdir()}/{name}.md"


def open_file(editor: typing.Sequence[str], path: str) -> None:
    log(f"formatting and running {editor!r} with {path!r}")

    try:
        subprocess.run([(token.replace("%s", path)) for token in editor])
    except Exception as e:
        sys.exit(err(f"failed to run editor : {e}"))


def trunc(data: str, length: int, end: str = " ...") -> str:
    return data[:length] + (end if len(data) > length else "")


def read_post(path: str) -> str:
    log(f"reading {path!r}")

    try:
        with open(path, "r") as data:
            return data.read().strip()
    except Exception as e:
        err(f"failed to read {path!r} : {e.__class__.__name__} {e}")
        return ""


def min_css_file(file: str, out: str) -> None:
    with open(file, "r") as icss:
        with open(out, "w") as ocss:
            ocss.write(web_mini.css.minify_css(icss.read()))


def sorted_post_counter(
    c: Counter[int],
    pcount: int,
    fix: str,
) -> typing.Dict[str, typing.Any]:
    s: int = sum(c.values())
    avg: float = s / len(c)

    return {
        f"posts_by_{fix}": " ".join(
            f"<li><time>{v}</time> -- <code>{p}</code> post{'' if p == 1 else 's'}, <code>{p / pcount * 100:.2f}%</code></li>"
            for v, p in c.most_common()
        ),
        f"posts_by_{fix}_avg": f"<code>{round(avg, 2)}</code>, <code>{round(avg / s * 100, 2)}%</code>",
    }


def s_to_str(seconds: float) -> str:
    minutes, sec = divmod(seconds, 60)
    hours, minutes = divmod(minutes, 60)
    days, hours = divmod(hours, 24)

    periods: typing.Tuple[typing.Tuple[float, str, str], ...] = (
        (round(days, 2), "day", "days"),
        (round(hours, 2), "hour", "hours"),
        (round(minutes, 2), "minute", "minutes"),
        (round(sec, 2), "second", "seconds"),
    )

    time_periods: typing.List[str] = []

    for period in periods:
        if period[0] != 0:
            time_periods.append(
                "{} {}".format(period[0], period[1] if period[0] == 1 else period[2])
            )

    readable_text: str = ", ".join(time_periods[:-1])

    if len(time_periods) > 1:
        readable_text += " and " + time_periods[-1]
    else:
        readable_text = time_periods[0]

    return (
        f"{readable_text} ( {round(seconds, 2)} second{'' if seconds == 1 else 's'} )"
    )


# markdown

TITLE_LINKS_RE: typing.Final[str] = r"<#:[^>]+?>"


def parse_inline_titlelink(
    _: mistune.inline_parser.InlineParser,
    m: re.Match[str],
    state: mistune.core.InlineState,
) -> int:
    text: str = m.group(0)[3:-1]

    state.append_token(
        {
            "type": "link",
            "children": [{"type": "text", "raw": f"# {text}"}],
            "attrs": {"url": f"#{slugify(text, [], 768, 768)}"},
        }
    )

    return m.end()


def titlelink(md: mistune.Markdown) -> None:
    md.inline.register("titlelink", TITLE_LINKS_RE, parse_inline_titlelink, before="link")  # type: ignore


class BlogRenderer(mistune.HTMLRenderer):
    def heading(self, text: str, level: int, **_: typing.Any) -> str:
        slug: str = slugify(text, [], 768, 768)
        level = max(2, level)

        return f'<h{level} id="{slug}" h><a href="#{slug}">#</a> {text}</h{level}>'


def markdown(md: str, plugins: list[typing.Any]) -> str:
    return mistune.create_markdown(plugins=plugins + [titlelink], renderer=BlogRenderer())(md)  # type: ignore


# edit commands


@ecmds.new
def title(post: dict[str, typing.Any]) -> int:
    post["title"] = iinput("post title", post["title"])
    return OK


@ecmds.new
def description(post: dict[str, typing.Any]) -> int:
    post["description"] = iinput("post description", post["description"])
    return OK


@ecmds.new
def content(post: dict[str, typing.Any]) -> int:
    """edit posts"""

    log("getting post markdown path")
    path: str = get_tmpfile(post["slug"])

    log("writing content")
    with open(path, "w") as p:
        p.write(post["content"])

    open_file(post["editor"], path)

    if not (content := read_post(path)):
        return err("post content cannot be empty")

    post["content"] = content

    return OK


@ecmds.new
def keywords(post: dict[str, typing.Any]) -> int:
    """edit keywords"""

    post["keywords"] = tuple(
        map(
            lambda k: unidecode.unidecode(k.strip()),
            filter(
                bool,
                set(
                    iinput("post keywords", ", ".join(post["keywords"]), force=False)
                    .lower()
                    .split(",")
                ),
            ),
        )
    )
    return OK


# normal commands


@cmds.new
def help(_: dict[str, typing.Any]) -> int:
    """print help"""

    return llog(
        "\n\n"
        + "\n".join(
            f"{cmd} -- {fn.__doc__ or 'no help provided'}"
            for cmd, fn in cmds.commands.items()
        )
    )


@cmds.new
def sort(config: dict[str, typing.Any]) -> int:
    """sort blog posts by creation time"""

    log("sorting posts by creation time")

    config["posts"] = dict(
        map(
            lambda k: (k, config["posts"][k]),
            sorted(
                config["posts"],
                key=lambda k: config["posts"][k]["created"],
                reverse=True,
            ),
        )
    )

    return lnew("sorted blog posts by creation time")


@cmds.new
def new(config: dict[str, typing.Any]) -> int:
    """create a new blog post"""

    title: str = iinput("post title")

    log("creating a slug from the given title")
    slug: str = slugify(
        title,
        config["context-words"],
        config["wslug-limit"],
        config["slug-limit"],
    )

    if slug in (posts := config["posts"]):
        slug += f"-{sum(map(lambda k: k.startswith(slug), posts))}"

    log("getting post markdown path")
    post_path: str = get_tmpfile(slug)

    open_file(config["editor"], post_path)

    if not (content := read_post(post_path)):
        return err("content cannot be empty")

    keywords: tuple[str, ...] = tuple(
        map(
            lambda k: unidecode.unidecode(k.strip()),
            filter(
                bool,
                set(
                    iinput("post keywords ( separated by `,` )", force=False)
                    .lower()
                    .split(",")
                ),
            ),
        )
    )

    description: str = iinput("post description")

    lnew(f"saving blog post {slug!r}")

    posts[slug] = {
        "title": title,
        "description": description.strip(),
        "content": content,
        "keywords": keywords,
        "created": datetime.datetime.utcnow().timestamp(),
    }

    return OK


@cmds.new
def ls(config: dict[str, typing.Any]) -> int:
    """list all posts"""

    for slug, post in reversed(config["posts"].items()):
        llog(
            f"""post({slug})

title : {post["title"]!r}
description : {post["description"]!r}
content : {trunc(post["content"], config["post-preview-size"])!r}
keywords : {", ".join(post["keywords"])}
created : {format_time(post["created"])}"""
            + (
                ""
                if (ed := post.get("edited")) is None
                else f"\nedited : {format_time(ed)}"
            )
        )

    return OK


@cmds.new
def ed(config: dict[str, typing.Any], major: bool = True) -> int:
    """edit posts"""

    fields: list[str] = select_multi(tuple(ecmds.commands.keys()))

    for slug in select_posts(config["posts"]):
        llog(f"editing {slug!r}")

        for field in fields:
            log(f"editing field {field!r}")

            post: dict[str, typing.Any] = config["posts"][slug]

            post["slug"] = slug
            post["editor"] = config["editor"]

            code: int = ecmds[field](post)

            del post["slug"]
            del post["editor"]

            if code is not OK:
                return code

            if major:
                post["edited"] = datetime.datetime.utcnow().timestamp()

    return OK


@cmds.new
def med(config: dict[str, typing.Any]) -> int:
    """minor edit posts"""
    return ed(config=config, major=False)  # type: ignore


@cmds.new
def rm(config: dict[str, typing.Any]) -> int:
    """remove posts"""

    for slug in select_posts(config["posts"]):
        imp(f"deleting {slug!r}")
        del config["posts"][slug]

    return OK


@cmds.new
def build(config: dict[str, typing.Any]) -> int:
    """build blog posts"""

    if not config["posts"]:
        return err("no posts to be built")

    log("compiling regex")

    web_mini.html.html_fns.compileall()

    log("setting up posts directory")

    if os.path.exists(config["posts-dir"]):
        shutil.rmtree(config["posts-dir"])

    if os.path.exists("stats"):
        shutil.rmtree("stats")

    os.makedirs(config["posts-dir"], exist_ok=True)
    os.makedirs("stats", exist_ok=True)

    llog("building blog")

    t: Thread

    blog_title: str = html_escape(config["title"])
    author: str = html_escape(config["author"])
    styles: str = f"{config['assets-dir']}/styles.min.css"
    lang: str = config["locale"][:2]

    crit_css: str = ""
    post_crit_css: str = ""

    if os.path.isfile(critp := f"{config['assets-dir']}/critical.css"):
        with open(critp, "r") as fp:
            crit_css = web_mini.css.minify_css(fp.read())

    if os.path.isfile(critp := f"{config['assets-dir']}/post_critical.css"):
        with open(critp, "r") as fp:
            post_crit_css = web_mini.css.minify_css(fp.read())

    rt: typing.List[int] = []
    cc: typing.List[int] = []
    ws: Counter[str] = Counter()
    tgs: Counter[str] = Counter()

    py: Counter[int] = Counter()
    pm: Counter[int] = Counter()
    pd: Counter[int] = Counter()
    ph: Counter[int] = Counter()

    w_regex: re.Pattern[str] = re.compile(r"\b[a-zA-Z']+\b")
    url_regex: re.Pattern[str] = re.compile(r"https?://\S+|www\.\S+")

    def build_post(slug: str, post: dict[str, typing.Any]) -> None:
        ct: float = ctimer()

        post_dir: str = f"{config['posts-dir']}/{slug}"
        os.makedirs(post_dir)

        rtm: MarkdownResult = read_time_of_markdown(post["content"], config["read-wpm"])
        cont: str = url_regex.sub("", post["content"]) + " " + post["title"]

        rt.append(rtm.seconds)
        cc.append(len(cont))
        ws.update(Counter(w_regex.findall(cont.lower().strip())))
        tgs.update(Counter(list(map(str.lower, post["keywords"]))))

        dt, s = rf_format_time(post["created"])

        py[dt.year] += 1
        pm[dt.month] += 1
        pd[dt.day] += 1
        ph[dt.hour] += 1

        with open(f"{post_dir}/index.html", "w") as html:
            html.write(
                web_mini.html.minify_html(
                    POST_TEMPLATE.format(
                        lang=lang,
                        keywords=html_escape(
                            ", ".join(
                                set(post["keywords"] + config["default-keywords"])
                            )
                        ),
                        theme_type=config["theme"]["type"],
                        theme_primary=config["theme"]["primary"],
                        theme_secondary=config["theme"]["secondary"],
                        styles=styles,
                        critical_css=crit_css,
                        post_critical_css=post_crit_css,
                        gen=GEN,
                        rss=config["rss-file"],
                        blog_title=blog_title,
                        post_title=html_escape(post["title"]),
                        author=author,
                        locale=config["locale"],
                        post_creation_time=s,
                        post_description=html_escape(post["description"]),
                        post_read_time=rtm.text,
                        post_edit_time=(
                            ""
                            if "edited" not in post
                            else f', edited on <time>{rformat_time(post["edited"])}</time> GMT'
                        ),
                        visitor_count=config["visitor-count"],
                        comment=config["comment"],
                        website=config["website"],
                        source=config["source"],
                        post_content=markdown(
                            post["content"], config["markdown-plugins"]
                        ),
                        blog=config["blog"],
                        path=f"{config['posts-dir']}/{slug}/",
                        license=config["license"],
                        email=config["email"],
                        legal=config["legal"],
                    ),
                )
            )

        lnew(f"built post {post['title']!r} in {ctimer() - ct} s")

    ts: list[Thread] = []

    for slug, post in tuple(config["posts"].items()):
        ts.append(t := Thread(target=build_post, args=(slug, post), daemon=True))
        t.start()

    latest_post: tuple[str, dict[str, typing.Any]] = tuple(config["posts"].items())[0]

    with open("index.html", "w") as index:
        index.write(
            web_mini.html.minify_html(
                INDEX_TEMPLATE.format(  # type: ignore
                    lang=lang,
                    keywords=(bkw := html_escape(", ".join(config["blog-keywords"]))),
                    theme_type=config["theme"]["type"],
                    theme_primary=config["theme"]["primary"],
                    theme_secondary=config["theme"]["secondary"],
                    blog=config["blog"],
                    path="",
                    styles=styles,
                    critical_css=crit_css,
                    gen=GEN,
                    rss=config["rss-file"],
                    blog_title=blog_title,
                    author=author,
                    locale=config["locale"],
                    license=config["license"],
                    blog_description=(bd := html_escape(config["description"])),
                    blog_header=(bh := html_escape(config["header"])),
                    latest_post_path=f"{config['posts-dir']}/{latest_post[0]}",
                    latest_post_title_trunc=html_escape(
                        trunc(latest_post[1]["title"], config["recent-title-trunc"])
                    ),
                    latest_post_creation_time=rformat_time(latest_post[1]["created"]),
                    visitor_count=config["visitor-count"],
                    comment=config["comment"],
                    website=config["website"],
                    source=config["source"],
                    blog_list=" ".join(
                        f'<li><a href="/{config["posts-dir"]}/{slug}">{html_escape(post["title"])}</a></li>'
                        for slug, post in config["posts"].items()
                    ),
                    email=config["email"],
                ),
            )
        )

        lnew(f"generated {index.name!r}")

    for t in ts:
        t.join()

    char_count: int = sum(cc)
    post_count: int = len(config["posts"])
    epost_count: int = sum("edited" in p for p in config["posts"].values())

    rts: int = sum(rt)

    wcs: int = sum(ws.values())
    wcl: int = len(ws)

    tcs: int = sum(tgs.values())
    tcl: int = len(tgs)

    avg_chars: float = char_count / post_count
    avg_words: float = wcs / post_count
    avg_tags: float = tcs / post_count

    with open("stats/index.html", "w") as stats:
        stats.write(
            web_mini.html.minify_html(
                STATS_TEMPLATE.format(
                    lang=lang,
                    keywords=bkw + ", stats, statistics",
                    theme_type=config["theme"]["type"],
                    theme_primary=config["theme"]["primary"],
                    theme_secondary=config["theme"]["secondary"],
                    blog=config["blog"],
                    path="stats/",
                    styles=styles,
                    critical_css=crit_css,
                    gen=GEN,
                    locale=config["locale"],
                    blog_title=blog_title,
                    blog_description=bd,
                    blog_header=bh,
                    visitor_count=config["visitor-count"],
                    comment=config["comment"],
                    website=config["website"],
                    source=config["source"],
                    rss=config["rss-file"],
                    post_count=post_count,
                    edited_post_count=epost_count,
                    edited_post_count_p=epost_count / post_count * 100,
                    read_time=s_to_str(rts),
                    avg_read_time=s_to_str(rts / post_count),
                    char_count=char_count,
                    avg_chars=avg_chars,
                    word_count=wcs,
                    avg_words=avg_words,
                    avg_word_len=avg_chars / avg_words,
                    top_words=config["top-words"],
                    word_most_used=" ".join(
                        f"<li><code>{html_escape(w)}</code>, <code>{u}</code> use{'' if u == 1 else 's'}, <code>{u / wcl * 100:.2f}%</code></li>"
                        for w, u in ws.most_common(config["top-words"])
                    ),
                    tag_count=tcs,
                    avg_tags=avg_tags,
                    top_tags=config["top-tags"],
                    tags_most_used=" ".join(
                        f"<li><code>{html_escape(w)}</code>, <code>{u}</code> use{'' if u == 1 else 's'}, <code>{u / tcl * 100:.2f}%</code></li>"
                        for w, u in tgs.most_common(config["top-tags"])
                    ),
                    default_tags=" ".join(
                        f"<li><code>{html_escape(t)}</code></li>"
                        for t in config["default-keywords"]
                    ),
                    **sorted_post_counter(py, post_count, "yr"),
                    **sorted_post_counter(pm, post_count, "month"),
                    **sorted_post_counter(pd, post_count, "day"),
                    **sorted_post_counter(ph, post_count, "hr"),
                    author=config["author"],
                    email=config["email"],
                    license=config["license"],
                )
            )
        )

        lnew(f"generated {stats.name!r}")

    return 0


@cmds.new
def css(config: dict[str, typing.Any]) -> int:
    """build and minify css"""

    t: Thread
    ts: list[Thread] = []

    log("compiling regex")
    web_mini.css.css_fns.compileall()

    def _thread(c: typing.Callable[..., typing.Any], *args: str) -> None:
        def _c() -> None:
            ct: float = ctimer()
            c(*args)
            lnew(f"processed {args[0]!r} in {ctimer() - ct} s")

        ts.append(t := Thread(target=_c, daemon=True))
        t.start()

    if os.path.isfile(styles := f"{config['assets-dir']}/styles.css"):
        lnew(f"minifying {styles!r}")
        _thread(min_css_file, styles, f"{config['assets-dir']}/styles.min.css")  # type: ignore

    if os.path.isdir(fonts := f"{config['assets-dir']}/fonts"):
        log(f"minifying fonts in {fonts!r}")

        for fcss in iglob(f"{fonts}/*.css"):
            if fcss.endswith(".min.css"):
                continue

            _thread(min_css_file, fcss, f"{os.path.splitext(fcss)[0]}.min.css")  # type: ignore

    for t in ts:
        t.join()

    return OK


@cmds.new
def robots(config: dict[str, typing.Any]) -> int:
    """generate a robots.txt"""

    llog("generating robots")

    with open("robots.txt", "w") as r:
        r.write(
            f"""User-agent: *
Disallow: /{config["assets-dir"]}/*
Allow: *
Sitemap: {config["blog"]}/sitemap.xml"""
        )

        lnew(f"generated {r.name!r}")

    return OK


@cmds.new
def manifest(config: dict[str, typing.Any]) -> int:
    """generate a manifest.json"""

    llog("generating a manifest")

    with open("manifest.json", "w") as m:
        json.dump(
            {
                "$schema": "https://json.schemastore.org/web-manifest-combined.json",
                "short_name": config["header"],
                "name": config["title"],
                "description": config["description"],
                "start_url": ".",
                "display": "standalone",
                "theme_color": config["theme"]["primary"],
                "background_color": config["theme"]["secondary"],
                **config["manifest"],
            },
            m,
        )

        lnew(f"generated {m.name!r}")

    return OK


@cmds.new
def sitemap(config: dict[str, typing.Any]) -> int:
    """generate a sitemap.xml"""

    llog("generating a sitemap")

    now: float = datetime.datetime.utcnow().timestamp()

    root: etree.Element = etree.Element("urlset")
    root.set("xmlns", "http://www.sitemaps.org/schemas/sitemap/0.9")

    for slug, post in (
        ("", config["website"]),
        ("", config["blog"]),
        ("", f"{config['blog']}/{config['rss-file']}"),
        ("", f'{config["blog"]}/stats'),
    ) + tuple(config["posts"].items()):
        llog(f"adding {slug or post!r} to sitemap")

        url: etree.Element = etree.SubElement(root, "url")

        etree.SubElement(url, "loc").text = (
            f"{config['blog']}/{config['posts-dir']}/{slug}" if slug else post
        )
        etree.SubElement(url, "lastmod").text = datetime.datetime.utcfromtimestamp(
            post.get("edited", post["created"]) if slug else now,  # type: ignore
        ).strftime("%Y-%m-%dT%H:%M:%S+00:00")
        etree.SubElement(url, "priority").text = "1.0"

    etree.ElementTree(root).write("sitemap.xml", encoding="UTF-8", xml_declaration=True)
    lnew("generated 'sitemap.xml'")

    return OK


@cmds.new
def rss(config: dict[str, typing.Any]) -> int:
    """generate an rss feed"""

    llog("generating an rss feed")

    ftime: str = "%a, %d %b %Y %H:%M:%S GMT"
    now: datetime.datetime = datetime.datetime.utcnow()

    root: etree.Element = etree.Element("rss")
    root.set("version", "2.0")

    channel: etree.Element = etree.SubElement(root, "channel")

    etree.SubElement(channel, "title").text = config["title"]
    etree.SubElement(channel, "link").text = config["blog"]
    etree.SubElement(channel, "description").text = config["description"]
    etree.SubElement(channel, "generator").text = GEN
    etree.SubElement(channel, "language").text = (
        config["locale"].lower().replace("_", "-")
    )
    etree.SubElement(channel, "lastBuildDate").text = now.strftime(ftime)

    for slug, post in config["posts"].items():
        llog(f"adding {slug!r} to rss")

        created: float | None = post.get("edited")

        item: etree.Element = etree.SubElement(channel, "item")

        etree.SubElement(item, "title").text = post["title"]
        etree.SubElement(item, "link").text = (
            link := f"{config['blog']}/{config['posts-dir']}/{slug}"
        )
        etree.SubElement(item, "description").text = post["description"] + (
            f" [edited at {datetime.datetime.utcfromtimestamp(created).strftime(ftime)}]"
            if created
            else ""
        )
        etree.SubElement(item, "pubDate").text = datetime.datetime.utcfromtimestamp(
            post["created"],
        ).strftime(ftime)
        etree.SubElement(item, "guid").text = link
        etree.SubElement(item, "author").text = (
            f"{config['email']} ({config['author']})"
        )

    etree.ElementTree(root).write(
        config["rss-file"], encoding="UTF-8", xml_declaration=True
    )

    lnew(f"generated {config['rss-file']!r}")

    return OK


@cmds.new
def apis(config: dict[str, typing.Any]) -> int:
    """generate and hash apis"""

    with open("recents.json", "w") as recents:
        json.dump(
            dict(
                map(
                    lambda kv: (  # type: ignore
                        kv[0],
                        {
                            "title": kv[1]["title"],
                            "content": trunc(
                                kv[1]["content"], config["post-preview-size"], ""
                            ),
                            "created": kv[1]["created"],
                        },
                    ),
                    tuple(config["posts"].items())[: config["recents"]],
                )
            ),
            recents,
        )
        lnew(f"generated {recents.name!r}")

    for api in recents.name, CONFIG_FILE:
        with open(api, "rb") as content:
            h: str = hashlib.sha256(content.read()).hexdigest()

        with open(f"{api.replace('.', '_')}_hash.txt", "w") as hf:
            hf.write(h)
            lnew(f"generated {hf.name!r}")

    return OK


@cmds.new
def clean(config: dict[str, typing.Any]) -> int:
    """clean up the site"""

    def remove(file: str) -> None:
        imp(f"removing {file!r}")

        try:
            os.remove(file)
        except IsADirectoryError:
            shutil.rmtree(file)

    for pattern in (
        config["posts-dir"],
        "index.html",
        f"{config['assets-dir']}/*.min.*",
        "blog_json_hash.txt",
        "manifest.json",
        f"{config['assets-dir']}/fonts/*.min.*",
        "recents_json_hash.txt",
        "recents.json",
        config["rss-file"],
        "robots.txt",
        "sitemap.xml",
        "stats",
    ):
        if os.path.exists(pattern):
            remove(pattern)
        else:
            for file in iglob(pattern, recursive=True):
                remove(file)

    return OK


@cmds.new
def static(config: dict[str, typing.Any]) -> int:
    """generate a full static site"""

    ct: float = ctimer()

    for stage in clean, build, css, robots, manifest, sitemap, rss, apis:
        imp(f"running stage {stage.__name__!r} : {stage.__doc__ or stage.__name__!r}")

        st: float = ctimer()

        if (code := stage(config)) is not OK:
            return code

        imp(f"stage finished in {ctimer() - st} s")

    return log(f"site built in {ctimer() - ct} s")


@cmds.new
def serve(config: dict[str, typing.Any]) -> int:
    """simple server"""

    class RequestHandler(http.server.SimpleHTTPRequestHandler):
        def log_message(self, format: str, *args: typing.Any) -> None:
            llog(format % args)

        def do_GET(self) -> None:
            file_path: str = self.translate_path(self.path)  # type: ignore

            if os.path.isdir(file_path):  # type: ignore
                file_path = f"{file_path}/index.html"

            try:
                with open(file_path, "rb") as fp:  # type: ignore
                    self.send_response(200)  # type: ignore
                    self.send_header(
                        "Cache-Control", "no-store, no-cache, must-revalidate"
                    )
                    self.send_header("Pragma", "no-cache")
                    self.end_headers()  # type: ignore
                    self.wfile.write(fp.read())  # type: ignore
            except Exception as e:
                self.send_response(404)  # type: ignore
                self.send_header("Cache-Control", "no-store, no-cache, must-revalidate")
                self.send_header("Pragma", "no-cache")
                self.end_headers()  # type: ignore
                self.wfile.write(f"{e.__class__.__name__} : {e}".encode())  # type: ignore

    httpd: typing.Any = http.server.HTTPServer(
        (config["server-host"], config["server-port"]), RequestHandler
    )
    httpd.RequestHandlerClass.directory = "."

    try:
        imp(
            f"server running on http://{httpd.server_address[0]}:{httpd.server_address[1]}/ ^C to close it"
        )
        httpd.serve_forever()
    except KeyboardInterrupt:
        httpd.server_close()
        imp("server shut down")

    return OK


@cmds.new
def dev(config: dict[str, typing.Any]) -> int:
    """generate a full static site + serve it"""

    if (code := static(config)) is not OK:
        return code

    return serve(config)


@cmds.new
def blog(config: dict[str, typing.Any]) -> int:
    """generate a new blog"""

    log("changing config")
    config.update(DEFAULT_CONFIG)
    lnew("blog set to default values")

    return OK


def main() -> int:
    """entry / main function"""

    main_t: float = ctimer()

    log("hello world")

    if len(sys.argv) < 2:
        return err("no arguments provided, see `help`")

    cfg: dict[str, typing.Any] = DEFAULT_CONFIG.copy()

    if os.path.exists(CONFIG_FILE):
        with open(CONFIG_FILE, "r") as config:
            log(f"using {config.name!r} config")
            cfg.update(json.load(config))
    else:
        lnew("using the default config")

    sort(cfg)

    log(f"looking command {sys.argv[1]!r} up")

    try:
        cmd: typing.Callable[[dict[str, typing.Any]], int] = cmds[sys.argv[1]]
    except KeyError:
        return err(f"command {sys.argv[1]!r} does not exist")

    for file in ".editorconfig", ".clang-format":
        if os.path.isfile(file):
            log(f"copying {file!r} to /tmp")
            shutil.copy(file, f"/tmp/{file}")

    log("calling and timing the command")
    if NCI:
        print()
        timer: float = ctimer()

    code: int = cmd(cfg)

    if NCI:
        print()
        log(f"command finished in {ctimer() - timer} s")  # type: ignore
        sort(cfg)

    with open(CONFIG_FILE, "w") as config:
        log(f"dumping config to {config.name!r}")
        json.dump(cfg, config, indent=cfg["indent"] if NCI else None)

    log(f"goodbye world, return {code}, total {ctimer() - main_t} s")

    return code


if __name__ == "__main__":
    assert (
        main.__annotations__.get("return") == "int"
    ), "main() should return an integer"

    # filter_warnings("error", category=Warning)
    raise SystemExit(main())