devplacepy/devplacepy/rendering.py

# retoor <retoor@molodetz.nl>

import html
import json
import re
from functools import lru_cache
from html.parser import HTMLParser
from pathlib import Path
from urllib.parse import unquote, urlparse

import emoji
import mistune
from markupsafe import Markup

EMOJI_JS_PATH = Path(__file__).parent / "static" / "js" / "emoji-shortcodes.js"

CUSTOM_SHORTCODES = {
    "party": "\U0001F973",
    "tools": "\U0001F6E0️",
}


def build_emoji_shortcodes() -> dict[str, str]:
    shortcodes: dict[str, str] = {}
    for char, data in emoji.EMOJI_DATA.items():
        if data.get("status", 0) > emoji.STATUS["fully_qualified"]:
            continue
        names = [data["en"]] if data.get("en") else []
        names += data.get("alias", [])
        for name in names:
            shortcodes.setdefault(name.strip(":").lower(), char)
    shortcodes.update(CUSTOM_SHORTCODES)
    return shortcodes


def write_emoji_module(path: Path = EMOJI_JS_PATH) -> int:
    shortcodes = build_emoji_shortcodes()
    body = json.dumps(shortcodes, ensure_ascii=False, sort_keys=True)
    path.write_text(
        f"// retoor <retoor@molodetz.nl>\n\nexport const EMOJI_SHORTCODES = {body};\n",
        encoding="utf-8",
    )
    return len(shortcodes)


EMOJI_MAP = build_emoji_shortcodes()

_SHORTCODE_RE = re.compile(r":([A-Za-z0-9_+\-]+):")
_YOUTUBE_RE = re.compile(
    r"(?:https?://)?(?:www\.)?"
    r"(?:youtube\.com/(?:watch\?v=|embed/|v/|shorts/)|youtu\.be/)"
    r"([A-Za-z0-9_-]{11})"
)
_IMAGE_RE = re.compile(r"\.(jpg|jpeg|png|gif|webp|svg|bmp)(\?.*)?$", re.I)
_VIDEO_RE = re.compile(r"\.(mp4|webm|ogg|ogv|mov|m4v)(\?.*)?$", re.I)
_AUDIO_RE = re.compile(r"\.(mp3|wav|flac|ogg|aac|m4a|wma|opus)(\?.*)?$", re.I)
_TOKEN_RE = re.compile(
    r"(?P<url>https?://[^\s<]+)"
    r"|(?<![A-Za-z0-9_@])@(?P<mention>[A-Za-z0-9_-]+)"
)
_SCHEME_RE = re.compile(r"^([a-z][a-z0-9+.\-]*):", re.I)
_ALLOWED_SCHEMES = {"http", "https", "mailto", "tel"}

_YOUTUBE_ALLOW = (
    "accelerometer; autoplay; clipboard-write; encrypted-media; "
    "gyroscope; picture-in-picture"
)

_MEDIA_SKIP_TAGS = {"a", "code", "pre"}
_TITLE_INLINE_TAGS = {
    "b", "strong", "i", "em", "code", "del", "s", "mark", "sub", "sup", "span", "br",
}


def _safe_url(url: str) -> str:
    value = re.sub(r"[\x00-\x20]", "", url or "")
    match = _SCHEME_RE.match(value)
    if match and match.group(1).lower() not in _ALLOWED_SCHEMES:
        return "#"
    return value


class _ContentRenderer(mistune.HTMLRenderer):
    def link(self, text: str, url: str, title=None) -> str:
        return super().link(text, _safe_url(url), title)

    def image(self, text: str, url: str, title=None) -> str:
        src = html.escape(_safe_url(url), quote=True)
        alt = html.escape(text or "", quote=True)
        title_attr = f' title="{html.escape(title, quote=True)}"' if title else ""
        return (
            f'<img src="{src}" alt="{alt}"{title_attr} '
            f'loading="lazy" data-lightbox>'
        )


_content_markdown = mistune.create_markdown(
    hard_wrap=True,
    renderer=_ContentRenderer(escape=True),
    plugins=["strikethrough", "table"],
)


def _normalize_dashes(text: str) -> str:
    return text.replace("\u2014", "-")


def _replace_shortcodes(text: str) -> str:
    return _SHORTCODE_RE.sub(
        lambda m: EMOJI_MAP.get(m.group(1).lower(), m.group(0)), text
    )


def _alt_from_url(url: str) -> str:
    name = Path(unquote(urlparse(url).path)).stem.replace("-", " ").replace("_", " ").strip()
    return name if name else "Embedded image"


def _embed_url(url: str) -> str:
    youtube = _YOUTUBE_RE.search(url)
    if youtube:
        video_id = youtube.group(1)
        return (
            f'<div class="embed-youtube"><iframe '
            f'src="https://www.youtube.com/embed/{video_id}" '
            f'frameborder="0" allowfullscreen allow="{_YOUTUBE_ALLOW}">'
            f"</iframe></div>"
        )
    escaped = html.escape(url, quote=True)
    if _IMAGE_RE.search(url):
        alt = html.escape(_alt_from_url(url), quote=True)
        return f'<img src="{escaped}" alt="{alt}" loading="lazy" data-lightbox>'
    if _VIDEO_RE.search(url):
        return f'<video src="{escaped}" controls preload="metadata"></video>'
    if _AUDIO_RE.search(url):
        return f'<audio src="{escaped}" controls preload="metadata"></audio>'
    return (
        f'<a href="{escaped}" target="_blank" rel="noopener noreferrer">'
        f"{html.escape(url)}</a>"
    )


def _transform_text(text: str) -> str:
    out: list[str] = []
    pos = 0
    for match in _TOKEN_RE.finditer(text):
        if match.start() > pos:
            out.append(html.escape(text[pos:match.start()]))
        if match.group("url"):
            out.append(_embed_url(match.group("url")))
        else:
            user = match.group("mention")
            out.append(
                f'<a href="/profile/{html.escape(user, quote=True)}" '
                f'class="mention-link">@{html.escape(user)}</a>'
            )
        pos = match.end()
    if pos < len(text):
        out.append(html.escape(text[pos:]))
    return "".join(out)


def _emit_tag(tag: str, attrs: list, self_closing: bool = False) -> str:
    parts = [f"<{tag}"]
    for key, value in attrs:
        if value is None:
            parts.append(f" {key}")
        else:
            parts.append(f' {key}="{html.escape(value, quote=True)}"')
    parts.append(" />" if self_closing else ">")
    return "".join(parts)


class _MediaProcessor(HTMLParser):
    def __init__(self) -> None:
        super().__init__(convert_charrefs=True)
        self._out: list[str] = []
        self._skip_depth = 0

    def handle_starttag(self, tag: str, attrs: list) -> None:
        self._out.append(_emit_tag(tag, attrs))
        if tag in _MEDIA_SKIP_TAGS:
            self._skip_depth += 1

    def handle_startendtag(self, tag: str, attrs: list) -> None:
        self._out.append(_emit_tag(tag, attrs, self_closing=True))

    def handle_endtag(self, tag: str) -> None:
        if tag in _MEDIA_SKIP_TAGS and self._skip_depth > 0:
            self._skip_depth -= 1
        self._out.append(f"</{tag}>")

    def handle_data(self, data: str) -> None:
        if self._skip_depth > 0:
            self._out.append(html.escape(data))
        else:
            self._out.append(_transform_text(data))

    def result(self) -> str:
        return "".join(self._out)


class _InlineFilter(HTMLParser):
    def __init__(self) -> None:
        super().__init__(convert_charrefs=True)
        self._out: list[str] = []

    def handle_starttag(self, tag: str, attrs: list) -> None:
        if tag in _TITLE_INLINE_TAGS:
            self._out.append(f"<{tag}>")

    def handle_startendtag(self, tag: str, attrs: list) -> None:
        if tag in _TITLE_INLINE_TAGS:
            self._out.append(f"<{tag}>")

    def handle_endtag(self, tag: str) -> None:
        if tag in _TITLE_INLINE_TAGS and tag != "br":
            self._out.append(f"</{tag}>")

    def handle_data(self, data: str) -> None:
        self._out.append(html.escape(data))

    def result(self) -> str:
        return "".join(self._out).strip()


def _process_media(rendered: str) -> str:
    processor = _MediaProcessor()
    processor.feed(rendered)
    processor.close()
    return processor.result()


def _keep_inline(rendered: str) -> str:
    inline = _InlineFilter()
    inline.feed(rendered)
    inline.close()
    return inline.result()


@lru_cache(maxsize=4096)
def _render_content(text: str) -> str:
    text = _normalize_dashes(text)
    text = _replace_shortcodes(text)
    return _process_media(_content_markdown(text))


@lru_cache(maxsize=4096)
def _render_title(text: str) -> str:
    text = _normalize_dashes(text)
    text = _replace_shortcodes(text)
    return _keep_inline(_content_markdown(text))


def render_content(text) -> Markup:
    if not text:
        return Markup("")
    return Markup(_render_content(str(text)))


def render_title(text) -> Markup:
    if not text:
        return Markup("")
    return Markup(_render_title(str(text)))


def content_preview(text, length: int = 60) -> str:
    if not text:
        return ""
    from devplacepy.utils import strip_html

    plain = strip_html(_render_content(str(text)))
    if len(plain) <= length:
        return plain
    return plain[:length].rstrip() + "..."