from __future__ import annotations

import re

_ALLOWED_TAGS = {"b", "i", "u", "s", "a", "br", "code", "pre", "blockquote", "img"}

_UNSUPPORTED_REPLACEMENTS: dict[str, tuple[str, str]] = {
    "strong": ("<b>", "</b>"),
    "em": ("<i>", "</i>"),
    "strike": ("<s>", "</s>"),
    "del": ("<s>", "</s>"),
    "ins": ("<u>", "</u>"),
    "h1": ("<b>", "</b>\n"),
    "h2": ("<b>", "</b>\n"),
    "h3": ("<b>", "</b>\n"),
    "h4": ("<b>", "</b>\n"),
    "h5": ("<b>", "</b>\n"),
    "h6": ("<b>", "</b>\n"),
    "p": ("\n", ""),
    "ul": ("", ""),
    "ol": ("", ""),
    "li": ("• ", "\n"),
    "span": ("", ""),
    "div": ("\n", ""),
    "hr": ("\n---\n", ""),
}

_STRIP_TAGS = {"script", "style", "head", "meta", "link", "iframe", "object", "embed"}


def render_html_to_imessage(html: str) -> str:
    if not html:
        return ""

    for tag in _STRIP_TAGS:
        html = re.sub(rf"<{tag}[^>]*>.*?</{tag}>", "", html, flags=re.DOTALL | re.IGNORECASE)
        html = re.sub(rf"<{tag}[^>]*/>", "", html, flags=re.IGNORECASE)

    for tag, (open_tag, close_tag) in _UNSUPPORTED_REPLACEMENTS.items():
        html = re.sub(rf"<{tag}(\s[^>]*)?>", open_tag, html, flags=re.IGNORECASE)
        html = re.sub(rf"</{tag}>", close_tag, html, flags=re.IGNORECASE)

    html = _strip_unsafe_attributes(html)

    html = _normalize_links(html)

    html = html.replace("\r\n", "\n").replace("\r", "\n")

    return html.strip()


def _strip_unsafe_attributes(html: str) -> str:
    def _clean_tag(tag_match: re.Match) -> str:
        full_tag = tag_match.group(0)
        tag_name_match = re.match(r"</?(\w+)", full_tag)
        if not tag_name_match:
            return full_tag
        tag_name = tag_name_match.group(1).lower()
        if tag_name not in _ALLOWED_TAGS:
            return full_tag
        safe_attrs = ["href", "src", "alt"]
        result = full_tag
        for attr in re.finditer(r'(\w+)="([^"]*)"', full_tag):
            if attr.group(1).lower() not in safe_attrs:
                result = result.replace(attr.group(0), "")
        return result

    return re.sub(r"<[^>]+>", _clean_tag, html)


def _normalize_links(text: str) -> str:
    url_pattern = re.compile(r"(?<!href=\")(https?://[^\s<>\"\)]+)")
    return url_pattern.sub(r'<a href="\1">\1</a>', text)


def render_markdown_to_imessage(text: str) -> str:
    if not text:
        return ""

    text = text.replace("**", "<b>", 1)
    while "**" in text:
        text = text.replace("**", "</b>", 1)
        if "**" in text:
            text = text.replace("**", "<b>", 1)

    text = text.replace("__", "<b>", 1)
    while "__" in text:
        text = text.replace("__", "</b>", 1)
        if "__" in text:
            text = text.replace("__", "<b>", 1)

    text = text.replace("*", "<i>", 1)
    while "*" in text:
        text = text.replace("*", "</i>", 1)

    text = text.replace("_", "<i>", 1)
    while "_" in text:
        text = text.replace("_", "</i>", 1)

    return text