import re

import bs4
import pydantic

from ows_text_campaigns.campaigns.exceptions import InvalidShortLinkTagError
from ows_text_campaigns.campaigns.types import (
    ShortenedUrl,
)

URL_REGEX = re.compile(r"""(?i)\b((?:https?://|www\.)[^\s<>'"()]+)""")


def extract_shortened_urls_from_markup(
    markup: str, *, raise_on_error: bool = False
) -> list[ShortenedUrl]:
    result: list[ShortenedUrl] = []
    soup = bs4.BeautifulSoup(markup, "html.parser")
    for tag in soup.select("a[data-short-link-id]"):
        try:
            result.append(
                ShortenedUrl.model_validate(
                    {
                        "id": tag.get("data-short-link-id"),
                        "url": tag.get("data-short-link-url"),
                        "method": tag.get("data-short-link-method"),
                        "path": tag.get("data-short-link-path"),
                        "domain": tag.get("data-short-link-domain"),
                    },
                )
            )
        except pydantic.ValidationError as exc:
            if raise_on_error:
                attr_names = ", ".join(
                    ["".join(map(str, error["loc"])) for error in exc.errors()]
                )
                raise InvalidShortLinkTagError(
                    f"Missing or invalid `data-short-link-*` attribute(s) in short link tag: {attr_names}"
                ) from exc
            continue
    return result


def extract_non_shortened_urls_from_content(content: str) -> list[str]:
    soup = bs4.BeautifulSoup(content, "html.parser")
    for tag in soup.select("a[data-short-link-id]"):
        tag.decompose()
    return list(set(URL_REGEX.findall(str(soup))))


def create_short_link_tag(shortened_url: ShortenedUrl) -> str:
    return (
        f'<a data-short-link-id="{shortened_url.id}" '
        f'data-short-link-url="{shortened_url.url}" '
        f'data-short-link-method="{shortened_url.method.value}" '
        f'data-short-link-path="{shortened_url.path}" '
        f'data-short-link-domain="{shortened_url.domain}">'
        f"{shortened_url.preview_url}</a>"
    )


def replace_short_link_tags(markup: str, *, shortened_urls: dict[str, str]) -> str:
    soup = bs4.BeautifulSoup(markup, "html.parser")

    for tag in soup.select("a[data-short-link-id]"):
        shortened_url_id = tag.get("data-short-link-id")
        if not isinstance(shortened_url_id, str):
            continue

        shortened_url = shortened_urls.get(shortened_url_id)
        if not shortened_url:
            continue

        # Create a new <a> tag
        new_tag = soup.new_tag("a", href=shortened_url)
        new_tag.string = shortened_url

        tag.replace_with(new_tag)

    return str(soup)


def replace_standard_links_tags(markup: str) -> str:
    soup = bs4.BeautifulSoup(markup, "html.parser")

    for tag in soup.select('a[data-short-link-method="STANDARD"]'):
        shortened_url = tag.get("data-short-link-url")
        if not isinstance(shortened_url, str):
            continue

        # Create a new <a> tag
        new_tag = soup.new_tag("a", href=shortened_url)
        new_tag.string = shortened_url

        tag.replace_with(new_tag)

    return str(soup)