from __future__ import annotations

from datetime import datetime
from time import mktime

import feedparser
from bs4 import BeautifulSoup

from .base import ExtractedJob, ExtractionResult, FieldEvidence


class RssAdapter:
    parser_key = "rss-atom"
    parser_version = "1.0.0"

    def extract(self, content: str, *, url: str) -> ExtractionResult:
        feed = feedparser.parse(content)
        jobs: list[ExtractedJob] = []
        for entry in feed.entries:
            title = str(entry.get("title") or "").strip()
            link = str(entry.get("link") or "").strip()
            if not title or not link:
                continue
            summary = str(entry.get("summary") or entry.get("description") or "")
            text = BeautifulSoup(summary, "lxml").get_text("\n", strip=True)
            published = ""
            if entry.get("published_parsed"):
                published = datetime.fromtimestamp(mktime(entry.published_parsed)).isoformat()
            jobs.append(
                ExtractedJob(
                    url=link,
                    title=title,
                    external_id=str(entry.get("id") or ""),
                    description_html=summary,
                    description_text=text,
                    date_posted=published,
                    raw=dict(entry),
                    evidence=[
                        FieldEvidence("title", "rss", 0.88, title[:240]),
                        FieldEvidence("description", "rss", 0.80, text[:300]),
                    ],
                )
            )
        warnings: list[str] = []
        if getattr(feed, "bozo", False):
            warnings.append(str(getattr(feed, "bozo_exception", "Ongeldige feed")))
        return ExtractionResult(
            jobs,
            self.parser_key,
            self.parser_version,
            0.82 if jobs else 0.0,
            warnings,
        )