Dovodi_Backend/apps/hadis/utils/links.py

import re
from urllib.parse import urlparse


def extract_domain_from_url(url: str) -> str:
    """
    Extracts the root domain / hostname without 'www.' from a URL.
    Examples:
        'https://www.digikala.com/product/123' -> 'digikala.com'
        'http://dorar.net/hadith/3420' -> 'dorar.net'
        'www.al-islam.org/path' -> 'al-islam.org'
        'digikala.com/item' -> 'digikala.com'
    """
    if not url or not isinstance(url, str):
        return ""
    trimmed = url.strip()
    if not trimmed:
        return ""

    candidate = trimmed
    if not re.match(r"^[a-zA-Z][a-zA-Z0-9+\-.]*://", candidate):
        candidate = "https://" + candidate

    try:
        parsed = urlparse(candidate)
        hostname = (parsed.hostname or "").lower().strip()
        if hostname.startswith("www."):
            hostname = hostname[4:]
        if hostname:
            return hostname
    except Exception:
        pass

    # Regex fallback if urlparse failed
    match = re.match(r"^(?:https?://)?(?:www\.)?([^/\s:?#]+)", trimmed, re.IGNORECASE)
    if match:
        return match.group(1).lower().strip()

    return ""


def normalize_links_for_storage(links_data):
    """
    Ensures any link without a title (or with an empty / generic placeholder title)
    has its title automatically set to the domain of the URL.
    Returns the normalized structure ready to be stored in JSONField.
    """
    if not links_data:
        return {}

    if isinstance(links_data, str):
        trimmed = links_data.strip()
        if trimmed.startswith(("{", "[")):
            try:
                import json
                parsed = json.loads(trimmed)
                return normalize_links_for_storage(parsed)
            except (ValueError, TypeError, json.JSONDecodeError):
                pass
        return {}

    # Case 1: Dict format { title: url }
    if isinstance(links_data, dict):
        normalized = {}
        for title, url_or_obj in links_data.items():
            if isinstance(url_or_obj, str) and url_or_obj.strip():
                url = url_or_obj.strip()
                t = str(title).strip() if title else ""
                if not t or t.lower() in ("link", "untitled", "url", "default"):
                    t = extract_domain_from_url(url) or "Link"

                # Deduplicate key if multiple links share the same domain/title
                key = t
                counter = 2
                while key in normalized and normalized[key] != url:
                    key = f"{t} ({counter})"
                    counter += 1
                normalized[key] = url

            elif isinstance(url_or_obj, dict):
                url = url_or_obj.get("url") or url_or_obj.get("link") or url_or_obj.get("href")
                if url and isinstance(url, str) and url.strip():
                    url = url.strip()
                    t = str(url_or_obj.get("title") or title or "").strip()
                    if not t or t.lower() in ("link", "untitled", "url", "default"):
                        t = extract_domain_from_url(url) or "Link"
                    key = t
                    counter = 2
                    while key in normalized and normalized[key] != url:
                        key = f"{t} ({counter})"
                        counter += 1
                    normalized[key] = url
        return normalized

    # Case 2: List format [ {"title": "...", "url": "..."}, ... ]
    if isinstance(links_data, list):
        normalized = []
        for item in links_data:
            if isinstance(item, str) and item.strip():
                url = item.strip()
                title = extract_domain_from_url(url) or "Link"
                normalized.append({"title": title, "url": url})
            elif isinstance(item, dict):
                url = item.get("url") or item.get("link") or item.get("href") or item.get("value")
                if url and isinstance(url, str) and url.strip():
                    url = url.strip()
                    title = str(item.get("title") or item.get("name") or "").strip()
                    if not title or title.lower() in ("link", "untitled", "url", "default"):
                        title = extract_domain_from_url(url) or "Link"
                    normalized.append({"title": title, "url": url})
        return normalized

    return links_data


def format_links_output(links_data):
    """
    Standardizes links into a list of structured objects:
    [
        {"title": "dorar.net", "url": "https://dorar.net/hadith/3420"},
        {"title": "islamweb.net", "url": "https://www.islamweb.net/ar/fatwa/1909"}
    ]
    Handles dict, list of dicts, list of strings, and raw JSON strings cleanly.
    Always ensures title defaults to domain if title is empty or generic 'Link'.
    """
    if not links_data:
        return []

    if isinstance(links_data, str):
        trimmed = links_data.strip()
        if trimmed.startswith(("[", "{")):
            try:
                import json
                parsed = json.loads(trimmed)
                return format_links_output(parsed)
            except (ValueError, TypeError, json.JSONDecodeError):
                pass
        if trimmed:
            return [{"title": extract_domain_from_url(trimmed) or "Link", "url": trimmed}]
        return []

    if isinstance(links_data, dict):
        out = []
        for title, url_or_obj in links_data.items():
            if isinstance(url_or_obj, str) and url_or_obj.strip():
                url = url_or_obj.strip()
                t = str(title).strip() if title else ""
                if not t or t.lower() in ("link", "untitled", "url", "default"):
                    t = extract_domain_from_url(url) or "Link"
                out.append({"title": t, "url": url})
            elif isinstance(url_or_obj, dict):
                url = url_or_obj.get("url") or url_or_obj.get("link") or url_or_obj.get("href")
                if url and isinstance(url, str) and url.strip():
                    url = url.strip()
                    t = str(url_or_obj.get("title") or title or "").strip()
                    if not t or t.lower() in ("link", "untitled", "url", "default"):
                        t = extract_domain_from_url(url) or "Link"
                    out.append({"title": t, "url": url})
            elif isinstance(url_or_obj, list):
                out.extend(format_links_output(url_or_obj))
        return out

    if isinstance(links_data, list):
        out = []
        for item in links_data:
            if isinstance(item, str) and item.strip():
                url = item.strip()
                out.append({
                    "title": extract_domain_from_url(url) or "Link",
                    "url": url
                })
            elif isinstance(item, dict):
                url = item.get("url") or item.get("link") or item.get("href") or item.get("value")
                if url and isinstance(url, str) and url.strip():
                    url = url.strip()
                    t = str(item.get("title") or item.get("name") or "").strip()
                    if not t or t.lower() in ("link", "untitled", "url", "default"):
                        t = extract_domain_from_url(url) or "Link"
                    out.append({"title": t, "url": url})
            elif isinstance(item, list):
                out.extend(format_links_output(item))
        return out

    return []