Dovodi_Backend/apps/hadis/utils/links.py

import re
from urllib.parse import urlparse


def ensure_external_url(url: str) -> str:
    """
    Ensures a URL string has an explicit http:// or https:// scheme so browsers
    never treat domain-only links (e.g. 'mamad.com') as relative paths.
    """
    if not url or not isinstance(url, str):
        return ""
    trimmed = url.strip()
    if not trimmed:
        return ""
    if re.match(r"^[a-zA-Z][a-zA-Z0-9+\-.]*://", trimmed):
        return trimmed
    if trimmed.startswith("//"):
        return "https:" + trimmed
    if trimmed.startswith("/"):
        return trimmed
    return "https://" + trimmed.lstrip("/")


def is_dummy_placeholder_link(title: str, url: str) -> bool:
    """
    Detects legacy seeded dummy links ('Первая ссылка...' -> 'google.com')
    so they are never stored or returned.
    """
    t = (title or "").strip()
    u = (url or "").strip().lower()
    if "Первая ссылка" in t:
        return True
    if u in (
        "google.com",
        "https://google.com",
        "http://google.com",
        "https://www.google.com",
        "http://www.google.com",
    ) and ("Первая" in t or t in ("google.com", "Link", "")):
        return True
    return False


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 = ensure_external_url(trimmed)

    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, and ensures the URL
    starts with http:// or https://.
    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():
                raw_url = url_or_obj.strip()
                t = str(title).strip() if title else ""
                if is_dummy_placeholder_link(t, raw_url):
                    continue
                url = ensure_external_url(raw_url)
                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):
                raw_url = url_or_obj.get("url") or url_or_obj.get("link") or url_or_obj.get("href")
                if raw_url and isinstance(raw_url, str) and raw_url.strip():
                    raw_url = raw_url.strip()
                    t = str(url_or_obj.get("title") or title or "").strip()
                    if is_dummy_placeholder_link(t, raw_url):
                        continue
                    url = ensure_external_url(raw_url)
                    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():
                raw_url = item.strip()
                if is_dummy_placeholder_link("", raw_url):
                    continue
                url = ensure_external_url(raw_url)
                title = extract_domain_from_url(url) or "Link"
                normalized.append({"title": title, "url": url})
            elif isinstance(item, dict):
                raw_url = item.get("url") or item.get("link") or item.get("href") or item.get("value")
                if raw_url and isinstance(raw_url, str) and raw_url.strip():
                    raw_url = raw_url.strip()
                    title = str(item.get("title") or item.get("name") or "").strip()
                    if is_dummy_placeholder_link(title, raw_url):
                        continue
                    url = ensure_external_url(raw_url)
                    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',
    and ensures url has an explicit http(s):// scheme.
    """
    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 and not is_dummy_placeholder_link("", trimmed):
            url = ensure_external_url(trimmed)
            return [{"title": extract_domain_from_url(url) or "Link", "url": url}]
        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():
                raw_url = url_or_obj.strip()
                t = str(title).strip() if title else ""
                if is_dummy_placeholder_link(t, raw_url):
                    continue
                url = ensure_external_url(raw_url)
                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):
                raw_url = url_or_obj.get("url") or url_or_obj.get("link") or url_or_obj.get("href")
                if raw_url and isinstance(raw_url, str) and raw_url.strip():
                    raw_url = raw_url.strip()
                    t = str(url_or_obj.get("title") or title or "").strip()
                    if is_dummy_placeholder_link(t, raw_url):
                        continue
                    url = ensure_external_url(raw_url)
                    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():
                raw_url = item.strip()
                if is_dummy_placeholder_link("", raw_url):
                    continue
                url = ensure_external_url(raw_url)
                out.append({
                    "title": extract_domain_from_url(url) or "Link",
                    "url": url
                })
            elif isinstance(item, dict):
                raw_url = item.get("url") or item.get("link") or item.get("href") or item.get("value")
                if raw_url and isinstance(raw_url, str) and raw_url.strip():
                    raw_url = raw_url.strip()
                    t = str(item.get("title") or item.get("name") or "").strip()
                    if is_dummy_placeholder_link(t, raw_url):
                        continue
                    url = ensure_external_url(raw_url)
                    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 []