fix(hianime-provider): use a more specific css selector

.gitignore

@@ -210,3 +210,4 @@ repomix-output.xml
 .project/
 result
 .direnv
+fastanime/libs/provider/anime/hianime/extractors/js/node_modules

fastanime/cli/utils/exception.py

@@ -1,7 +1,10 @@
+import os
 import sys
 
 from rich.traceback import install as rich_install
 
+from ...core.constants import PROJECT_NAME
+
 
 def custom_exception_hook(exc_type, exc_value, exc_traceback):
     print(f"{exc_type.__name__}: {exc_value}")
@@ -16,6 +19,9 @@ def setup_exceptions_handler(
     rich_traceback: bool | None,
     rich_traceback_theme: str,
 ):
+    if dev:
+        # auto set env
+        os.environ[f"{PROJECT_NAME}_DEBUG"] = "1"
     if trace or dev:
         sys.excepthook = default_exception_hook
         if rich_traceback:

fastanime/libs/provider/anime/hianime/constants.py

@@ -0,0 +1,12 @@
+# The base domain for HiAnime.
+HIANIME_DOMAIN = "hianime.to"
+HIANIME_BASE_URL = f"https://{HIANIME_DOMAIN}"
+
+# The endpoint for making AJAX requests (fetching episodes, servers, etc.).
+HIANIME_AJAX_URL = f"{HIANIME_BASE_URL}/ajax"
+
+# The base URL for search queries.
+SEARCH_URL = f"{HIANIME_BASE_URL}/search"
+
+# The Referer header is crucial for making successful requests to the AJAX endpoints.
+AJAX_REFERER_HEADER = f"{HIANIME_BASE_URL}/"

fastanime/libs/provider/anime/hianime/extractors/__init__.py

@@ -0,0 +1,30 @@
+import logging
+from typing import Optional
+
+from ....anime.types import Server
+from .megacloud import MegaCloudExtractor
+
+logger = logging.getLogger(__name__)
+
+
+def extract_server(embed_url: str) -> Optional[Server]:
+    """
+    Acts as a router to select the correct extractor based on the embed URL.
+
+    Args:
+        embed_url: The URL of the video host's embed page.
+
+    Returns:
+        A Server object containing the stream links, or None if extraction fails.
+    """
+    hostname = embed_url.split("/")[2]
+
+    if "megacloud" in hostname or "megaplay" in hostname:
+        return MegaCloudExtractor().extract(embed_url)
+
+    # In the future, you could add other extractors here:
+    # if "streamsb" in hostname:
+    #     return StreamSbExtractor().extract(embed_url)
+
+    logger.warning(f"No extractor found for hostname: {hostname}")
+    return None

fastanime/libs/provider/anime/hianime/extractors/js/megacloud_decrypt.js

@@ -0,0 +1,55 @@
+const CryptoJS = require("crypto-js");
+
+/**
+ * Extracts a secret key from an encrypted string based on an array of index pairs,
+ * then uses that key to decrypt the rest of the string.
+ * @param {string} encryptedString - The full encrypted sources string.
+ * @param {string} varsJson - A JSON string representing an array of [start, length] pairs.
+ * @returns {string} The decrypted JSON string of video sources.
+ */
+function getSecretAndDecrypt(encryptedString, varsJson) {
+  const values = JSON.parse(varsJson);
+  let secret = "";
+  let encryptedSource = "";
+  let encryptedSourceArray = encryptedString.split("");
+  let currentIndex = 0;
+
+  for (const index of values) {
+    const start = index[0] + currentIndex;
+    const end = start + index[1];
+
+    for (let i = start; i < end; i++) {
+      secret += encryptedString[i];
+      encryptedSourceArray[i] = "";
+    }
+    currentIndex += index[1];
+  }
+
+  encryptedSource = encryptedSourceArray.join("");
+
+  const decrypted = CryptoJS.AES.decrypt(encryptedSource, secret).toString(
+    CryptoJS.enc.Utf8,
+  );
+  return decrypted;
+}
+
+// Main execution logic
+const args = process.argv.slice(2);
+if (args.length < 2) {
+  console.error(
+    "Usage: node megacloud_decrypt.js <encryptedString> '<varsJson>'",
+  );
+  process.exit(1);
+}
+
+const encryptedString = args[0];
+const varsJson = args[1];
+
+try {
+  const result = getSecretAndDecrypt(encryptedString, varsJson);
+  // The result is already a JSON string of the sources, just print it to stdout.
+  console.log(result);
+} catch (e) {
+  console.error(e.message);
+  process.exit(1);
+}

fastanime/libs/provider/anime/hianime/extractors/js/package-lock.json

@@ -0,0 +1,21 @@
+{
+  "name": "hianime-extractor-helper",
+  "version": "1.0.0",
+  "lockfileVersion": 3,
+  "requires": true,
+  "packages": {
+    "": {
+      "name": "hianime-extractor-helper",
+      "version": "1.0.0",
+      "dependencies": {
+        "crypto-js": "^4.2.0"
+      }
+    },
+    "node_modules/crypto-js": {
+      "version": "4.2.0",
+      "resolved": "https://registry.npmjs.org/crypto-js/-/crypto-js-4.2.0.tgz",
+      "integrity": "sha512-KALDyEYgpY+Rlob/iriUtjV6d5Eq+Y191A5g4UqLAi8CyGP9N1+FdVbkc1SxKc2r4YAYqG8JzO2KGL+AizD70Q==",
+      "license": "MIT"
+    }
+  }
+}

fastanime/libs/provider/anime/hianime/extractors/js/package.json

@@ -0,0 +1,9 @@
+{
+  "name": "hianime-extractor-helper",
+  "version": "1.0.0",
+  "description": "Helper script to decrypt MegaCloud sources for FastAnime.",
+  "main": "megacloud_decrypt.js",
+  "dependencies": {
+    "crypto-js": "^4.2.0"
+  }
+}

fastanime/libs/provider/anime/hianime/extractors/megacloud.py

@@ -0,0 +1,180 @@
+import json
+import logging
+import re
+import subprocess
+from pathlib import Path
+from typing import List, Optional
+
+import httpx
+
+from ...types import EpisodeStream, Server, Subtitle
+from ..types import HiAnimeSource
+
+logger = logging.getLogger(__name__)
+
+# The path to our Node.js decryption script, relative to this file.
+DECRYPT_SCRIPT_PATH = Path(__file__).parent / "js" / "megacloud_decrypt.js"
+
+
+class MegaCloudExtractor:
+    """
+    Extractor for MegaCloud streams.
+
+    It works by:
+    1. Fetching the embed page.
+    2. Finding the encrypted sources data and the URL to a JavaScript file.
+    3. Fetching the JavaScript file and using regex to find decryption keys.
+    4. Calling an external Node.js script to perform the decryption.
+    5. Parsing the decrypted result to get the final stream URLs.
+    """
+
+    def _run_node_script(self, encrypted_string: str, vars_json: str) -> Optional[dict]:
+        """
+        Executes the Node.js decryption script as a subprocess.
+
+        Args:
+            encrypted_string: The large encrypted sources string.
+            vars_json: A JSON string of the array of indexes for key extraction.
+
+        Returns:
+            The decrypted data as a dictionary, or None on failure.
+        """
+        if not DECRYPT_SCRIPT_PATH.exists():
+            logger.error(
+                f"Node.js decryption script not found at: {DECRYPT_SCRIPT_PATH}"
+            )
+            return None
+
+        command = ["node", str(DECRYPT_SCRIPT_PATH), encrypted_string, vars_json]
+
+        try:
+            process = subprocess.run(
+                command,
+                capture_output=True,
+                text=True,
+                check=True,
+                cwd=DECRYPT_SCRIPT_PATH.parent,  # Run from the 'js' directory
+            )
+            return json.loads(process.stdout)
+        except subprocess.CalledProcessError as e:
+            logger.error(f"Node.js script failed with error: {e.stderr}")
+        except json.JSONDecodeError:
+            logger.error("Failed to parse JSON output from Node.js script.")
+        except Exception as e:
+            logger.error(
+                f"An unexpected error occurred while running Node.js script: {e}"
+            )
+
+        return None
+
+    def extract_vars_from_script(self, script_content: str) -> Optional[str]:
+        """
+        Uses regex to find the variable array needed for decryption from the script content.
+        This pattern is based on the logic from the TypeScript project.
+        """
+        # This regex is a Python adaptation of the one in the TypeScript source.
+        # It looks for the specific pattern that initializes the decryption keys.
+        regex = r"case\s*0x[0-9a-f]+:(?![^;]*=partKey)\s*\w+\s*=\s*(\w+)\s*,\s*\w+\s*=\s*(\w+);"
+        matches = re.findall(regex, script_content)
+
+        if not matches:
+            logger.error("Could not find decryption variables in the script.")
+            return None
+
+        def matching_key(value: str, script: str) -> Optional[str]:
+            # This nested function replicates the `matchingKey` logic from the TS file.
+            key_regex = re.compile(f",{value}=((?:0x)?([0-9a-fA-F]+))")
+            match = key_regex.search(script)
+            return match.group(1) if match else None
+
+        vars_array = []
+        for match in matches:
+            try:
+                key1_hex = matching_key(match[0], script_content)
+                key2_hex = matching_key(match[1], script_content)
+                if key1_hex and key2_hex:
+                    vars_array.append([int(key1_hex, 16), int(key2_hex, 16)])
+            except (ValueError, TypeError):
+                logger.warning(
+                    f"Could not parse hex values from script for match: {match}"
+                )
+                continue
+
+        return json.dumps(vars_array) if vars_array else None
+
+    def extract(self, embed_url: str) -> Optional[Server]:
+        """
+        Main extraction method.
+
+        Args:
+            embed_url: The URL of the MegaCloud embed page.
+
+        Returns:
+            A Server object containing stream links and subtitles.
+        """
+        try:
+            with httpx.Client() as client:
+                # 1. Get the embed page content
+                embed_response = client.get(
+                    embed_url, headers={"Referer": constants.HIANIME_BASE_URL}
+                )
+                embed_response.raise_for_status()
+                embed_html = embed_response.text
+
+                # 2. Find the encrypted sources and the script URL
+                # The data is usually stored in a script tag as `var sources = [...]`.
+                sources_match = re.search(r"var sources = ([^;]+);", embed_html)
+                script_url_match = re.search(
+                    r'src="(/js/player/a/prod/e1-player.min.js\?[^"]+)"', embed_html
+                )
+
+                if not sources_match or not script_url_match:
+                    logger.error("Could not find sources or script URL in embed page.")
+                    return None
+
+                encrypted_sources_data = json.loads(sources_match.group(1))
+                script_url = "https:" + script_url_match.group(1)
+
+                encrypted_string = encrypted_sources_data.get("sources")
+                if not isinstance(encrypted_string, str) or not encrypted_string:
+                    logger.error("Encrypted sources string is missing or invalid.")
+                    return None
+
+                # 3. Fetch the script and extract decryption variables
+                script_response = client.get(script_url)
+                script_response.raise_for_status()
+                vars_json = self.extract_vars_from_script(script_response.text)
+
+                if not vars_json:
+                    return None
+
+                # 4. Decrypt using the Node.js script
+                decrypted_data = self._run_node_script(encrypted_string, vars_json)
+                if not decrypted_data or not isinstance(decrypted_data, list):
+                    logger.error("Decryption failed or returned invalid data.")
+                    return None
+
+                # 5. Map to generic models
+                streams = [
+                    EpisodeStream(
+                        link=source["file"], quality="auto", format=source["type"]
+                    )
+                    for source in decrypted_data
+                ]
+
+                subtitles = [
+                    Subtitle(url=track["file"], language=track.get("label", "en"))
+                    for track in encrypted_sources_data.get("tracks", [])
+                    if track.get("kind") == "captions"
+                ]
+
+                return Server(
+                    name="MegaCloud",
+                    links=streams,
+                    subtitles=subtitles,
+                    headers={"Referer": "https://megacloud.tv/"},
+                )
+
+        except Exception as e:
+            logger.error(f"MegaCloud extraction failed: {e}", exc_info=True)
+            return None

fastanime/libs/provider/anime/hianime/mappers.py

@@ -0,0 +1,149 @@
+import re
+from typing import List, Optional
+
+from ....provider.anime.types import (
+    Anime,
+    AnimeEpisodes,
+    PageInfo,
+    SearchResult,
+    SearchResults,
+)
+from ....provider.scraping.html_parser import (
+    extract_attributes,
+    get_element_by_class,
+    get_elements_by_class,
+)
+
+
+def _parse_episodes(element_html: str) -> AnimeEpisodes:
+    """Helper function to parse sub/dub episode counts from an anime item."""
+    sub_text = get_element_by_class("tick-sub", element_html)
+    dub_text = get_element_by_class("tick-dub", element_html)
+
+    sub_count = 0
+    dub_count = 0
+
+    if sub_text:
+        match = re.search(r"\d+", sub_text)
+        if match:
+            sub_count = int(match.group(0))
+
+    if dub_text:
+        match = re.search(r"\d+", dub_text)
+        if match:
+            dub_count = int(match.group(0))
+
+    # Generate a list of episode numbers as strings
+    sub_list = [str(i) for i in range(1, sub_count + 1)]
+    dub_list = [str(i) for i in range(1, dub_count + 1)]
+
+    return AnimeEpisodes(sub=sub_list, dub=dub_list, raw=[])
+
+
+def map_to_search_results(
+    anime_elements: List[str], full_html: str
+) -> Optional[SearchResults]:
+    """
+    Maps a list of HTML elements from a HiAnime search page to a generic SearchResults object.
+
+    Args:
+        anime_elements: A list of raw HTML strings, each representing an anime (.flw-item).
+        full_html: The full HTML content of the search page for parsing pagination.
+
+    Returns:
+        A SearchResults object or None if parsing fails.
+    """
+    results = []
+    for element in anime_elements:
+        title_element = get_element_by_class("dynamic-name", element)
+        if not title_element:
+            continue
+
+        attrs = extract_attributes(title_element)
+        title = title_element.split(">")[1].split("<")[0].strip()
+        anime_id = attrs.get("href", "").lstrip("/")
+
+        poster_element = get_element_by_class("film-poster-img", element)
+        poster_attrs = extract_attributes(poster_element or "")
+
+        results.append(
+            SearchResult(
+                id=anime_id,
+                title=title,
+                poster=poster_attrs.get("data-src"),
+                episodes=_parse_episodes(element),
+            )
+        )
+
+    # Parse pagination to determine total pages
+    total_pages = 1
+    # Use a simpler selector that is less prone to parsing issues.
+    pagination_elements = get_elements_by_class("page-item", full_html)
+    if pagination_elements:
+        # Find the last page number from all pagination links
+        last_page_num = 0
+        for el in pagination_elements:
+            attrs = extract_attributes(el)
+            href = attrs.get("href", "")
+            if "?page=" in href:
+                try:
+                    num = int(href.split("?page=")[-1])
+                    if num > last_page_num:
+                        last_page_num = num
+                except (ValueError, IndexError):
+                    continue
+        if last_page_num > 0:
+            total_pages = last_page_num
+    page_info = PageInfo(total=total_pages)
+    return SearchResults(page_info=page_info, results=results)
+
+
+def map_to_anime_result(anime_id_slug: str, episode_list_html: str) -> Optional[Anime]:
+    """
+    Maps the AJAX response for an episode list to a generic Anime object.
+
+    Args:
+        anime_id_slug: The anime's unique ID string (e.g., "steinsgate-3").
+        episode_list_html: The raw HTML snippet containing the list of episodes.
+
+    Returns:
+        An Anime object containing the episode list, or None.
+    """
+    episodes = get_elements_by_class("ssl-item", episode_list_html)
+
+    episode_numbers_sub = []
+    # Note: HiAnime's episode list doesn't differentiate sub/dub, so we assume all are sub for now.
+    # The user selects sub/dub when choosing a server later.
+    for ep_element in episodes:
+        attrs = extract_attributes(ep_element)
+        ep_num = attrs.get("data-number")
+        if ep_num:
+            episode_numbers_sub.append(ep_num)
+
+    # The title isn't in this AJAX response, so we derive a placeholder from the slug.
+    # The application's state usually carries the real title from the search/list step.
+    placeholder_title = anime_id_slug.replace("-", " ").title()
+
+    return Anime(
+        id=anime_id_slug,
+        title=placeholder_title,
+        episodes=AnimeEpisodes(
+            sub=episode_numbers_sub,
+            dub=[],  # We don't know dub count from this endpoint
+            raw=[],
+        ),
+    )
+
+
+def map_to_server_id(server_element_html: str) -> Optional[str]:
+    """
+    Extracts the server's unique data-id from its HTML element.
+
+    Args:
+        server_element_html: The raw HTML of a server-item.
+
+    Returns:
+        The server ID string, or None.
+    """
+    attrs = extract_attributes(server_element_html)
+    return attrs.get("data-id")

fastanime/libs/provider/anime/hianime/provider.py

@@ -0,0 +1,168 @@
+import logging
+from typing import Iterator, Optional
+
+from ....provider.anime.base import BaseAnimeProvider
+from ....provider.anime.params import AnimeParams, EpisodeStreamsParams, SearchParams
+from ....provider.anime.types import Anime, SearchResults, Server
+from ....provider.scraping.html_parser import get_elements_by_class
+from . import constants, mappers
+from .extractors import extract_server
+
+logger = logging.getLogger(__name__)
+
+
+class HiAnime(BaseAnimeProvider):
+    """
+    Provider for scraping anime data from HiAnime.
+
+    This provider implements the search, get, and episode_streams methods
+    to fetch anime information and video stream URLs from HiAnime's website
+    and internal AJAX APIs.
+    """
+
+    HEADERS = {"Referer": constants.HIANIME_BASE_URL}
+
+    def search(self, params: SearchParams) -> Optional[SearchResults]:
+        """
+        Searches HiAnime for a given query.
+
+        Args:
+            params: The search parameters containing the query.
+
+        Returns:
+            A SearchResults object containing the found anime, or None.
+        """
+        search_url = f"{constants.SEARCH_URL}?keyword={params.query}"
+        try:
+            response = self.client.get(search_url, follow_redirects=True)
+            response.raise_for_status()
+
+            # The search results are rendered in the HTML. We use our HTML parser
+            # to find all elements with the class '.flw-item', which represent
+            # individual anime search results.
+            anime_elements = get_elements_by_class("flw-item", response.text)
+            if not anime_elements:
+                return None
+
+            # The mapper will convert the raw HTML elements into our generic SearchResults model.
+            return mappers.map_to_search_results(anime_elements, response.text)
+        except Exception as e:
+            logger.error(
+                f"Failed to perform search on HiAnime for query '{params.query}': {e}"
+            )
+            return None
+
+    def get(self, params: AnimeParams) -> Optional[Anime]:
+        """
+        Retrieves detailed information and a list of episodes for a specific anime.
+
+        Args:
+            params: The parameters containing the anime ID (slug).
+
+        Returns:
+            An Anime object with a full episode list, or None.
+        """
+        try:
+            # The numeric ID is the last part of the slug.
+            clean_id_slug = params.id.split("?")[0]
+            anime_id_numeric = clean_id_slug.split("-")[-1]
+            if not anime_id_numeric.isdigit():
+                raise ValueError("Could not extract numeric ID from anime slug.")
+
+            # HiAnime loads episodes via an AJAX request.
+            episodes_url = (
+                f"{constants.HIANIME_AJAX_URL}/v2/episode/list/{anime_id_numeric}"
+            )
+            response = self.client.get(
+                episodes_url,
+                headers={
+                    "X-Requested-With": "XMLHttpRequest",
+                    "Referer": constants.AJAX_REFERER_HEADER,
+                },
+            )
+            response.raise_for_status()
+
+            # The response is JSON containing an 'html' key with the episode list.
+            html_snippet = response.json().get("html", "")
+            if not html_snippet:
+                return None
+
+            # We pass the original anime ID (slug) and the HTML snippet to the mapper.
+            return mappers.map_to_anime_result(params.id, html_snippet)
+        except Exception as e:
+            logger.error(f"Failed to get anime details for '{params.id}': {e}")
+            return None
+
+    def episode_streams(
+        self, params: EpisodeStreamsParams
+    ) -> Optional[Iterator[Server]]:
+        """
+        Fetches the actual video stream URLs for a given episode.
+
+        This is a multi-step process:
+        1. Get the list of available servers (e.g., MegaCloud, StreamSB).
+        2. For each server, get the embed URL.
+        3. Pass the embed URL to an extractor to get the final stream URL.
+
+        Args:
+            params: The parameters containing the episode ID.
+
+        Yields:
+            A Server object for each available video source.
+        """
+        try:
+            # The episode ID is in the format 'anime-slug?ep=12345'
+            episode_id_numeric = params.episode.split("?ep=")[-1]
+            if not episode_id_numeric.isdigit():
+                raise ValueError("Could not extract numeric episode ID.")
+
+            # 1. Get available servers for the episode.
+            servers_url = f"{constants.HIANIME_AJAX_URL}/v2/episode/servers?episodeId={episode_id_numeric}"
+            servers_response = self.client.get(
+                servers_url,
+                headers={
+                    "X-Requested-With": "XMLHttpRequest",
+                    "Referer": constants.AJAX_REFERER_HEADER,
+                },
+            )
+            servers_response.raise_for_status()
+
+            server_elements = get_elements_by_class(
+                "server-item", servers_response.json().get("html", "")
+            )
+
+            for server_element in server_elements:
+                try:
+                    # 2. Extract the server's unique ID.
+                    server_id = mappers.map_to_server_id(server_element)
+                    if not server_id:
+                        continue
+
+                    # 3. Get the embed URL for this server.
+                    sources_url = f"{constants.HIANIME_AJAX_URL}/v2/episode/sources?id={server_id}"
+                    sources_response = self.client.get(
+                        sources_url,
+                        headers={
+                            "X-Requested-With": "XMLHttpRequest",
+                            "Referer": constants.AJAX_REFERER_HEADER,
+                        },
+                    )
+                    sources_response.raise_for_status()
+
+                    embed_url = sources_response.json().get("link")
+                    if not embed_url:
+                        continue
+
+                    # 4. Use an extractor to get the final stream URLs from the embed page.
+                    # The extractor handles the complex, host-specific logic.
+                    server = extract_server(embed_url)
+                    if server:
+                        yield server
+                except Exception as e:
+                    logger.warning(
+                        f"Failed to process a server for episode '{params.episode}': {e}"
+                    )
+                    continue
+        except Exception as e:
+            logger.error(f"Failed to get episode streams for '{params.episode}': {e}")
+            return None

fastanime/libs/provider/anime/hianime/types.py

@@ -0,0 +1,33 @@
+from typing import List, Literal, TypedDict
+
+
+class HiAnimeEpisode(TypedDict):
+    """
+    Represents a single episode entry returned by the
+    `/ajax/v2/episode/list/{anime_id}` endpoint.
+    """
+
+    title: str | None
+    episodeId: str | None
+    number: int
+    isFiller: bool
+
+
+class HiAnimeEpisodeServer(TypedDict):
+    """
+    Represents a single server entry returned by the
+    `/ajax/v2/episode/servers?episodeId={episode_id}` endpoint.
+    """
+
+    serverName: str
+    serverId: int | None
+
+
+class HiAnimeSource(TypedDict):
+    """
+    Represents the JSON response from the
+    `/ajax/v2/episode/sources?id={server_id}` endpoint,
+    which contains the link to the extractor's embed page.
+    """
+
+    link: str

fastanime/libs/provider/anime/provider.py

@@ -12,8 +12,6 @@ PROVIDERS_AVAILABLE = {
     "allanime": "provider.AllAnime",
     "animepahe": "provider.AnimePahe",
     "hianime": "provider.HiAnime",
-    "nyaa": "provider.Nyaa",
-    "yugen": "provider.Yugen",
 }
 
 

fastanime/libs/provider/anime/types.py

@@ -11,6 +11,7 @@ from pydantic import BaseModel, ConfigDict
 class ProviderName(Enum):
     ALLANIME = "allanime"
     ANIMEPAHE = "animepahe"
+    HIANIME = "hianime"
 
 
 class ProviderServer(Enum):