-
-
Notifications
You must be signed in to change notification settings - Fork 288
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #2362 from dipu-bd/dev
Version 3.6.0
- Loading branch information
Showing
18 changed files
with
574 additions
and
540 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1 +1 @@ | ||
| 3.5.1 | ||
| 3.6.0 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,126 +1,105 @@ | ||
| """ | ||
| To search for novels in selected sources | ||
| """ | ||
| import random | ||
| import logging | ||
| import os | ||
| from concurrent import futures | ||
| from typing import Dict, List | ||
|
|
||
| from concurrent.futures import Future | ||
| from slugify import slugify | ||
| from tqdm import tqdm | ||
|
|
||
| from ..core.sources import crawler_list, prepare_crawler | ||
| from ..models import CombinedSearchResult, SearchResult | ||
| from .sources import crawler_list, prepare_crawler | ||
| from .taskman import TaskManager | ||
|
|
||
| SEARCH_TIMEOUT = 60 | ||
| MAX_RESULTS = 15 | ||
|
|
||
| logger = logging.getLogger(__name__) | ||
| executor = futures.ThreadPoolExecutor(20) | ||
| taskman = TaskManager(10) | ||
|
|
||
|
|
||
| def _perform_search(app, link, bar): | ||
| def _perform_search(app, link): | ||
| from .app import App | ||
| assert isinstance(app, App) | ||
| try: | ||
| crawler = prepare_crawler(link) | ||
| results = [] | ||
| for item in crawler.search_novel(app.user_input): | ||
| if not item.get("url"): | ||
| continue | ||
| if not isinstance(item, SearchResult): | ||
| item = SearchResult(**item) | ||
| if not (item.url and item.title): | ||
| continue | ||
| results.append(item) | ||
|
|
||
| logger.debug(results) | ||
| logger.info("%d results from %s", len(results), link) | ||
| logger.info(f"{len(results)} results from {link}") | ||
| return results | ||
| except KeyboardInterrupt as e: | ||
| raise e | ||
| except Exception: | ||
| if logger.isEnabledFor(logging.DEBUG): | ||
| logging.exception("<!> Search Failed! << %s >>", link) | ||
| return [] | ||
|
|
||
|
|
||
| def _combine_results(results: List[SearchResult]) -> List[CombinedSearchResult]: | ||
| combined: Dict[str, List[SearchResult]] = {} | ||
| for item in results: | ||
| key = slugify(item.title) | ||
| if len(key) <= 2: | ||
| continue | ||
| combined.setdefault(key, []) | ||
| combined[key].append(item) | ||
|
|
||
| processed: List[CombinedSearchResult] = [] | ||
| for key, value in combined.items(): | ||
| value.sort(key=lambda x: x.url) | ||
| processed.append( | ||
| CombinedSearchResult( | ||
| id=key, | ||
| title=value[0].title, | ||
| novels=value, | ||
| ) | ||
| ) | ||
|
|
||
| processed.sort(key=lambda x: -len(x.novels)) | ||
| return processed[:15] # Control the number of results | ||
| finally: | ||
| app.progress += 1 | ||
|
|
||
|
|
||
| def search_novels(app): | ||
| from .app import App | ||
|
|
||
| assert isinstance(app, App) | ||
|
|
||
| if not app.crawler_links: | ||
| return | ||
|
|
||
| sources = app.crawler_links.copy() | ||
| # random.shuffle(sources) | ||
|
|
||
| is_debug = os.getenv("debug_mode") | ||
| bar = tqdm( | ||
| desc="Searching", | ||
| total=len(sources), | ||
| unit="source", | ||
| disable=is_debug, | ||
| ) | ||
| random.shuffle(sources) | ||
|
|
||
| # Add future tasks | ||
| checked = {} | ||
| futures_to_check = [] | ||
| checked = set() | ||
| app.progress = 0 | ||
| futures: List[Future] = [] | ||
| for link in sources: | ||
| crawler = crawler_list[link] | ||
| if crawler in checked: | ||
| bar.update() | ||
| continue | ||
| checked[crawler] = True | ||
| future = executor.submit(_perform_search, app, link, bar) | ||
| futures_to_check.append(future) | ||
| checked.add(crawler) | ||
| f = taskman.submit_task(_perform_search, app, link) | ||
| futures.append(f) | ||
|
|
||
| # Resolve all futures | ||
| results: List[SearchResult] = [] | ||
| for i, f in enumerate(futures_to_check): | ||
| assert isinstance(f, futures.Future) | ||
| try: | ||
| f.result(SEARCH_TIMEOUT) | ||
| except KeyboardInterrupt: | ||
| break | ||
| except TimeoutError: | ||
| f.cancel() | ||
| except Exception as e: | ||
| if is_debug: | ||
| logger.error("Failed to complete search", e) | ||
| finally: | ||
| app.progress += 1 | ||
| bar.update() | ||
| try: | ||
| taskman.resolve_futures( | ||
| futures, | ||
| desc="Searching", | ||
| unit="source", | ||
| timeout=SEARCH_TIMEOUT, | ||
| ) | ||
| except Exception: | ||
| if logger.isEnabledFor(logging.DEBUG): | ||
| logging.exception("<!> Search Failed!") | ||
|
|
||
| # Cancel any remaining futures | ||
| for f in futures_to_check: | ||
| assert isinstance(f, futures.Future) | ||
| if not f.done(): | ||
| f.cancel() | ||
| elif not f.cancelled(): | ||
| results += f.result() | ||
| # Combine the search results | ||
| combined: Dict[str, List[SearchResult]] = {} | ||
| for f in futures: | ||
| if not f or not f.done() or f.cancelled(): | ||
| continue | ||
| for item in f.result() or []: | ||
| if not item: | ||
| continue | ||
| key = slugify(item.title) | ||
| if len(key) <= 2: | ||
| continue | ||
| combined.setdefault(key, []) | ||
| combined[key].append(item) | ||
|
|
||
| # Process combined search results | ||
| app.search_results = _combine_results(results) | ||
| bar.close() | ||
| processed: List[CombinedSearchResult] = [] | ||
| for key, value in combined.items(): | ||
| value.sort(key=lambda x: x.url) | ||
| processed.append( | ||
| CombinedSearchResult( | ||
| id=key, | ||
| title=value[0].title, | ||
| novels=value, | ||
| ) | ||
| ) | ||
| processed.sort(key=lambda x: -len(x.novels)) | ||
| app.search_results = processed[:MAX_RESULTS] |
Oops, something went wrong.