From 73f697f1db835be09bf5b717c25f3565d0c054b1 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Mon, 23 Jan 2023 01:10:06 +0400 Subject: [PATCH 01/22] Re-enable follow_imports. --- setup.cfg | 1 - 1 file changed, 1 deletion(-) diff --git a/setup.cfg b/setup.cfg index 1fab6fe22..c711e6e72 100644 --- a/setup.cfg +++ b/setup.cfg @@ -6,7 +6,6 @@ universal=1 [mypy] ignore_missing_imports = true -follow_imports = skip # FIXME: remove the following sections once the issues are solved From c0ea7fd4fd2726494adf6a84a9fd6b6d7fcf116f Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 22:14:24 +0400 Subject: [PATCH 02/22] Remove obsolete top-level run code in tests. --- tests/test_dependencies.py | 4 ---- tests/test_utils_iterators.py | 4 ---- tests/test_utils_python.py | 4 ---- 3 files changed, 12 deletions(-) diff --git a/tests/test_dependencies.py b/tests/test_dependencies.py index f9b2d853c..a39ed0694 100644 --- a/tests/test_dependencies.py +++ b/tests/test_dependencies.py @@ -37,7 +37,3 @@ class ScrapyUtilsTest(unittest.TestCase): pinned_twisted_version_string = match[1] self.assertEqual(twisted_version.short(), pinned_twisted_version_string) - - -if __name__ == "__main__": - unittest.main() diff --git a/tests/test_utils_iterators.py b/tests/test_utils_iterators.py index 893582a32..778f7162d 100644 --- a/tests/test_utils_iterators.py +++ b/tests/test_utils_iterators.py @@ -528,7 +528,3 @@ class TestHelper(unittest.TestCase): type(a) is type(b), f"Got {type(a)}, expected {type(b)} for { obj!r}" ) self.assertEqual(a, b) - - -if __name__ == "__main__": - unittest.main() diff --git a/tests/test_utils_python.py b/tests/test_utils_python.py index 5caa5b8f2..8d2695b0d 100644 --- a/tests/test_utils_python.py +++ b/tests/test_utils_python.py @@ -258,7 +258,3 @@ class UtilsPythonTestCase(unittest.TestCase): without_none_values({"one": 1, "none": None, "three": 3, "four": 4}), {"one": 1, "three": 3, "four": 4}, ) - - -if __name__ == "__main__": - unittest.main() From 2e33fb812b227b8f5ded8be9913467c31929a87b Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 22:15:07 +0400 Subject: [PATCH 03/22] Silence improper typing of twisted.internet.reactor in tests. --- scrapy/utils/benchserver.py | 6 +++--- tests/CrawlerRunner/ip_address.py | 6 +++--- tests/mockserver.py | 13 ++++++++----- tests/test_engine.py | 2 +- 4 files changed, 15 insertions(+), 12 deletions(-) diff --git a/scrapy/utils/benchserver.py b/scrapy/utils/benchserver.py index 32bc2e38c..750d3c093 100644 --- a/scrapy/utils/benchserver.py +++ b/scrapy/utils/benchserver.py @@ -37,11 +37,11 @@ if __name__ == "__main__": root = Root() factory = Site(root) - httpPort = reactor.listenTCP(8998, Site(root)) + httpPort = reactor.listenTCP(8998, Site(root)) # type: ignore[attr-defined] def _print_listening(): httpHost = httpPort.getHost() print(f"Bench server at http://{httpHost.host}:{httpHost.port}") - reactor.callWhenRunning(_print_listening) - reactor.run() + reactor.callWhenRunning(_print_listening) # type: ignore[attr-defined] + reactor.run() # type: ignore[attr-defined] diff --git a/tests/CrawlerRunner/ip_address.py b/tests/CrawlerRunner/ip_address.py index 26db16dd6..b9a4485a9 100644 --- a/tests/CrawlerRunner/ip_address.py +++ b/tests/CrawlerRunner/ip_address.py @@ -41,10 +41,10 @@ if __name__ == "__main__": url = f"http://not.a.real.domain:{port}/echo" servers = [(mock_dns_server.host, mock_dns_server.port)] - reactor.installResolver(createResolver(servers=servers)) + reactor.installResolver(createResolver(servers=servers)) # type: ignore[attr-defined] configure_logging() runner = CrawlerRunner() d = runner.crawl(LocalhostSpider, url=url) - d.addBoth(lambda _: reactor.stop()) - reactor.run() + d.addBoth(lambda _: reactor.stop()) # type: ignore[attr-defined] + reactor.run() # type: ignore[attr-defined] diff --git a/tests/mockserver.py b/tests/mockserver.py index e07ae8797..185897373 100644 --- a/tests/mockserver.py +++ b/tests/mockserver.py @@ -10,6 +10,7 @@ from urllib.parse import urlencode from OpenSSL import SSL from twisted.internet import defer, reactor, ssl +from twisted.internet.protocol import ServerFactory from twisted.internet.task import deferLater from twisted.names import dns, error from twisted.names.server import DNSServerFactory @@ -369,12 +370,14 @@ if __name__ == "__main__": ) args = parser.parse_args() + factory: ServerFactory + if args.type == "http": root = Root() factory = Site(root) - httpPort = reactor.listenTCP(0, factory) + httpPort = reactor.listenTCP(0, factory) # type: ignore[attr-defined] contextFactory = ssl_context_factory() - httpsPort = reactor.listenSSL(0, factory, contextFactory) + httpsPort = reactor.listenSSL(0, factory, contextFactory) # type: ignore[attr-defined] def print_listening(): httpHost = httpPort.getHost() @@ -388,11 +391,11 @@ if __name__ == "__main__": clients = [MockDNSResolver()] factory = DNSServerFactory(clients=clients) protocol = dns.DNSDatagramProtocol(controller=factory) - listener = reactor.listenUDP(0, protocol) + listener = reactor.listenUDP(0, protocol) # type: ignore[attr-defined] def print_listening(): host = listener.getHost() print(f"{host.host}:{host.port}") - reactor.callWhenRunning(print_listening) - reactor.run() + reactor.callWhenRunning(print_listening) # type: ignore[attr-defined] + reactor.run() # type: ignore[attr-defined] diff --git a/tests/test_engine.py b/tests/test_engine.py index 7ddb420ba..5fb87424b 100644 --- a/tests/test_engine.py +++ b/tests/test_engine.py @@ -557,4 +557,4 @@ class EngineTest(unittest.TestCase): if __name__ == "__main__": if len(sys.argv) > 1 and sys.argv[1] == "runserver": start_test_site(debug=True) - reactor.run() + reactor.run() # type: ignore[attr-defined] From afd5d8532093c25a6ed4453e3a10b275335c94bd Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Mon, 23 Jan 2023 01:12:59 +0400 Subject: [PATCH 04/22] Fix remaining issues in tests. --- tests/test_downloadermiddleware_httpproxy.py | 2 +- tests/test_http2_client_protocol.py | 5 ++++- tests/test_spidermiddleware.py | 1 + 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/tests/test_downloadermiddleware_httpproxy.py b/tests/test_downloadermiddleware_httpproxy.py index ca125ba36..0040f8d23 100644 --- a/tests/test_downloadermiddleware_httpproxy.py +++ b/tests/test_downloadermiddleware_httpproxy.py @@ -14,7 +14,7 @@ spider = Spider("foo") class TestHttpProxyMiddleware(TestCase): - failureException = AssertionError + failureException = AssertionError # type: ignore[assignment] def setUp(self): self._oldenv = os.environ.copy() diff --git a/tests/test_http2_client_protocol.py b/tests/test_http2_client_protocol.py index 88345d2bc..2a928eea0 100644 --- a/tests/test_http2_client_protocol.py +++ b/tests/test_http2_client_protocol.py @@ -5,6 +5,7 @@ import shutil import string from ipaddress import IPv4Address from pathlib import Path +from typing import Dict from unittest import mock, skipIf from urllib.parse import urlencode @@ -86,6 +87,7 @@ class GetDataHtmlLarge(LeafResource): class PostDataJsonMixin: @staticmethod def make_response(request: TxRequest, extra_data: str): + assert request.content is not None response = { "request-headers": {}, "request-body": json.loads(request.content.read()), @@ -144,7 +146,8 @@ class QueryParams(LeafResource): request.setHeader("Content-Type", "application/json; charset=UTF-8") request.setHeader("Content-Encoding", "UTF-8") - query_params = {} + query_params: Dict[str, str] = {} + assert request.args is not None for k, v in request.args.items(): query_params[str(k, "utf-8")] = str(v[0], "utf-8") diff --git a/tests/test_spidermiddleware.py b/tests/test_spidermiddleware.py index 760ee43df..ca6348913 100644 --- a/tests/test_spidermiddleware.py +++ b/tests/test_spidermiddleware.py @@ -112,6 +112,7 @@ class BaseAsyncSpiderMiddlewareTestCase(SpiderMiddlewareTestCase): Should work for process_spider_output and, when it's supported, process_start_requests. """ + ITEM_TYPE: type RESULT_COUNT = 3 # to simplify checks, let everything return 3 objects @staticmethod From 232aab53b38553262d60259461adf3fa8054e445 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 22:28:15 +0400 Subject: [PATCH 05/22] Fix most of the new typing issues in utils. --- scrapy/utils/defer.py | 19 +++++++++++-------- scrapy/utils/response.py | 7 ++++--- scrapy/utils/testsite.py | 4 ++-- 3 files changed, 17 insertions(+), 13 deletions(-) diff --git a/scrapy/utils/defer.py b/scrapy/utils/defer.py index 21cd5e78f..8fee5a7ed 100644 --- a/scrapy/utils/defer.py +++ b/scrapy/utils/defer.py @@ -10,6 +10,7 @@ from typing import ( AsyncGenerator, AsyncIterable, Callable, + cast, Coroutine, Generator, Iterable, @@ -38,8 +39,8 @@ def defer_fail(_failure: Failure) -> Deferred: """ from twisted.internet import reactor - d = Deferred() - reactor.callLater(0.1, d.errback, _failure) + d: Deferred = Deferred() + reactor.callLater(0.1, d.errback, _failure) # type: ignore[attr-defined] return d @@ -52,8 +53,8 @@ def defer_succeed(result) -> Deferred: """ from twisted.internet import reactor - d = Deferred() - reactor.callLater(0.1, d.callback, result) + d: Deferred = Deferred() + reactor.callLater(0.1, d.callback, result) # type: ignore[attr-defined] return d @@ -182,7 +183,9 @@ class _AsyncCooperatorAdapter(Iterator): def _call_anext(self) -> None: # This starts waiting for the next result from aiterator. # If aiterator is exhausted, _errback will be called. - self.anext_deferred = deferred_from_coro(self.aiterator.__anext__()) + self.anext_deferred = cast( + Deferred, deferred_from_coro(self.aiterator.__anext__()) + ) self.anext_deferred.addCallbacks(self._callback, self._errback) def __next__(self) -> Deferred: @@ -190,7 +193,7 @@ class _AsyncCooperatorAdapter(Iterator): # It also calls __anext__() if needed. if self.finished: raise StopIteration - d = Deferred() + d: Deferred = Deferred() self.waiting_deferreds.append(d) if not self.anext_deferred: self._call_anext() @@ -209,7 +212,7 @@ def parallel_async( def process_chain(callbacks: Iterable[Callable], input, *a, **kw) -> Deferred: """Return a Deferred built by chaining the given callbacks""" - d = Deferred() + d: Deferred = Deferred() for x in callbacks: d.addCallback(x, *a, **kw) d.callback(input) @@ -220,7 +223,7 @@ def process_chain_both( callbacks: Iterable[Callable], errbacks: Iterable[Callable], input, *a, **kw ) -> Deferred: """Return a Deferred built by chaining the given callbacks and errbacks""" - d = Deferred() + d: Deferred = Deferred() for cb, eb in zip(callbacks, errbacks): d.addCallbacks( callback=cb, diff --git a/scrapy/utils/response.py b/scrapy/utils/response.py index a91a49170..40a32309a 100644 --- a/scrapy/utils/response.py +++ b/scrapy/utils/response.py @@ -6,7 +6,7 @@ import os import re import tempfile import webbrowser -from typing import Any, Callable, Iterable, Optional, Tuple, Union +from typing import Any, Callable, Iterable, Tuple, Union from weakref import WeakKeyDictionary from twisted.web import http @@ -38,12 +38,13 @@ _metaref_cache: "WeakKeyDictionary[Response, Union[Tuple[None, None], Tuple[floa def get_meta_refresh( response: "scrapy.http.response.text.TextResponse", - ignore_tags: Optional[Iterable[str]] = ("script", "noscript"), + ignore_tags: Iterable[str] = ("script", "noscript"), ) -> Union[Tuple[None, None], Tuple[float, str]]: """Parse the http-equiv refresh parameter from the given response""" if response not in _metaref_cache: text = response.text[0:4096] - _metaref_cache[response] = html.get_meta_refresh( + # a w3lib typing bug here + _metaref_cache[response] = html.get_meta_refresh( # type: ignore[assignment] text, response.url, response.encoding, ignore_tags=ignore_tags ) return _metaref_cache[response] diff --git a/scrapy/utils/testsite.py b/scrapy/utils/testsite.py index a47756c4b..119be1dfb 100644 --- a/scrapy/utils/testsite.py +++ b/scrapy/utils/testsite.py @@ -50,6 +50,6 @@ def test_site(): if __name__ == "__main__": from twisted.internet import reactor - port = reactor.listenTCP(0, test_site(), interface="127.0.0.1") + port = reactor.listenTCP(0, test_site(), interface="127.0.0.1") # type: ignore[attr-defined] print(f"http://localhost:{port.getHost().port}/") - reactor.run() + reactor.run() # type: ignore[attr-defined] From 764a9d47bb95e87519a4fc2fe20ddf0c97b11739 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 22:31:12 +0400 Subject: [PATCH 06/22] Fix typing of inlineCallbacks-decorated functions. --- scrapy/core/engine.py | 6 +++--- scrapy/core/spidermw.py | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/scrapy/core/engine.py b/scrapy/core/engine.py index 19696415b..cafa7f6da 100644 --- a/scrapy/core/engine.py +++ b/scrapy/core/engine.py @@ -7,7 +7,7 @@ For more information see docs/topics/architecture.rst import logging import warnings from time import time -from typing import Callable, Iterable, Iterator, Optional, Set, Union +from typing import Any, Callable, Generator, Iterable, Iterator, Optional, Set, Union from twisted.internet.defer import Deferred, inlineCallbacks, succeed from twisted.internet.task import LoopingCall @@ -96,7 +96,7 @@ class ExecutionEngine: return scheduler_cls @inlineCallbacks - def start(self) -> Deferred: + def start(self) -> Generator[Deferred, Any, None]: if self.running: raise RuntimeError("Engine already running") self.start_time = time() @@ -109,7 +109,7 @@ class ExecutionEngine: """Gracefully stop the execution engine""" @inlineCallbacks - def _finish_stopping_engine(_) -> Deferred: + def _finish_stopping_engine(_) -> Generator[Deferred, Any, None]: yield self.signals.send_catch_log_deferred(signal=signals.engine_stopped) self._closewait.callback(None) diff --git a/scrapy/core/spidermw.py b/scrapy/core/spidermw.py index 1aaed5865..0bc8d54d2 100644 --- a/scrapy/core/spidermw.py +++ b/scrapy/core/spidermw.py @@ -182,7 +182,7 @@ class SpiderMiddlewareManager(MiddlewareManager): spider: Spider, result: Union[Iterable, AsyncIterable], start_index: int = 0, - ) -> Deferred: + ) -> Generator[Deferred, Any, Union[MutableChain, MutableAsyncChain]]: # items in this iterable do not need to go through the process_spider_output # chain, they went through it already from the process_spider_exception method recovered: Union[MutableChain, MutableAsyncChain] From 5fde6d533903441fe3d986030319de3415a9b956 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 22:54:46 +0400 Subject: [PATCH 07/22] Don't type-check Twisted interfaces. --- setup.cfg | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/setup.cfg b/setup.cfg index c711e6e72..af9c87945 100644 --- a/setup.cfg +++ b/setup.cfg @@ -7,6 +7,10 @@ universal=1 [mypy] ignore_missing_imports = true +# Interface classes are hard to support +[mypy-twisted.internet.interfaces] +follow_imports = skip + # FIXME: remove the following sections once the issues are solved [mypy-scrapy.downloadermiddlewares.httpproxy] From e9094d1f38f2ed2904a6955004a8069054c8e35b Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 23:22:18 +0400 Subject: [PATCH 08/22] Address remaining typing issues in scrapy.core. --- scrapy/core/downloader/__init__.py | 2 +- scrapy/core/downloader/handlers/http2.py | 8 +++---- scrapy/core/engine.py | 27 ++++++++++++++++++------ scrapy/core/http2/agent.py | 10 +++++---- scrapy/core/http2/protocol.py | 7 ++++++ scrapy/core/http2/stream.py | 2 +- scrapy/core/scraper.py | 8 ++++--- scrapy/core/spidermw.py | 2 +- 8 files changed, 45 insertions(+), 21 deletions(-) diff --git a/scrapy/core/downloader/__init__.py b/scrapy/core/downloader/__init__.py index 3a7de8072..e66156ba1 100644 --- a/scrapy/core/downloader/__init__.py +++ b/scrapy/core/downloader/__init__.py @@ -95,7 +95,7 @@ class Downloader: dfd = self.middleware.download(self._enqueue_request, request, spider) return dfd.addBoth(_deactivate) - def needs_backout(self): + def needs_backout(self) -> bool: return len(self.active) >= self.total_concurrency def _get_slot(self, request, spider): diff --git a/scrapy/core/downloader/handlers/http2.py b/scrapy/core/downloader/handlers/http2.py index 25ac0307b..20cd50c5a 100644 --- a/scrapy/core/downloader/handlers/http2.py +++ b/scrapy/core/downloader/handlers/http2.py @@ -28,7 +28,7 @@ class H2DownloadHandler: from twisted.internet import reactor - self._pool = H2ConnectionPool(reactor, settings) + self._pool = H2ConnectionPool(reactor, settings) # type: ignore[arg-type] self._context_factory = load_context_factory_from_settings(settings, crawler) @classmethod @@ -82,7 +82,7 @@ class ScrapyH2Agent: "Tunneling via CONNECT method using HTTP/2.0 is not yet supported" ) return self._ProxyAgent( - reactor=reactor, + reactor=reactor, # type: ignore[arg-type] context_factory=self._context_factory, proxy_uri=URI.fromBytes(to_bytes(proxy, encoding="ascii")), connect_timeout=timeout, @@ -91,7 +91,7 @@ class ScrapyH2Agent: ) return self._Agent( - reactor=reactor, + reactor=reactor, # type: ignore[arg-type] context_factory=self._context_factory, connect_timeout=timeout, bind_address=bind_address, @@ -108,7 +108,7 @@ class ScrapyH2Agent: d = agent.request(request, spider) d.addCallback(self._cb_latency, request, start_time) - timeout_cl = reactor.callLater(timeout, d.cancel) + timeout_cl = reactor.callLater(timeout, d.cancel) # type: ignore[attr-defined] d.addBoth(self._cb_timeout, request, timeout, timeout_cl) return d diff --git a/scrapy/core/engine.py b/scrapy/core/engine.py index cafa7f6da..e44675ef0 100644 --- a/scrapy/core/engine.py +++ b/scrapy/core/engine.py @@ -7,13 +7,24 @@ For more information see docs/topics/architecture.rst import logging import warnings from time import time -from typing import Any, Callable, Generator, Iterable, Iterator, Optional, Set, Union +from typing import ( + Any, + Callable, + cast, + Generator, + Iterable, + Iterator, + Optional, + Set, + Union, +) from twisted.internet.defer import Deferred, inlineCallbacks, succeed from twisted.internet.task import LoopingCall from twisted.python.failure import Failure from scrapy import signals +from scrapy.core.downloader import Downloader from scrapy.core.scraper import Scraper from scrapy.exceptions import ( CloseSpider, @@ -80,7 +91,7 @@ class ExecutionEngine: self.paused = False self.scheduler_cls = self._get_scheduler_class(crawler.settings) downloader_cls = load_object(self.settings["DOWNLOADER"]) - self.downloader = downloader_cls(crawler) + self.downloader: Downloader = downloader_cls(crawler) self.scraper = Scraper(crawler) self._spider_closed_callback = spider_closed_callback @@ -102,7 +113,7 @@ class ExecutionEngine: self.start_time = time() yield self.signals.send_catch_log_deferred(signal=signals.engine_started) self.running = True - self._closewait = Deferred() + self._closewait: Deferred = Deferred() yield self._closewait def stop(self) -> Deferred: @@ -177,11 +188,13 @@ class ExecutionEngine: self._spider_idle() def _needs_backout(self) -> bool: + assert self.slot is not None # typing + assert self.scraper.slot is not None # typing return ( not self.running - or self.slot.closing # type: ignore[union-attr] + or bool(self.slot.closing) or self.downloader.needs_backout() - or self.scraper.slot.needs_backout() # type: ignore[union-attr] + or self.scraper.slot.needs_backout() ) def _next_request_from_scheduler(self) -> Optional[Deferred]: @@ -201,7 +214,7 @@ class ExecutionEngine: extra={"spider": self.spider}, ) ) - d.addBoth(lambda _: self.slot.remove_request(request)) + d.addBoth(lambda _: cast(Slot, self.slot).remove_request(request)) d.addErrback( lambda f: logger.info( "Error while removing request from slot", @@ -429,7 +442,7 @@ class ExecutionEngine: dfd.addErrback(log_failure("Scraper close failure")) if hasattr(self.slot.scheduler, "close"): - dfd.addBoth(lambda _: self.slot.scheduler.close(reason)) + dfd.addBoth(lambda _: cast(Slot, self.slot).scheduler.close(reason)) dfd.addErrback(log_failure("Scheduler close failure")) dfd.addBoth( diff --git a/scrapy/core/http2/agent.py b/scrapy/core/http2/agent.py index 119443c80..3ceda9b61 100644 --- a/scrapy/core/http2/agent.py +++ b/scrapy/core/http2/agent.py @@ -35,7 +35,7 @@ class H2ConnectionPool: # Received a request while connecting to remote # Create a deferred which will fire with the H2ClientProtocol # instance - d = Deferred() + d: Deferred = Deferred() self._pending_requests[key].append(d) return d @@ -53,14 +53,14 @@ class H2ConnectionPool: ) -> Deferred: self._pending_requests[key] = deque() - conn_lost_deferred = Deferred() + conn_lost_deferred: Deferred = Deferred() conn_lost_deferred.addCallback(self._remove_connection, key) factory = H2ClientFactory(uri, self.settings, conn_lost_deferred) conn_d = endpoint.connect(factory) conn_d.addCallback(self.put_connection, key) - d = Deferred() + d: Deferred = Deferred() self._pending_requests[key].append(d) return d @@ -83,7 +83,8 @@ class H2ConnectionPool: pending_requests = self._pending_requests.pop(key, None) while pending_requests: d = pending_requests.popleft() - d.errback(errors) + # TODO: this is incorrect, errback takes a single exception + d.errback(errors) # type: ignore[arg-type] def close_connections(self) -> None: """Close all the HTTP/2 connections and remove them from pool @@ -92,6 +93,7 @@ class H2ConnectionPool: Deferred that fires when all connections have been closed """ for conn in self._connections.values(): + assert conn.transport is not None # typing conn.transport.abortConnection() diff --git a/scrapy/core/http2/protocol.py b/scrapy/core/http2/protocol.py index 214deeed0..e44aed43c 100644 --- a/scrapy/core/http2/protocol.py +++ b/scrapy/core/http2/protocol.py @@ -141,6 +141,7 @@ class H2ClientProtocol(Protocol, TimeoutMixin): This is used while initiating pending streams to make sure that we initiate stream only during active HTTP/2 Connection """ + assert self.transport is not None # typing return bool(self.transport.connected) and self.metadata["settings_acknowledged"] @property @@ -197,6 +198,7 @@ class H2ClientProtocol(Protocol, TimeoutMixin): """Write data to the underlying transport connection from the HTTP2 connection instance if any """ + assert self.transport is not None # typing # Reset the idle timeout as connection is still actively sending data self.resetTimeout() @@ -227,6 +229,7 @@ class H2ClientProtocol(Protocol, TimeoutMixin): # Initialize the timeout self.setTimeout(self.IDLE_TIMEOUT) + assert self.transport is not None # typing destination = self.transport.getPeer() self.metadata["ip_address"] = ipaddress.ip_address(destination.host) @@ -238,12 +241,14 @@ class H2ClientProtocol(Protocol, TimeoutMixin): """Helper function to lose the connection with the error sent as a reason""" self._conn_lost_errors += errors + assert self.transport is not None # typing self.transport.loseConnection() def handshakeCompleted(self) -> None: """ Close the connection if it's not made via the expected protocol """ + assert self.transport is not None # typing if ( self.transport.negotiatedProtocol is not None and self.transport.negotiatedProtocol != PROTOCOL_NAME @@ -276,6 +281,7 @@ class H2ClientProtocol(Protocol, TimeoutMixin): # hyper-h2 does not drop the connection in this scenario, we # need to abort the connection manually. self._conn_lost_errors += [e] + assert self.transport is not None # typing self.transport.abortConnection() return @@ -389,6 +395,7 @@ class H2ClientProtocol(Protocol, TimeoutMixin): self._send_pending_requests() # Update certificate when our HTTP/2 connection is established + assert self.transport is not None # typing self.metadata["certificate"] = Certificate(self.transport.getPeerCertificate()) def stream_ended(self, event: StreamEnded) -> None: diff --git a/scrapy/core/http2/stream.py b/scrapy/core/http2/stream.py index 2b5c98e5f..5324a5ca1 100644 --- a/scrapy/core/http2/stream.py +++ b/scrapy/core/http2/stream.py @@ -151,7 +151,7 @@ class Stream: else: self.close(StreamCloseReason.CANCELLED) - self._deferred_response = Deferred(_cancel) + self._deferred_response: Deferred = Deferred(_cancel) def __repr__(self) -> str: return f"Stream(id={self.stream_id!r})" diff --git a/scrapy/core/scraper.py b/scrapy/core/scraper.py index 7c2eefbe6..c1488a46c 100644 --- a/scrapy/core/scraper.py +++ b/scrapy/core/scraper.py @@ -66,7 +66,7 @@ class Slot: def add_response_request( self, result: Union[Response, Failure], request: Request ) -> Deferred: - deferred = Deferred() + deferred: Deferred = Deferred() self.queue.append((result, request, deferred)) if isinstance(result, Response): self.active_size += max(len(result.body), self.MIN_RESPONSE_SIZE) @@ -205,10 +205,12 @@ class Scraper: callback=callback, callbackKeywords=result.request.cb_kwargs ) else: # result is a Failure - result.request = request + # TODO: properly type adding this attribute to a Failure + result.request = request # type: ignore[attr-defined] warn_on_generator_with_return_value(spider, request.errback) dfd = defer_fail(result) - dfd.addErrback(request.errback) + if request.errback: + dfd.addErrback(request.errback) return dfd.addCallback(iterate_spider_output) def handle_spider_error( diff --git a/scrapy/core/spidermw.py b/scrapy/core/spidermw.py index 0bc8d54d2..31e815b19 100644 --- a/scrapy/core/spidermw.py +++ b/scrapy/core/spidermw.py @@ -157,7 +157,7 @@ class SpiderMiddlewareManager(MiddlewareManager): # simplified when downgrading is removed. if dfd.called: # the result is available immediately if _process_spider_output didn't do downgrading - return dfd.result + return cast(MutableChain, dfd.result) # we forbid waiting here because otherwise we would need to return a deferred from # _process_spider_exception too, which complicates the architecture msg = f"Async iterable returned from {method.__qualname__} cannot be downgraded" From a4edff31b91c4647a02bb392020110856b669fd9 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 23:30:21 +0400 Subject: [PATCH 09/22] Replace lxml-stubs with types-lxml. --- tox.ini | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tox.ini b/tox.ini index eaf29173a..f05ebc45a 100644 --- a/tox.ini +++ b/tox.ini @@ -37,9 +37,9 @@ install_command = [testenv:typing] basepython = python3 deps = - lxml-stubs==0.2.0 mypy==0.991 types-attrs==19.1.0 + types-lxml==2022.11.8 types-pyOpenSSL==21.0.0 types-setuptools==57.0.0 commands = @@ -201,4 +201,4 @@ deps = black==22.12.0 commands = black {posargs:--check .} - \ No newline at end of file + From 69bb9a78594ed864291d45d36108cc0ad960f048 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Wed, 25 Jan 2023 23:52:46 +0400 Subject: [PATCH 10/22] Install types-Pillow and types-Pygments. --- tox.ini | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tox.ini b/tox.ini index f05ebc45a..7048391ac 100644 --- a/tox.ini +++ b/tox.ini @@ -40,6 +40,8 @@ deps = mypy==0.991 types-attrs==19.1.0 types-lxml==2022.11.8 + types-Pillow==9.4.0.5 + types-Pygments==2.14.0.1 types-pyOpenSSL==21.0.0 types-setuptools==57.0.0 commands = From c0efb271a23218656b5f629fde6d23c115a0c3de Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 27 Jan 2023 23:19:44 +0400 Subject: [PATCH 11/22] Ignore typing for twisted.internet.reactor globally. --- scrapy/core/downloader/handlers/http2.py | 8 ++++---- scrapy/utils/benchserver.py | 6 +++--- scrapy/utils/defer.py | 4 ++-- scrapy/utils/testsite.py | 4 ++-- setup.cfg | 3 +++ tests/CrawlerRunner/ip_address.py | 6 +++--- tests/mockserver.py | 10 +++++----- tests/test_engine.py | 2 +- 8 files changed, 23 insertions(+), 20 deletions(-) diff --git a/scrapy/core/downloader/handlers/http2.py b/scrapy/core/downloader/handlers/http2.py index 20cd50c5a..25ac0307b 100644 --- a/scrapy/core/downloader/handlers/http2.py +++ b/scrapy/core/downloader/handlers/http2.py @@ -28,7 +28,7 @@ class H2DownloadHandler: from twisted.internet import reactor - self._pool = H2ConnectionPool(reactor, settings) # type: ignore[arg-type] + self._pool = H2ConnectionPool(reactor, settings) self._context_factory = load_context_factory_from_settings(settings, crawler) @classmethod @@ -82,7 +82,7 @@ class ScrapyH2Agent: "Tunneling via CONNECT method using HTTP/2.0 is not yet supported" ) return self._ProxyAgent( - reactor=reactor, # type: ignore[arg-type] + reactor=reactor, context_factory=self._context_factory, proxy_uri=URI.fromBytes(to_bytes(proxy, encoding="ascii")), connect_timeout=timeout, @@ -91,7 +91,7 @@ class ScrapyH2Agent: ) return self._Agent( - reactor=reactor, # type: ignore[arg-type] + reactor=reactor, context_factory=self._context_factory, connect_timeout=timeout, bind_address=bind_address, @@ -108,7 +108,7 @@ class ScrapyH2Agent: d = agent.request(request, spider) d.addCallback(self._cb_latency, request, start_time) - timeout_cl = reactor.callLater(timeout, d.cancel) # type: ignore[attr-defined] + timeout_cl = reactor.callLater(timeout, d.cancel) d.addBoth(self._cb_timeout, request, timeout, timeout_cl) return d diff --git a/scrapy/utils/benchserver.py b/scrapy/utils/benchserver.py index 750d3c093..32bc2e38c 100644 --- a/scrapy/utils/benchserver.py +++ b/scrapy/utils/benchserver.py @@ -37,11 +37,11 @@ if __name__ == "__main__": root = Root() factory = Site(root) - httpPort = reactor.listenTCP(8998, Site(root)) # type: ignore[attr-defined] + httpPort = reactor.listenTCP(8998, Site(root)) def _print_listening(): httpHost = httpPort.getHost() print(f"Bench server at http://{httpHost.host}:{httpHost.port}") - reactor.callWhenRunning(_print_listening) # type: ignore[attr-defined] - reactor.run() # type: ignore[attr-defined] + reactor.callWhenRunning(_print_listening) + reactor.run() diff --git a/scrapy/utils/defer.py b/scrapy/utils/defer.py index 8fee5a7ed..7f2211877 100644 --- a/scrapy/utils/defer.py +++ b/scrapy/utils/defer.py @@ -40,7 +40,7 @@ def defer_fail(_failure: Failure) -> Deferred: from twisted.internet import reactor d: Deferred = Deferred() - reactor.callLater(0.1, d.errback, _failure) # type: ignore[attr-defined] + reactor.callLater(0.1, d.errback, _failure) return d @@ -54,7 +54,7 @@ def defer_succeed(result) -> Deferred: from twisted.internet import reactor d: Deferred = Deferred() - reactor.callLater(0.1, d.callback, result) # type: ignore[attr-defined] + reactor.callLater(0.1, d.callback, result) return d diff --git a/scrapy/utils/testsite.py b/scrapy/utils/testsite.py index 119be1dfb..a47756c4b 100644 --- a/scrapy/utils/testsite.py +++ b/scrapy/utils/testsite.py @@ -50,6 +50,6 @@ def test_site(): if __name__ == "__main__": from twisted.internet import reactor - port = reactor.listenTCP(0, test_site(), interface="127.0.0.1") # type: ignore[attr-defined] + port = reactor.listenTCP(0, test_site(), interface="127.0.0.1") print(f"http://localhost:{port.getHost().port}/") - reactor.run() # type: ignore[attr-defined] + reactor.run() diff --git a/setup.cfg b/setup.cfg index af9c87945..db79c5821 100644 --- a/setup.cfg +++ b/setup.cfg @@ -11,6 +11,9 @@ ignore_missing_imports = true [mypy-twisted.internet.interfaces] follow_imports = skip +[mypy-twisted.internet.reactor] +follow_imports = skip + # FIXME: remove the following sections once the issues are solved [mypy-scrapy.downloadermiddlewares.httpproxy] diff --git a/tests/CrawlerRunner/ip_address.py b/tests/CrawlerRunner/ip_address.py index b9a4485a9..26db16dd6 100644 --- a/tests/CrawlerRunner/ip_address.py +++ b/tests/CrawlerRunner/ip_address.py @@ -41,10 +41,10 @@ if __name__ == "__main__": url = f"http://not.a.real.domain:{port}/echo" servers = [(mock_dns_server.host, mock_dns_server.port)] - reactor.installResolver(createResolver(servers=servers)) # type: ignore[attr-defined] + reactor.installResolver(createResolver(servers=servers)) configure_logging() runner = CrawlerRunner() d = runner.crawl(LocalhostSpider, url=url) - d.addBoth(lambda _: reactor.stop()) # type: ignore[attr-defined] - reactor.run() # type: ignore[attr-defined] + d.addBoth(lambda _: reactor.stop()) + reactor.run() diff --git a/tests/mockserver.py b/tests/mockserver.py index 185897373..eb1c03db7 100644 --- a/tests/mockserver.py +++ b/tests/mockserver.py @@ -375,9 +375,9 @@ if __name__ == "__main__": if args.type == "http": root = Root() factory = Site(root) - httpPort = reactor.listenTCP(0, factory) # type: ignore[attr-defined] + httpPort = reactor.listenTCP(0, factory) contextFactory = ssl_context_factory() - httpsPort = reactor.listenSSL(0, factory, contextFactory) # type: ignore[attr-defined] + httpsPort = reactor.listenSSL(0, factory, contextFactory) def print_listening(): httpHost = httpPort.getHost() @@ -391,11 +391,11 @@ if __name__ == "__main__": clients = [MockDNSResolver()] factory = DNSServerFactory(clients=clients) protocol = dns.DNSDatagramProtocol(controller=factory) - listener = reactor.listenUDP(0, protocol) # type: ignore[attr-defined] + listener = reactor.listenUDP(0, protocol) def print_listening(): host = listener.getHost() print(f"{host.host}:{host.port}") - reactor.callWhenRunning(print_listening) # type: ignore[attr-defined] - reactor.run() # type: ignore[attr-defined] + reactor.callWhenRunning(print_listening) + reactor.run() diff --git a/tests/test_engine.py b/tests/test_engine.py index 5fb87424b..7ddb420ba 100644 --- a/tests/test_engine.py +++ b/tests/test_engine.py @@ -557,4 +557,4 @@ class EngineTest(unittest.TestCase): if __name__ == "__main__": if len(sys.argv) > 1 and sys.argv[1] == "runserver": start_test_site(debug=True) - reactor.run() # type: ignore[attr-defined] + reactor.run() From 5fa0f64db5d4b7bf3f3a7c83578ce15ec534d3d3 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 27 Jan 2023 23:29:23 +0400 Subject: [PATCH 12/22] Ru typing on 3.8 as types-lxml isn't available for 3.7. --- .github/workflows/checks.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/checks.yml b/.github/workflows/checks.yml index 740092dab..bd26e8bb2 100644 --- a/.github/workflows/checks.yml +++ b/.github/workflows/checks.yml @@ -17,7 +17,7 @@ jobs: - python-version: "3.11" env: TOXENV: pylint - - python-version: 3.7 + - python-version: 3.8 env: TOXENV: typing - python-version: "3.11" # Keep in sync with .readthedocs.yml From 80453d53b19bbe288dc1e5b721f2f29acb89706a Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Sat, 28 Jan 2023 01:29:09 +0400 Subject: [PATCH 13/22] Type DeferredLists as Deferreds. --- scrapy/utils/defer.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/scrapy/utils/defer.py b/scrapy/utils/defer.py index 7f2211877..79b96856c 100644 --- a/scrapy/utils/defer.py +++ b/scrapy/utils/defer.py @@ -85,7 +85,7 @@ def mustbe_deferred(f: Callable, *args, **kw) -> Deferred: def parallel( iterable: Iterable, count: int, callable: Callable, *args, **named -) -> DeferredList: +) -> Deferred: """Execute a callable over the objects in the given iterable, in parallel, using no more than ``count`` concurrent calls. @@ -202,11 +202,11 @@ class _AsyncCooperatorAdapter(Iterator): def parallel_async( async_iterable: AsyncIterable, count: int, callable: Callable, *args, **named -) -> DeferredList: +) -> Deferred: """Like parallel but for async iterators""" coop = Cooperator() work = _AsyncCooperatorAdapter(async_iterable, callable, *args, **named) - dl = DeferredList([coop.coiterate(work) for _ in range(count)]) + dl: Deferred = DeferredList([coop.coiterate(work) for _ in range(count)]) return dl @@ -245,7 +245,7 @@ def process_parallel(callbacks: Iterable[Callable], input, *a, **kw) -> Deferred callbacks """ dfds = [defer.succeed(input).addCallback(x, *a, **kw) for x in callbacks] - d = DeferredList(dfds, fireOnOneErrback=True, consumeErrors=True) + d: Deferred = DeferredList(dfds, fireOnOneErrback=True, consumeErrors=True) d.addCallbacks(lambda r: [x[1] for x in r], lambda f: f.value.subFailure) return d From b1f33a68aceb0e0d599f006506ab7d74a23e95f3 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Sat, 11 Feb 2023 21:39:27 +0400 Subject: [PATCH 14/22] Bump mypy and types-*. --- tox.ini | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tox.ini b/tox.ini index 7048391ac..b1fd06b09 100644 --- a/tox.ini +++ b/tox.ini @@ -37,13 +37,13 @@ install_command = [testenv:typing] basepython = python3 deps = - mypy==0.991 + mypy==1.0.0 types-attrs==19.1.0 types-lxml==2022.11.8 - types-Pillow==9.4.0.5 - types-Pygments==2.14.0.1 - types-pyOpenSSL==21.0.0 - types-setuptools==57.0.0 + types-Pillow==9.4.0.10 + types-Pygments==2.14.0.2 + types-pyOpenSSL==23.0.0.2 + types-setuptools==67.2.0.1 commands = mypy --show-error-codes {posargs: scrapy tests} From fa9897282fc571bbac64eaba06433362ad233530 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Sat, 11 Feb 2023 22:11:30 +0400 Subject: [PATCH 15/22] Add a cast to deferred_from_coro, that code path has different typing. --- scrapy/utils/defer.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scrapy/utils/defer.py b/scrapy/utils/defer.py index a3bd273e3..d25ebbdf4 100644 --- a/scrapy/utils/defer.py +++ b/scrapy/utils/defer.py @@ -288,7 +288,7 @@ def deferred_from_coro(o) -> Any: if not is_asyncio_reactor_installed(): # wrapping the coroutine directly into a Deferred, this doesn't work correctly with coroutines # that use asyncio, e.g. "await asyncio.sleep(1)" - return ensureDeferred(o) + return ensureDeferred(cast(Coroutine[Deferred, Any, Any], o)) # wrapping the coroutine into a Future and then into a Deferred, this requires AsyncioSelectorReactor event_loop = _get_asyncio_event_loop() return Deferred.fromFuture(asyncio.ensure_future(o, loop=event_loop)) From b7daa2624d698a2fcf64923b4d1c51d859673006 Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Sat, 11 Feb 2023 23:46:33 +0400 Subject: [PATCH 16/22] Remove a workaround for old lxml. --- scrapy/http/request/form.py | 6 ------ 1 file changed, 6 deletions(-) diff --git a/scrapy/http/request/form.py b/scrapy/http/request/form.py index bdc6a3e39..342e0c7ad 100644 --- a/scrapy/http/request/form.py +++ b/scrapy/http/request/form.py @@ -198,12 +198,6 @@ def _select_value(ele: SelectElement, n: str, v: str): # And for select tags without options o = ele.value_options return (n, o[0]) if o else (None, None) - if v is not None and multiple: - # This is a workaround to bug in lxml fixed 2.3.1 - # fix https://github.com/lxml/lxml/commit/57f49eed82068a20da3db8f1b18ae00c1bab8b12#L1L1139 - selected_options = ele.xpath(".//option[@selected]") - values = [(o.get("value") or o.text or "").strip() for o in selected_options] - return n, values return n, v From a1717aa48c42fdeb9c45739a82b00b9cadf7b58c Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Sun, 12 Feb 2023 00:42:39 +0400 Subject: [PATCH 17/22] Fix most of FormRequest typing issues. --- scrapy/http/request/form.py | 33 +++++++++++++++++++++++---------- scrapy/utils/python.py | 4 ++-- 2 files changed, 25 insertions(+), 12 deletions(-) diff --git a/scrapy/http/request/form.py b/scrapy/http/request/form.py index 342e0c7ad..ef6589361 100644 --- a/scrapy/http/request/form.py +++ b/scrapy/http/request/form.py @@ -5,10 +5,17 @@ This module implements the FormRequest class which is a more convenient class See documentation in docs/topics/request-response.rst """ -from typing import Iterable, List, Optional, Tuple, Type, TypeVar, Union +from typing import Iterable, List, Optional, Tuple, Type, TypeVar, Union, cast from urllib.parse import urlencode, urljoin, urlsplit, urlunsplit -from lxml.html import FormElement, HtmlElement, HTMLParser, SelectElement +from lxml.html import ( + FormElement, + HTMLParser, + InputElement, + MultipleSelectOptions, + SelectElement, + TextareaElement, +) from parsel.selector import create_root_node from w3lib.html import strip_html5_whitespace @@ -19,7 +26,8 @@ from scrapy.utils.response import get_base_url FormRequestTypeVar = TypeVar("FormRequestTypeVar", bound="FormRequest") -FormdataType = Optional[Union[dict, List[Tuple[str, str]]]] +FormdataKVType = Tuple[str, Union[str, Iterable[str]]] +FormdataType = Optional[Union[dict, List[FormdataKVType]]] class FormRequest(Request): @@ -79,6 +87,7 @@ class FormRequest(Request): def _get_form_url(form: FormElement, url: Optional[str]) -> str: + assert form.base_url is not None # typing if url is None: action = form.get("action") if action is None: @@ -87,11 +96,11 @@ def _get_form_url(form: FormElement, url: Optional[str]) -> str: return urljoin(form.base_url, url) -def _urlencode(seq: Iterable, enc: str) -> str: +def _urlencode(seq: Iterable[FormdataKVType], enc: str) -> str: values = [ (to_bytes(k, enc), to_bytes(v, enc)) for k, vs in seq - for v in (vs if is_listlike(vs) else [vs]) + for v in (cast(Iterable[str], vs) if is_listlike(vs) else [cast(str, vs)]) ] return urlencode(values, doseq=True) @@ -147,7 +156,7 @@ def _get_inputs( formdata: FormdataType, dont_click: bool, clickdata: Optional[dict], -) -> List[Tuple[str, str]]: +) -> List[FormdataKVType]: """Return a list of key-value pairs for the inputs found in the given form.""" try: formdata_keys = dict(formdata or ()).keys() @@ -165,7 +174,7 @@ def _get_inputs( ' not(re:test(., "^(?:checkbox|radio)$", "i")))]]', namespaces={"re": "http://exslt.org/regular-expressions"}, ) - values = [ + values: List[FormdataKVType] = [ (k, "" if v is None else v) for k, v in (_value(e) for e in inputs) if k and k not in formdata_keys @@ -183,15 +192,19 @@ def _get_inputs( return values -def _value(ele: HtmlElement): +def _value( + ele: Union[InputElement, SelectElement, TextareaElement] +) -> Tuple[Optional[str], Union[None, str, MultipleSelectOptions]]: n = ele.name v = ele.value if ele.tag == "select": - return _select_value(ele, n, v) + return _select_value(cast(SelectElement, ele), n, v) return n, v -def _select_value(ele: SelectElement, n: str, v: str): +def _select_value( + ele: SelectElement, n: Optional[str], v: Union[None, str, MultipleSelectOptions] +) -> Tuple[Optional[str], Union[None, str, MultipleSelectOptions]]: multiple = ele.multiple if v is None and not multiple: # Match browser behaviour on simple select tag without options selected diff --git a/scrapy/utils/python.py b/scrapy/utils/python.py index 1771e5550..fc50e0f12 100644 --- a/scrapy/utils/python.py +++ b/scrapy/utils/python.py @@ -8,7 +8,7 @@ import sys import weakref from functools import partial, wraps from itertools import chain -from typing import AsyncGenerator, AsyncIterable, Iterable, Union +from typing import Any, AsyncGenerator, AsyncIterable, Iterable, Union from scrapy.utils.asyncgen import as_async_generator @@ -45,7 +45,7 @@ def iflatten(x): yield el -def is_listlike(x): +def is_listlike(x: Any) -> bool: """ >>> is_listlike("foo") False From 733309affa0c01239841899a790ec4734ea173dc Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 24 Feb 2023 00:05:59 +0400 Subject: [PATCH 18/22] Don't assume formnumber can be None. --- scrapy/http/request/form.py | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/scrapy/http/request/form.py b/scrapy/http/request/form.py index ef6589361..8da779b7c 100644 --- a/scrapy/http/request/form.py +++ b/scrapy/http/request/form.py @@ -58,7 +58,7 @@ class FormRequest(Request): response: TextResponse, formname: Optional[str] = None, formid: Optional[str] = None, - formnumber: Optional[int] = 0, + formnumber: int = 0, formdata: FormdataType = None, clickdata: Optional[dict] = None, dont_click: bool = False, @@ -109,7 +109,7 @@ def _get_form( response: TextResponse, formname: Optional[str], formid: Optional[str], - formnumber: Optional[int], + formnumber: int, formxpath: Optional[str], ) -> FormElement: """Find the wanted form element within the given response.""" @@ -142,13 +142,12 @@ def _get_form( raise ValueError(f"No
element found with {formxpath}") # If we get here, it means that either formname was None or invalid - if formnumber is not None: - try: - form = forms[formnumber] - except IndexError: - raise IndexError(f"Form number {formnumber} not found in {response}") - else: - return form + try: + form = forms[formnumber] + except IndexError: + raise IndexError(f"Form number {formnumber} not found in {response}") + else: + return form def _get_inputs( From 3a40c06ed9e4e7bf110f0eebc17a9ec77195bd1b Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 24 Feb 2023 00:16:49 +0400 Subject: [PATCH 19/22] Bump mypy and types-*. --- tox.ini | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/tox.ini b/tox.ini index b78b7c125..7ca26ec49 100644 --- a/tox.ini +++ b/tox.ini @@ -37,13 +37,13 @@ install_command = [testenv:typing] basepython = python3 deps = - mypy==1.0.0 + mypy==1.0.1 types-attrs==19.1.0 - types-lxml==2022.11.8 - types-Pillow==9.4.0.10 - types-Pygments==2.14.0.2 - types-pyOpenSSL==23.0.0.2 - types-setuptools==67.2.0.1 + types-lxml==2023.2.11 + types-Pillow==9.4.0.16 + types-Pygments==2.14.0.5 + types-pyOpenSSL==23.0.0.4 + types-setuptools==67.4.0.1 commands = mypy --show-error-codes {posargs: scrapy tests} From d7bf39ee78acdf6342fbbad3d4ba0fdb6d8c17de Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 24 Feb 2023 00:28:06 +0400 Subject: [PATCH 20/22] Fix a merge problem. --- tox.ini | 6 ------ 1 file changed, 6 deletions(-) diff --git a/tox.ini b/tox.ini index 7ca26ec49..5a9d9cf29 100644 --- a/tox.ini +++ b/tox.ini @@ -187,9 +187,3 @@ deps = {[docs]deps} setenv = {[docs]setenv} commands = sphinx-build -W -b linkcheck . {envtmpdir}/linkcheck - -[testenv:black] -deps = - black==22.12.0 -commands = - black {posargs:--check .} From d20f2788822eabc513a4c55a341cfe65b6bce37d Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 24 Feb 2023 00:28:24 +0400 Subject: [PATCH 21/22] Add a link to the w3lib typing fix. --- scrapy/utils/response.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scrapy/utils/response.py b/scrapy/utils/response.py index b26eeb0d4..794678c48 100644 --- a/scrapy/utils/response.py +++ b/scrapy/utils/response.py @@ -42,7 +42,7 @@ def get_meta_refresh( """Parse the http-equiv refresh parameter from the given response""" if response not in _metaref_cache: text = response.text[0:4096] - # a w3lib typing bug here + # a w3lib typing bug here, fixed in https://github.com/scrapy/w3lib/pull/211 _metaref_cache[response] = html.get_meta_refresh( # type: ignore[assignment] text, response.url, response.encoding, ignore_tags=ignore_tags ) From 3dd9d71c328fea0f4b2ab4bc668e65ecd1eb1d8e Mon Sep 17 00:00:00 2001 From: Andrey Rakhmatullin Date: Fri, 24 Feb 2023 01:13:48 +0400 Subject: [PATCH 22/22] Correctly wrap connection errors into ResponseFailed. --- scrapy/core/http2/agent.py | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/scrapy/core/http2/agent.py b/scrapy/core/http2/agent.py index 75731f525..215ea9716 100644 --- a/scrapy/core/http2/agent.py +++ b/scrapy/core/http2/agent.py @@ -6,7 +6,12 @@ from twisted.internet.base import ReactorBase from twisted.internet.defer import Deferred from twisted.internet.endpoints import HostnameEndpoint from twisted.python.failure import Failure -from twisted.web.client import URI, BrowserLikePolicyForHTTPS, _StandardEndpointFactory +from twisted.web.client import ( + URI, + BrowserLikePolicyForHTTPS, + ResponseFailed, + _StandardEndpointFactory, +) from twisted.web.error import SchemeNotSupported from scrapy.core.downloader.contextfactory import AcceptableProtocolsContextFactory @@ -83,8 +88,7 @@ class H2ConnectionPool: pending_requests = self._pending_requests.pop(key, None) while pending_requests: d = pending_requests.popleft() - # TODO: this is incorrect, errback takes a single exception - d.errback(errors) # type: ignore[arg-type] + d.errback(ResponseFailed(errors)) def close_connections(self) -> None: """Close all the HTTP/2 connections and remove them from pool