sanic/tests/test_keep_alive_timeout.py

317 lines
10 KiB
Python
Raw Normal View History

import asyncio
from asyncio import sleep as aio_sleep
from json import JSONDecodeError
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
from os import environ
2020-09-27 00:58:36 +01:00
import httpcore
import httpx
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
import pytest
from sanic_testing.testing import HOST, SanicTestClient
from sanic import Sanic, server
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
from sanic.compat import OS_IS_WINDOWS
from sanic.response import text
CONFIG_FOR_TESTS = {"KEEP_ALIVE_TIMEOUT": 2, "KEEP_ALIVE": True}
Fix Ctrl+C and tests on Windows. (#1808) * Fix Ctrl+C on Windows. * Disable testing of a function N/A on Windows. * Add test for coverage, avoid crash on missing _stopping. * Initialise StreamingHTTPResponse.protocol = None * Improved comments. * Reduce amount of data in test_request_stream to avoid failures on Windows. * The Windows test doesn't work on Windows :( * Use port numbers more likely to be free than 8000. * Disable the other signal tests on Windows as well. * Windows doesn't properly support SO_REUSEADDR, so that's disabled in Python, and thus rebinding fails. For successful testing, reuse port instead. * app.run argument handling: added server kwargs (alike create_server), added warning on extra kwargs, made auto_reload explicit argument. Another go at Windows tests * Revert "app.run argument handling: added server kwargs (alike create_server), added warning on extra kwargs, made auto_reload explicit argument. Another go at Windows tests" This reverts commit dc5d682448e3f6595bdca5cb764e5f26ca29e295. * Use random test server port on most tests. Should avoid port/addr reuse issues. * Another test to random port instead of 8000. * Fix deprecation warnings about missing name on Sanic() in tests. * Linter and typing * Increase test coverage * Rewrite test for ctrlc_windows_workaround * py36 compat * py36 compat * py36 compat * Don't rely on loop internals but add a stopping flag to app. * App may be restarted. * py36 compat * Linter * Add a constant for OS checking. Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com>
2020-03-26 04:42:46 +00:00
PORT = 42101 # test_keep_alive_timeout_reuse doesn't work with random port
2020-09-27 00:58:36 +01:00
class ReusableSanicConnectionPool(httpcore.AsyncConnectionPool):
last_reused_connection = None
2020-09-27 00:58:36 +01:00
async def _get_connection_from_pool(self, *args, **kwargs):
conn = await super()._get_connection_from_pool(*args, **kwargs)
self.__class__.last_reused_connection = conn
return conn
class ResusableSanicSession(httpx.AsyncClient):
def __init__(self, *args, **kwargs) -> None:
2020-09-27 00:58:36 +01:00
transport = ReusableSanicConnectionPool()
super().__init__(transport=transport, *args, **kwargs)
class ReuseableSanicTestClient(SanicTestClient):
2017-09-12 04:09:42 +01:00
def __init__(self, app, loop=None):
super().__init__(app)
2017-09-12 04:09:42 +01:00
if loop is None:
loop = asyncio.get_event_loop()
self._loop = loop
self._server = None
self._tcp_connector = None
self._session = None
def get_new_session(self):
return ResusableSanicSession()
2017-09-12 04:09:42 +01:00
# Copied from SanicTestClient, but with some changes to reuse the
# same loop for the same app.
def _sanic_endpoint_test(
2018-12-30 11:18:06 +00:00
self,
method="get",
uri="/",
gather_request=True,
debug=False,
server_kwargs=None,
2018-12-30 11:18:06 +00:00
*request_args,
**request_kwargs,
2018-12-30 11:18:06 +00:00
):
2017-09-12 04:09:42 +01:00
loop = self._loop
results = [None, None]
exceptions = []
server_kwargs = server_kwargs or {"return_asyncio_server": True}
if gather_request:
2018-12-30 11:18:06 +00:00
def _collect_request(request):
if results[0] is None:
results[0] = request
self.app.request_middleware.appendleft(_collect_request)
if uri.startswith(
("http:", "https:", "ftp:", "ftps://", "//", "ws:", "wss:")
):
url = uri
else:
uri = uri if uri.startswith("/") else f"/{uri}"
scheme = "http"
url = f"{scheme}://{HOST}:{PORT}{uri}"
2018-12-30 11:18:06 +00:00
@self.app.listener("after_server_start")
2017-09-12 04:09:42 +01:00
async def _collect_response(loop):
try:
response = await self._local_request(
method, url, *request_args, **request_kwargs
2018-12-30 11:18:06 +00:00
)
results[-1] = response
except Exception as e2:
exceptions.append(e2)
2017-09-12 04:09:42 +01:00
if self._server is not None:
_server = self._server
else:
2018-12-30 11:18:06 +00:00
_server_co = self.app.create_server(
host=HOST, debug=debug, port=PORT, **server_kwargs
)
2017-09-12 04:09:42 +01:00
server.trigger_events(
2018-12-30 11:18:06 +00:00
self.app.listeners["before_server_start"], loop
)
2017-09-12 04:09:42 +01:00
try:
loop._stopping = False
_server = loop.run_until_complete(_server_co)
except Exception as e1:
raise e1
self._server = _server
2018-12-30 11:18:06 +00:00
server.trigger_events(self.app.listeners["after_server_start"], loop)
self.app.listeners["after_server_start"].pop()
if exceptions:
raise ValueError(f"Exception during request: {exceptions}")
if gather_request:
2017-09-12 04:09:42 +01:00
self.app.request_middleware.pop()
try:
request, response = results
return request, response
except Exception:
raise ValueError(
f"Request and response object expected, got ({results})"
2018-12-30 11:18:06 +00:00
)
else:
try:
return results[-1]
except Exception:
raise ValueError(f"Request object expected, got ({results})")
def kill_server(self):
try:
if self._server:
self._server.close()
self._loop.run_until_complete(self._server.wait_closed())
self._server = None
if self._session:
self._loop.run_until_complete(self._session.aclose())
self._session = None
except Exception as e3:
raise e3
2017-09-12 04:09:42 +01:00
# Copied from SanicTestClient, but with some changes to reuse the
# same TCPConnection and the sane ClientSession more than once.
# Note, you cannot use the same session if you are in a _different_
# loop, so the changes above are required too.
async def _local_request(self, method, url, *args, **kwargs):
raw_cookies = kwargs.pop("raw_cookies", None)
2018-12-30 11:18:06 +00:00
request_keepalive = kwargs.pop(
"request_keepalive", CONFIG_FOR_TESTS["KEEP_ALIVE_TIMEOUT"]
2018-12-30 11:18:06 +00:00
)
if not self._session:
self._session = self.get_new_session()
try:
response = await getattr(self._session, method.lower())(
url, timeout=request_keepalive, *args, **kwargs
)
except NameError:
raise Exception(response.status_code)
try:
response.json = response.json()
except (JSONDecodeError, UnicodeDecodeError):
response.json = None
response.body = await response.aread()
response.status = response.status_code
response.content_type = response.headers.get("content-type")
if raw_cookies:
response.raw_cookies = {}
for cookie in response.cookies:
response.raw_cookies[cookie.name] = cookie
2017-09-12 04:09:42 +01:00
return response
2018-12-30 11:18:06 +00:00
keep_alive_timeout_app_reuse = Sanic("test_ka_timeout_reuse")
keep_alive_app_client_timeout = Sanic("test_ka_client_timeout")
keep_alive_app_server_timeout = Sanic("test_ka_server_timeout")
keep_alive_app_context = Sanic("keep_alive_app_context")
2017-09-12 04:09:42 +01:00
keep_alive_timeout_app_reuse.config.update(CONFIG_FOR_TESTS)
keep_alive_app_client_timeout.config.update(CONFIG_FOR_TESTS)
keep_alive_app_server_timeout.config.update(CONFIG_FOR_TESTS)
keep_alive_app_context.config.update(CONFIG_FOR_TESTS)
2017-09-12 04:09:42 +01:00
2018-12-30 11:18:06 +00:00
@keep_alive_timeout_app_reuse.route("/1")
2017-09-12 04:09:42 +01:00
async def handler1(request):
2018-12-30 11:18:06 +00:00
return text("OK")
2018-12-30 11:18:06 +00:00
@keep_alive_app_client_timeout.route("/1")
2017-09-12 04:09:42 +01:00
async def handler2(request):
2018-12-30 11:18:06 +00:00
return text("OK")
2018-12-30 11:18:06 +00:00
@keep_alive_app_server_timeout.route("/1")
2017-09-12 04:09:42 +01:00
async def handler3(request):
2018-12-30 11:18:06 +00:00
return text("OK")
2017-09-12 04:09:42 +01:00
@keep_alive_app_context.post("/ctx")
def set_ctx(request):
request.conn_info.ctx.foo = "hello"
return text("OK")
@keep_alive_app_context.get("/ctx")
def get_ctx(request):
return text(request.conn_info.ctx.foo)
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
@pytest.mark.skipif(
bool(environ.get("SANIC_NO_UVLOOP")) or OS_IS_WINDOWS,
reason="Not testable with current client",
)
2018-02-27 06:18:21 +00:00
def test_keep_alive_timeout_reuse():
"""If the server keep-alive timeout and client keep-alive timeout are
both longer than the delay, the client _and_ server will successfully
reuse the existing connection."""
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
client = ReuseableSanicTestClient(keep_alive_timeout_app_reuse, loop)
headers = {"Connection": "keep-alive"}
request, response = client.get("/1", headers=headers)
assert response.status == 200
assert response.text == "OK"
loop.run_until_complete(aio_sleep(1))
request, response = client.get("/1")
assert response.status == 200
assert response.text == "OK"
2020-09-27 00:58:36 +01:00
assert ReusableSanicConnectionPool.last_reused_connection
finally:
client.kill_server()
2018-02-27 06:18:21 +00:00
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
@pytest.mark.skipif(
bool(environ.get("SANIC_NO_UVLOOP")) or OS_IS_WINDOWS,
reason="Not testable with current client",
)
2018-02-27 06:18:21 +00:00
def test_keep_alive_client_timeout():
"""If the server keep-alive timeout is longer than the client
keep-alive timeout, client will try to create a new connection here."""
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
client = ReuseableSanicTestClient(keep_alive_app_client_timeout, loop)
headers = {"Connection": "keep-alive"}
_, response = client.get("/1", headers=headers, request_keepalive=1)
2020-09-27 00:58:36 +01:00
assert response.status == 200
assert response.text == "OK"
2020-09-27 00:58:36 +01:00
loop.run_until_complete(aio_sleep(2))
_, response = client.get("/1", request_keepalive=1)
2020-09-27 00:58:36 +01:00
assert ReusableSanicConnectionPool.last_reused_connection is None
finally:
client.kill_server()
2018-02-27 06:18:21 +00:00
Streaming Server (#1876) * Streaming request by async for. * Make all requests streaming and preload body for non-streaming handlers. * Cleanup of code and avoid mixing streaming responses. * Async http protocol loop. * Change of test: don't require early bad request error but only after CRLF-CRLF. * Add back streaming requests. * Rewritten request body parser. * Misc. cleanup, down to 4 failing tests. * All tests OK. * Entirely remove request body queue. * Let black f*ckup the layout * Better testing error messages on protocol errors. * Remove StreamBuffer tests because the type is about to be removed. * Remove tests using the deprecated get_headers function that can no longer be supported. Chunked mode is now autodetected, so do not put content-length header if chunked mode is preferred. * Major refactoring of HTTP protocol handling (new module http.py added), all requests made streaming. A few compatibility issues and a lot of cleanup to be done remain, 16 tests failing. * Terminate check_timeouts once connection_task finishes. * Code cleanup, 14 tests failing. * Much cleanup, 12 failing... * Even more cleanup and error checking, 8 failing tests. * Remove keep-alive header from responses. First of all, it should say timeout=<value> which wasn't the case with existing implementation, and secondly none of the other web servers I tried include this header. * Everything but CustomServer OK. * Linter * Disable custom protocol test * Remove unnecessary variables, optimise performance. * A test was missing that body_init/body_push/body_finish are never called. Rewritten using receive_body and case switching to make it fail if bypassed. * Minor fixes. * Remove unused code. * Py 3.8 check for deprecated loop argument. * Fix a middleware cancellation handling test with py38. * Linter 'n fixes * Typing * Stricter handling of request header size * More specific error messages on Payload Too Large. * Init http.response = None * Messages further tuned. * Always try to consume request body, plus minor cleanup. * Add a missing check in case of close_if_idle on a dead connection. * Avoid error messages on PayloadTooLarge. * Add test for new API. * json takes str, not bytes * Default to no maximum request size for streaming handlers. * Fix chunked mode crash. * Header values should be strictly ASCII but both UTF-8 and Latin-1 exist. Use UTF-8B to cope with all. * Refactoring and cleanup. * Unify response header processing of ASGI and asyncio modes. * Avoid special handling of StreamingHTTPResponse. * 35 % speedup in HTTP/1.1 response formatting (not so much overall effect). * Duplicate set-cookie headers were being produced. * Cleanup processed_headers some more. * Linting * Import ordering * Response middleware ran by async request.respond(). * Need to check if transport is closing to avoid getting stuck in sending loops after peer has disconnected. * Middleware and error handling refactoring. * Linter * Fix tracking of HTTP stage when writing to transport fails. * Add clarifying comment * Add a check for request body functions and a test for NotImplementedError. * Linter and typing * These must be tuples + hack mypy warnings away. * New streaming test and minor fixes. * Constant receive buffer size. * 256 KiB send and receive buffers. * Revert "256 KiB send and receive buffers." This reverts commit abc1e3edb21a5e6925fa4c856657559608a8d65b. * app.handle_exception already sends the response. * Improved handling of errors during request. * An odd hack to avoid an httpx limitation that causes test failures. * Limit request header size to 8 KiB at most. * Remove unnecessary use of format string. * Cleanup tests * Remove artifact * Fix type checking * Mark test for skipping * Cleanup some edge cases * Add ignore_body flag to safe methods * Add unit tests for timeout logic * Add unit tests for timeout logic * Fix Mock usage in timeout test * Change logging test to only logger in handler * Windows py3.8 logging issue with current testing client * Add test_header_size_exceeded * Resolve merge conflicts * Add request middleware to hard exception handling * Add request middleware to hard exception handling * Request middleware on exception handlers * Linting * Cleanup deprecations Co-authored-by: L. Kärkkäinen <tronic@users.noreply.github.com> Co-authored-by: Adam Hopkins <admhpkns@gmail.com>
2021-01-10 22:45:36 +00:00
@pytest.mark.skipif(
bool(environ.get("SANIC_NO_UVLOOP")) or OS_IS_WINDOWS,
reason="Not testable with current client",
)
2018-02-27 06:18:21 +00:00
def test_keep_alive_server_timeout():
"""If the client keep-alive timeout is longer than the server
keep-alive timeout, the client will either a 'Connection reset' error
_or_ a new connection. Depending on how the event-loop handles the
broken server connection."""
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
client = ReuseableSanicTestClient(keep_alive_app_server_timeout, loop)
headers = {"Connection": "keep-alive"}
_, response = client.get("/1", headers=headers, request_keepalive=60)
2020-09-27 00:58:36 +01:00
assert response.status == 200
assert response.text == "OK"
2020-09-27 00:58:36 +01:00
loop.run_until_complete(aio_sleep(3))
_, response = client.get("/1", request_keepalive=60)
2020-09-27 00:58:36 +01:00
assert ReusableSanicConnectionPool.last_reused_connection is None
finally:
client.kill_server()
@pytest.mark.skipif(
bool(environ.get("SANIC_NO_UVLOOP")) or OS_IS_WINDOWS,
reason="Not testable with current client",
)
def test_keep_alive_connection_context():
try:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
client = ReuseableSanicTestClient(keep_alive_app_context, loop)
headers = {"Connection": "keep-alive"}
request1, _ = client.post("/ctx", headers=headers)
loop.run_until_complete(aio_sleep(1))
request2, response = client.get("/ctx")
assert response.text == "hello"
assert id(request1.conn_info.ctx) == id(request2.conn_info.ctx)
assert (
request1.conn_info.ctx.foo == request2.conn_info.ctx.foo == "hello"
)
finally:
client.kill_server()