cowboy.git - Small, fast, modern HTTP server for Erlang/OTP.

Age	Commit message (Collapse)	Author
2017-10-21	Fix sending of final chunk in HTTP/1.1	Loïc Hoguin
	I broke this when fixing stream handlers earlier.
2017-10-21	Add a test for early_error metrics	Loïc Hoguin

2017-10-21	Add a metrics test when the handler didn't send anything	Loïc Hoguin
	Also fix a test group to use h2 instead of HTTP/1.1.
2017-10-21	Fix stream handler state being discarded on terminate	Loïc Hoguin
	When we have to send a response before terminating a stream, we call info. The state returned by this info call was discarded when we called terminate after that. This commit fixes it. There are no tests for this, however the new metrics test in the next commit requires the correct behavior so this is ultimately covered.
2017-10-20	Add a test for metrics with a request body	Loïc Hoguin

2017-10-20	Add experimental metrics stream handler	Loïc Hoguin
	It collects metrics and passes them to a configurable callback once the stream terminates. It will be documented in a future release. More tests incoming.
2017-10-20	Fix HTTP/1.1 stopping streams too early	Loïc Hoguin
	It is possible in some cases to move on to the next request without waiting, but that can be done as an optimization later on if necessary.
2017-10-20	Make the getting started use 2.0.0	Loïc Hoguin

2017-10-13	Add OTP-20.1.2 to CI	Loïc Hoguin

2017-10-13	Small guide fixes	Loïc Hoguin

2017-10-03	Make a few link use https2.0.0	Loïc Hoguin

2017-10-03	Update version, CI and Cowlib dependency version	Loïc Hoguin
	Hello 2.0.0!
2017-10-03	Make sure a path parsing error in the router gives a 400	Loïc Hoguin

2017-10-02	Document cow_ws:frame in Cowboy for now	Loïc Hoguin

2017-10-02	Fix cowboy:set_env when the env value is missing	Loïc Hoguin

2017-10-02	Fix infinite loop on incomplete multipart body	Jeffrey Griffin
	I have amended a lot of changes from the original commit to make it behave as expected, including returning a 400 error. LH
2017-10-02	Fix the documentation for the command flow	Loïc Hoguin

2017-09-27	Clarify a multipart example	Loïc Hoguin

2017-09-27	Cowboy 2.0.0-rc.42.0.0-rc.4	Loïc Hoguin

2017-09-27	Fix error response when constraint validation fails	Loïc Hoguin

2017-09-27	Add OTP-20.1 to CI	Loïc Hoguin

2017-09-27	Ensure the behavior on stream handler crash is consistent	Loïc Hoguin
	Also corrects the lack of error response when HTTP/1.1 is used.
2017-09-25	Cleanup various comments	Loïc Hoguin

2017-09-25	Avoid some crashes when HTTP/2 streams flush their response	Loïc Hoguin

2017-09-25	Move body length count to cowboy_stream_h instead of protocols	Loïc Hoguin
	The documentation was correct, the code was not. This should make it easier to implement new protocols. Note that for HTTP/2 we will need to add some form of counting later on to check for malformed requests, but we can do simpler and just reduce from the expected length and then check if that's 0 when IsFin=fin.
2017-09-21	Centralize stream handler error reporting in cowboy_stream	Loïc Hoguin

2017-09-15	Cowboy 2.0.0-rc.32.0.0-rc.3	Loïc Hoguin

2017-09-15	Add OTP 20.0.5 to CI	Loïc Hoguin

2017-09-15	Add two new RFCs, remove one obsolete	Loïc Hoguin

2017-09-14	Fix propagating the stacktrace on errors for OTP 19	Loïc Hoguin

2017-09-14	Rework the proc_lib_hack	Loïc Hoguin
	It is completely removed for connection processes, because assuming Cowboy is written properly this should bring us nothing anymore in 2.0. It is reworked for request processes, there we want to always propagate the stacktrace (including for exits) because we will print a report to help with debugging and proc_lib doesn't propagate it for exits. At the same time the initial callback for connection and request processes has been changed to connection_process and request_process, which should help with identifying processes when inspecting.
2017-09-14	Improve how we detect request errors	Loïc Hoguin
	When the request process exits with a {request_error, Reason, Human} exit reason, Cowboy will return a 400 status code instead of 500. Cowboy may also return a more specific status code depending on the error. Currently it may also return 408 or 413. This should prove to be more solid that looking inside the stack trace.
2017-09-05	Update naming in REST flowcharts	Loïc Hoguin

2017-09-05	Add tests for direct Req access	Loïc Hoguin

2017-09-05	Remove transfer-encoding parsing from cowboy_req	Loïc Hoguin
	The header never reaches this point.
2017-09-05	Remove an unused variable	Loïc Hoguin

2017-09-04	Accept sendfile tuple with 0 length in cowboy_req	Loïc Hoguin
	This will result in no data being sent. It's simply easier to do this than to have to handle 0 size cases in user code.
2017-09-04	Add a static handler test with an empty file	Loïc Hoguin

2017-09-04	Alphabetical order some tests	Loïc Hoguin

2017-09-04	Fix more documentation todos	Loïc Hoguin
	I have decided not to include a manual page for cowboy_stream_h at this point because it clashes with the cowboy_stream manual page. This decision will be revisited in the future.
2017-08-29	Add OTP 20.0.4 to CI	Loïc Hoguin

2017-08-23	Cowboy 2.0.0-rc.22.0.0-rc.2	Loïc Hoguin

2017-08-21	Tentative fix for out of order queued data	Loïc Hoguin

2017-08-21	Properly clean up timers when terminating	Loïc Hoguin

2017-08-14	Add tests for the streams shutdown mechanism	Loïc Hoguin

2017-08-09	Remove broken links from examples README	Loïc Hoguin

2017-08-08	Add missing cowboy_children module	Loïc Hoguin
	My bad! Still new at this.
2017-08-08	Add OTP-20.0.2 to CI	Loïc Hoguin

2017-08-08	Implement the shutdown timeout for request processes	Loïc Hoguin
	This should work very similar to normal supervisors, in particular during the shutdown sequence when the connection process goes down or switches to Websocket. Processes that need to enforce the shutdown timeout will be required to trap exits, just like in a supervisor. In a vanilla Cowboy, this only matters at connection shutdown, as Cowboy will otherwise wait for the request process to be down before stopping the stream. Tests are currently missing.
2017-07-26	Fix HTTP/2 server push	Loïc Hoguin
	Cowboy was encoding the headers then decoding them when initializing the request. The problem is that the encoding and decoding contexts are not the same. Now, Cowboy will directly use the headers it received in the push command for the new request. This is also more efficient. I am surprised it worked at all considering the issue.