Previously http_SUITE:echo_body/1 and http_SUITE:check_raw_status/1
returned large values. Common test would then try to write these to the
logs causing tests to take a long time and use lots of memory.
Instead it will always go through content_types_accepted and it is
up to the resource code to do any creation and to return the created
path if the method is POST and the client should be redirected to the
created resource's location.
This removes the meta value 'put_path' as it is not needed anymore.
This fixes an issue with PATCH where content types were not normalized.
For get_type_provided:
'*' will be match any parameters of media-range in "accept" header.
If '*' matched, then '*' is replaced by the matching parameters.
If Accept header is missing and '*' using, then in media_type in parameters
will be '*' and reply content-type will be without any parameters.
For content_types_accepted:
'*' will be match any parameters in "content-type" header.
We now read from the socket to be able to detect errors or TCP close
events, and buffer the data if any. Once the data receive goes over
a certain limit, which defaults to 5000 bytes, we simply close the
connection with an {error, overflow} reason.
Ultimately few things change, it's mostly just a nicer syntax and
slightly different expectations. The name of the value `dispatch`
did not change, because the previous dispatch values will now fail
if the code is not updated to using `cowboy_router:compile/1`.
No constraints have been implemented in this commit.
This behavior can be enabled with the `compress` protocol option.
See the `compress_response` example for more details.
All tests are now ran with and without compression for both HTTP
and HTTPS.
This changes the behavior of the `timeout` protocol option to
mean "Time in which the full request line and headers must be
received". The default of 5s should be fine for all normal uses.
This change has no noticeable impact on performance and is thus
enabled by default for everyone. It can be disabled by setting
`timeout` to `infinity` although that is definitely not encouraged.
Inspired by the contribution from @naryl on github.
Middlewares allow customizing the request processing.
All existing Cowboy project are incompatible with this commit.
You need to change `{dispatch, Dispatch}` in the protocol options
to `{env, [{dispatch, Dispatch}]}` to fix your code.
* #state{} changes are avoided where possible
* #state{} is now smaller and use less memory
* the Req object is created only after the whole request is parsed
* parsing makes use of a single binary match context
* external calls are avoided in the critical path
* URL fragment is now extracted properly (retrieval API next commit)
* argument orders to local functions modified to avoid extra operations
* dispatching waits as long as possible before tokenizing host/path
* handler opts are no longer shown in the error messages except in init
The code may not look as beautiful as it was before. But it really
is, for parsing code. The parsing section of the file may be skipped
if your eyes start to burn.
Header names are now binaries. Since header names are case insensitive
they are all converted to lowercase. For example: <<"content-length">>.
The max_line_length option was removed. Three new options have been
added instead:
* max_request_line_length (defaults to 4096)
* max_header_name_length (defaults to 64)
* max_header_value_length (defaults to 4096)
First step in making all methods and header names binaries to
get rid of many inconsistencies caused by decode_packet/3.
Methods are all binary now. Note that since they are case
sensitive, the usual methods become <<"GET">>, <<"POST">> and so on.
