aboutsummaryrefslogtreecommitdiffstats
path: root/src
AgeCommit message (Collapse)Author
2012-01-26Handle delete better when no delete_resource function is implementedTom Burdick
2012-01-23Fix a case where request body wouldn't get cleaned up on keepaliveLoïc Hoguin
The body was still in the buffer that's being used for the next request and was thus used as a request, causing errors.
2012-01-23Rename the type http_status/0 to cowboy_http:status/0Loïc Hoguin
2012-01-23Rename the type http_header/0 to cowboy_http:header/0Loïc Hoguin
At the same time rename http_headers/0 to cowboy_http:headers/0.
2012-01-23Rename the type http_version/0 to cowboy_http:version/0Loïc Hoguin
2012-01-23Rename the type http_uri/0 to cowboy_http:uri/0Loïc Hoguin
2012-01-23Rename the type http_method/0 to cowboy_http:method/0Loïc Hoguin
Exported types are much better than include files.
2012-01-23Fix typespecs for calendar:datetime/0 in cowboy_http_staticLoïc Hoguin
2012-01-23Add {halt, Req, State} return value to most REST callbacksLoïc Hoguin
Excluding generate_etag, last_modified, expires and variances.
2012-01-23Rename a lot of HandlerState2 variables to HandlerStateLoïc Hoguin
2012-01-23Improve the return value for cowboy_http_rest:upgrade/4Loïc Hoguin
2012-01-23Merge branch 'multipart' of https://github.com/nox/cowboyLoïc Hoguin
Conflicts: src/cowboy_http_req.erl test/http_SUITE.erl
2012-01-23Add multipart supportAnthony Ramine
2012-01-12Welcome to new developments with 0.5.0Loïc Hoguin
2012-01-09Update version to 0.4.00.4.0Loïc Hoguin
2012-01-09Make REST handlers' put_resource accept true or false return valuesLoïc Hoguin
Followup to 0bb23f2400ed0b65834913c8522a978d986f1f92. As discussed in #119.
2012-01-07Add etag option to cowboy_http_static handler.Magnus Klaar
2012-01-06Make REST handlers' process_post accept true or false return valuesLoïc Hoguin
They should return true when it has been processed successfully, or false otherwise, in which case a 500 error is sent. Fixes #119.
2012-01-06Fix handler crashes handlingLoïc Hoguin
We try to send a 500 error only if we didn't send the response headers yet. If they were, then we have no way to be sure the response was fully sent, nor should we assume anything about how this will be handled client-side, so we do nothing more and in both cases close the connection.
2012-01-06Move the ensure_response call before ensure_body_processedLoïc Hoguin
We want to reply as soon as possible.
2012-01-05Add test and doc for content types function.Magnus Klaar
2012-01-04inet_address:parse/1 does not accept binaries as an input, only stringsYurii Rashkovskii
2012-01-03Merge branch 'static-handler-split' of https://github.com/klaar/cowboyLoïc Hoguin
2012-01-03Add note on async threads to cowboy_http_static.Magnus Klaar
2011-12-28Add built-in cowboy_http_static handler.Magnus Klaar
2011-12-28Add cowboy_http_req:set_resp_body_fun/3 to restMagnus Klaar
2011-12-28Add cowboy_http_req:set_resp_body_fun/3.Magnus Klaar
2011-12-28Add cowboy_http_req:transport/1 function.Magnus Klaar
2011-12-26Use calendar date and time types exported since R14B04Loïc Hoguin
Removes the cowboy_clock:date/0, time/0 and datetime/0 exported types.
2011-12-22Add 'Upgrade' header parsingLoïc Hoguin
Not parsing product tokens, I don't see much point.
2011-12-22Remove outdated @todo notes and update anotherLoïc Hoguin
2011-12-22Add a max_keepalive HTTP protocol optionLoïc Hoguin
Based on the patch by Louis-Philippe Gauthier.
2011-12-22Websocket control frames payload length must be lower than 126 bytesLoïc Hoguin
2011-12-19create_path now returns a path, not an URILoïc Hoguin
Also add the new 'put_path' metadata information to the request before entering the put_resource step (which proceeds to call content_types_accepted and then the chosen content type's callback, which is expected to use the 'put_path' value as the path to the new resource instead of the original request path, because it may have been changed by the create_path call.
2011-12-19Add meta/2 and meta/3 to cowboy_http_req to save useful protocol informationLoïc Hoguin
* cowboy_http_protocol now defines 'websocket_version' as metadata. * cowboy_http_rest now defines 'media_type', 'language', 'charset' as metadata.
2011-12-19Add missing p_headers and cookies to compacted valuesLoïc Hoguin
2011-12-16update type of resp_body to iodataMagnus Klaar
resp_body is set by cowboy_http_req:set_resp_body/2 which accepts the input type iodata(). Update the resp_body field to use the same type. Also update the cowboy_http_req:has_resp_body/2 function to use the iolist_size/1 which works with both binary() and iolist() values instead of the byte_size/1 function which only works with binary() values.
2011-12-15Fix a dialyzer warningLoïc Hoguin
2011-12-15Reply status 400 if we receive an unexpected value or error for headersLoïc Hoguin
2011-12-15Add cowboy:accept_ack/1 for a cleaner handling of the shoot messageLoïc Hoguin
2011-12-12Merge branch 'variances-fix' of github.com:klaar/cowboyLoïc Hoguin
2011-12-12Fixes match_media_type in cowboy_http_restLoïc Hoguin
Based on the patch from Heinz N. Gies. The main change is that his patch accept values such as */plain which we do not want. Fixes issue #105.
2011-12-12Merge branch 'hipe-compilation-fix' of github.com:klaar/cowboyLoïc Hoguin
2011-12-12Merge branch 'master' of github.com:talentdeficit/cowboyLoïc Hoguin
2011-12-12adds webmachine format content type handlingalisdair sullivan
2011-12-11update cowboy_http_rest:variances/2Magnus Klaar
Fix pattern in case statement that was intended to strip away the first comma separating the values in the variance header. Update generation of variance list to use more idiomatic erlang. Pattern match on list structure over using erlang:length/1 to compute length.
2011-12-10update /bits in binary expressions to /binaryMagnus Klaar
This is a workaround for an issue in the hipe compiler where compilation of code matching on binaries using patterns of the form Data = <<Char, _Ignore/bits>> fails because _Ignore is expected to be used. Using /binary instead of /bits resolves the issue. We're applying this change to all binary expressions in cowboy_http because all functions in this module process human readable text formats which are always expected to be bytestrings. Two uses of /bits has been changed to /binary in cowboy_http_websocket in two clauses of a single function, websocket_data/4. This is safe to change because the data is later passed to binary:match/2 which will always fail with a badarg error if the input is a bitstring instead of a bytestring.
2011-12-09add cowboy_http:urlencode/2 and urlencode/1Magnus Klaar
This function complements the cowboy_http:urldecode/2 and urldecode/1 functions. We should have this encoding covered now. urldecode is only expected to be invoked by user code, no further changes needed to integrate this.
2011-12-08Fix the process_post/2 function in cowboy_http_restLoïc Hoguin
2011-12-08Add more documentation to the REST protocol handlerLoïc Hoguin