Commit Graph

162 Commits

Author SHA1 Message Date
Erik Dubbelboer 1d6a7e0493 The Authorization header should include the Basic keyword
Thanks to https://github.com/paween1980
2019-09-27 17:29:44 +02:00
Erik Dubbelboer 2edabf3b76 Add support for user:pass in URLs (#614)
Fixes #609
2019-08-18 11:23:33 +02:00
Marcelo Pires ccaae97f5b Support {readTimeout,maxBodySize,writeTimeout} per request based on the headers. (#598) 2019-07-12 14:42:07 +02:00
Vladimir Shteinman 74ad0f8e9b Export struct var and add specs 2019-05-22 23:23:02 +03:00
Vladimir Shteinman 9551d9544d Option for immediate header flush 2019-05-18 17:04:30 +03:00
Ciprian Dorin Craciun 733a6505a9 Support huge read-only []byte response bodies (#477)
* Add `Response.SetBodyRaw` method that serves a `[]byte` slice without touching it  (as an alternative to `SetBody`)
* Update various response related functions that are impacted after the incoduction of `Response.bodyRaw`
* Add a few test-cases in relation to `Response.SetBodyRaw`
2019-02-24 08:32:54 +00:00
xuecai 10b98c2cdf add conn's address info in Response (#537)
* reset commit
* fix response copy bug; add tests;
2019-02-16 10:50:14 +00:00
Erik Dubbelboer 65955d6208 Don't reset headers when reading the body fails
Not doing this means people using Client can still inspect the response
headers when reading the body fails (for example when it is too big).

fixes #456
2018-11-16 00:24:33 +08:00
Berezhnoy Pavel e771b6fe43 #457: allow to rewrite system error response (#458)
* #457: allow to rewrite system error response

* #457: review fixes

* #457: more review issues fixed
2018-11-13 15:41:42 +03:00
Gabriel Pérez S 996610f021 Schema changes detection with HostClient
Related with the issue https://github.com/valyala/fasthttp/issues/244
2018-10-14 16:19:09 +03:00
Tommy Chen d459e257bf Use form headers in WriteMultipartForm
CreateFormFile always set Content-Type header to application/octet-stream.
Use CreatePart instead so all headers in the multipart form can be used.
2018-10-05 19:02:16 +03:00
Erik Dubbelboer d4f0cf56d8 Remove fasthttp.ByteBuffer
As advertised in https://github.com/valyala/fasthttp/commit/b5f96d4b4120bb1e09c23ac32baf21a14da4a71d
2018-10-01 14:15:29 +08:00
Erik Dubbelboer c67f81e5f9 Use PostArgs in client
See: https://github.com/erikdubbelboer/fasthttp/issues/17
2018-09-20 15:22:18 +08:00
Erik Dubbelboer 26aa8e51e8 Allow whitespace after chunk size
There seems to be servers/load balancers that insert whitespaces
between the chunk-size number and \r\n.
2018-08-27 21:42:46 +00:00
Erik Dubbelboer cf6f6e7fca Read GET/HEAD body if it exists
Previously, GET/HEAD bodies were not read. The HTTP 1.1 specification
states:

"A server SHOULD read and forward a message-body on any request; if the
request method does not include defined semantics for an entity-body,
then the message-body SHOULD be ignored when handling the request.
I suspect this code is at fault."

This change reads the body on such request and continues the previous
behavior of returning a "Content-Length" of 0 to the application.

See: https://github.com/valyala/fasthttp/issues/159
2018-08-24 16:21:17 +03:00
xPushkin f24d00fcc6 A lot of typo fixes 2017-10-08 13:30:35 +01:00
Aliaksandr Valialkin 5116aa64e0 Mention that the body returned from Request.Body and Response.Body is valid until the Request/Response modification 2017-09-12 13:15:57 +03:00
Aliaksandr Valialkin 8948e047bf Issue #176: reset Content-Length when compression is enabled for streamed response body 2017-06-27 20:50:04 +03:00
Aliaksandr Valialkin b154429ca9 Compress responses only if their content-type starts with text/ or application/ 2017-05-17 17:10:29 +03:00
Aliaksandr Valialkin 103adc311e Do not compress responses with sizes smaller than 200 bytes
Compressing small bodies has little sense, since the compressed result size
may exceed the original body size.

This should save CPU time when the server responds with small responses.
2017-05-17 16:17:53 +03:00
Aliaksandr Valialkin 30e92af08f Limit heap memory usage when compressing high number of concurrent responses
Previously each concurrent compression could allocate huge compression state
with the size up to 1Mb each. So 10K concurrent connections could result in
10Gb of compression state in the heap.

This CL limits the number of compression states among concurrent requests
when {Append,Write}{Gzip,Deflate}* functions are called to O(GOMAXPROCS).
These functions are used by CompressHandler* for non-streaming responses,
i.e. it should cover the majority of use cases.

Memory usage for 10K concurrent connections that compress responses drops
from 10Gb to 200Mb after this CL.
2017-05-17 14:45:31 +03:00
Aliaksandr Valialkin b0de56d13b Properly set "https" scheme in RequestCtx.URI() for TLS connections 2017-01-15 00:26:21 +02:00
Aliaksandr Valialkin 4ee6517626 Issue #193: properly handle nil body when calling WriteGzip* and WriteDeflate* on Response struct 2016-10-31 13:41:50 +02:00
Aliaksandr Valialkin 3e1cb31dcd Added SwapBody to Request and Response for zero-copy body pass between Request and Response objects 2016-09-12 13:42:59 +03:00
Aliaksandr Valialkin 834fb48f10 Immediately write compressed response data to the connection after the user flushes the data from StreamWriter 2016-08-18 13:03:57 +03:00
Aliaksandr Valialkin 8e21bc7f60 do not compress response body again if Content-Encoding is set 2016-08-17 14:50:58 +03:00
Aliaksandr Valialkin d055141f64 Propagate 'https' scheme to request URI for TLS connections 2016-08-17 14:01:35 +03:00
Aliaksandr Valialkin 45915c69f6 Optimization: recycle request and response body buffers only if Server.ReduceMemoryUsage is set 2016-07-18 17:46:52 +03:00
Aliaksandr Valialkin 778f83ed89 Discourage using ReleaseBody, since now bytebufferpool should properly handle the case mentioned in #102 2016-06-27 18:26:56 +03:00
Aliaksandr Valialkin 00868f2226 Use separate pools for request and response bodies.
This should reduce potential memory waste related to byte buffers' pool usage.
2016-06-24 15:15:59 +03:00
Aliaksandr Valialkin 8ca66d3de8 Use github.com/valyala/bytebufferpool, which should protect against memory fragmentation in the pool of byte buffers like mentioned in the PR #102 2016-06-22 20:32:00 +03:00
Aliaksandr Valialkin 83a24c301b Issue #114: added SetHost and Host helper methods to Request 2016-06-15 20:32:54 +03:00
Aliaksandr Valialkin 8db7d05839 Close base response body streams when wrapping them into gzip or deflate stream writers 2016-06-10 20:19:15 +03:00
Aliaksandr Valialkin 90baa054bc Export IsBodyStream on RequestCtx, Request and Response
These methods may help writing proper request handler wrappers
for request handlers, which may set body either via SetBodyStream*
or via usual body methods.
2016-06-03 17:07:13 +03:00
Aliaksandr Valialkin 1fc1f4cbf5 Reduce memory usage under high load by pooling request/response bodies.
- Use request/response body pools.
- Defer request/response body allocation until it is really required.
- Return request/response bodies to the pool as soon as they become unused.

This minimizes the total amount of memory occupied by active request/response
bodies under high load.
2016-06-03 16:54:22 +03:00
Aliaksandr Valialkin bdd73c3261 small cleanup after 117f109c29 2016-05-27 16:11:29 +03:00
Jason Fesler 117f109c29 Add func (resp *Response) ReleaseBody(size int) (#102) 2016-05-24 15:49:08 +03:00
Aliaksandr Valialkin fa50251b46 Issue #100: accept POST query args with arbitrary charset set via Content-Type 2016-05-20 18:14:06 +03:00
Aliaksandr Valialkin c9c8d72607 Issue #78: immediately flush chunked data to client 2016-04-15 23:14:38 +03:00
Aliaksandr Valialkin 8280b7a162 Moved empty noCopy struct to the top of container structs. See @stemar94 's comment at https://github.com/golang/go/issues/12884 for details 2016-03-06 00:17:08 +02:00
Aliaksandr Valialkin 9f43aa1601 Do not expose noCopy.Lock 2016-03-04 21:20:52 +02:00
Aliaksandr Valialkin 9fa69c74af Embed noCopy struct into structs, which mustn't be copied
This should help `go vet` detecting invalid structs' copyings.
See https://github.com/golang/go/issues/8005#issuecomment-190753527 for details.
2016-03-04 16:57:24 +02:00
Aliaksandr Valialkin 8581513fa2 Issue #61: server: properly read multipart/form-data 2016-03-01 18:07:52 +02:00
Aliaksandr Valialkin 3546c31c2b Issue #60: skip body copying in DoTimeout. This should improve DoTimeout performance when dealing with big request and/or response bodies 2016-03-01 01:13:50 +02:00
Aliaksandr Valialkin 70316d0535 Issue #60: give up to find an optimal strategy for body buffers' re-use 2016-02-29 18:35:15 +02:00
Aliaksandr Valialkin 8985a565b2 Issue #60: throughput tuning: re-use body buffers if their size is smaller than 8Kb 2016-02-29 18:11:17 +02:00
Aliaksandr Valialkin fcfda9f312 Issue #60: increased client and server throughput when working with big bodies 2016-02-29 17:31:47 +02:00
Aliaksandr Valialkin c1437a71e6 Shuffled Request and Response members in order to reduce object sizes 2016-02-25 13:57:21 +02:00
Aliaksandr Valialkin f3d9e7ae9b golint fix 2016-02-25 13:30:25 +02:00
Sebastian Schepens 889c65bae6 Added BodyInflate and WriteInflate to match BodyGunzip and WriteGunzip 2016-02-22 14:36:58 -03:00