drakma-devel

drakma-devel@common-lisp.net

182 discussions

[drakma-devel] Patch - Accept-Encoding: gzip
by Red Daly 05 Feb '10

05 Feb '10

Dear drakma users, I made drakma accept gzipped message bodies by default, and also automatically decode them. A patch is attached. Unfortunately this adds a new dependency on gzip-sequence (which in turn depends on salza2). See the docstrings for more details. To sum it up: if WANT-STREAM is T, It wraps the returned stream in a flexi-stream that wraps a gzip-input-stream. Otherwise, it decodes the sequence before returning from HTTP-REQUEST. Thanks, Red

2 3

[drakma-devel] Drakma doesn't work on CLISP because of unsupported TCP_NODELAY
by Anton Vodonosov 22 Jan '10

22 Jan '10

Hello, The recent Drakma (1.1.0) doesn't work on CLISP because of :NODELAY T passed to USOCKET:SOCKET-CONNECT. :NODELAY is not supported on CLISP and usocket signals an error. IMHO it would be better to remove the :NODELAY T, not only because of the CLISP support, but also because the Nagle's algorithm disabled by :NODELAY T seems to be a feature rather than a problem. Best regards, - Anton

1 0

[drakma-devel] post requests and content-type
by Adam Dodek Michalik 08 Jan '10

08 Jan '10

Hey, I noticed that when using POST method, Drakma completely ignores content-type provided by user. That's kinda bad for me, since I'd like to be able to send the content's charset as well - so that Content-Type header looks for instance like this: application/x-www-form-urlencoded; charset=utf-8 Modifying drakma so that it uses content-type provided by user is trivial, but maybe there are other reasons behind this behaviour I'm not aware of. -- Adam Michalik vel Dodek Dodecki <dodek(a)dodecki.net>

1 0

[drakma-devel] Drakma - problem with http-request on SBCL 1.0.34
by Fred Gibson 08 Jan '10

08 Jan '10

I've found a small problem in Drakma's http-request where in a call to usocket:socket-connect it sent the key :nodelay which doesn't exist for sbcl's socket connect version, so I changed that call in drakma to this: (usocket:socket-stream (usocket:socket-connect host port :element-type 'octet #+:openmcl :deadline #+:openmcl deadline #-:sbcl :nodelay #-:sbcl t)) Then it seemed to work just fine. My best, Fred Gibson Founder / Software Developer http://www.streamfocus.com (c)2010 Organon Technologies LLC

3 5

[drakma-devel] Several problems
by Евгений Дмитренко 03 Dec '09

03 Dec '09

In the official manual http://weitz.de/drakma/#connection-timeoutconnection-timeout key parameter exists. But my SBCL interpreter says: debugger invoked on a SB-INT:SIMPLE-PROGRAM-ERROR: unknown &KEY argument: :CONNECTION-TIMEOUT Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL. restarts (invokable by number or by possibly-abbreviated name): 0: [ABORT] Exit debugger, returning to top level. (DRAKMA:HTTP-REQUEST "http://bcheck.scanit.be/bcheck/")[:EXTERNAL] 0] when I try to call function: (drakma:http-request "http://bcheck.scanit.be/bcheck/" :method :get :content-length t :connection-timeout 10) Please help me to solve this problem. Best regards, Evgen Dmitrenko

4 4

[drakma-devel] New releases (1.1.0) of Drakma and Chunga...
by Edi Weitz 01 Dec '09

01 Dec '09

...can be found at the usual location. If you update one of the libraries, please update both. What's new? http://weitz.de/drakma/CHANGELOG.txt http://weitz.de/chunga/CHANGELOG.txt Thanks, Edi.

1 0

[drakma-devel] redirect in http-request
by Xiangjun Wu 01 Dec '09

01 Dec '09

Hi, All, I'm building a oauth library with drakma, oauth specification need every request is signed with its token and token secret in additional-headers, but it also need a unique timestamp and nonce(a random string). It is OK when there is not redirect, but http-request will use the same additional-headers in the redirect request, so oauth server throw an error: "Invalid nonce/timestamp: This nonce has been used ..." I think there are two solutions, one is there is a option that http-request do not automatically follow redirect, the other is http-reqeust evaluate the additional-headers again when redirecting. Is it acceptable, thank you in advance. Plato Wu 片云天共远永夜月同孤

2 6

[drakma-devel] GET :PARAMETERS added again during redirects
by Eugene Ossintsev 12 Nov '09

12 Nov '09

I use DRAKMA 1.0, Clozure CL 1.4-dev (Ubuntu Linux x86). According to the DRAKMA documentation, parameters is an alist of name/value pairs (the car and the cdr each being a string) which denotes the parameters which are added to the query part of the URI or (in the case of a POST request) comprise the request body. In my example QUOTES1 sends a GET request passing the request parameters in the URI, and QUOTES2 uses the :PARAMETERS instead. The results, unfortunately, are not the same. (defun quotes1 () (multiple-value-bind (body code) (drakma:http-request "http://finance.yahoo.com/d/quotes.csv?s=XIC.TO&f=sxl1d1t1") (when (= code 200) (split-sequence:split-sequence #\, body)))) ("\"XIC.TO\"" "\"Toronto\"" "18.03" "\"10/9/2009\"" "\"3:59pm\" ^M ") <=== It's fine :-) (defun quotes2 () (multiple-value-bind (body code) (drakma:http-request "http://finance.yahoo.com/d/quotes.csv" :method :get :parameters '(("s" . "XIC.TO") ("f" . "sxl1d1t1"))) (when (= code 200) (split-sequence:split-sequence #\, body)))) ("\"XIC.TO\"" "\"Toronto\"" "18.03" "\"10/9/2009\"" "\"3:59pm\"" "\"XIC.TO\"" "\"XIC.TO\"" "\"XIC.TO\"" "\"Toronto\"" "18.03" "\"10/9/2009\"" "\"3:59pm\" ^M \"XIC.TO\"" "\"Toronto\"" "18.03" "\"10/9/2009\"" "\"3:59pm\"" "\"XIC.TO\"" "\"XIC.TO\"" "\"XIC.TO\"" "\"Toronto\"" "18.03" "\"10/9/2009\"" "\"3:59pm\" ^M ") <=== It's a mess :-( TRACE DRAKMA:HTTP-REQUEST shows that in case of redirects of GET requests the parameters from :PARAMETERS are added again to the URI. Thus, the same parameters may be added again and again, giving unpredictable results. 0> Calling (DRAKMA:HTTP-REQUEST "http://finance.yahoo.com/d/quotes.csv" :PARAMETERS (("s" . "XIC.TO") ("f" . "sxl1d1t1"))) 1> Calling (DRAKMA:HTTP-REQUEST #<URI http://download.finance.yahoo.com/d/quotes.csv?s=XIC.TO&f=sxl1d1t1> :REDIRECT 4 :STREAM NIL :ADDITIONAL-HEADERS NIL :PARAMETERS (("s" . "XIC.TO") ("f" . "sxl1d1t1"))) <1 DRAKMA:HTTP-REQUEST returned 7 values : <1 #(34 88 73 67 46 84 79 34 44 34 84 111 114 111 110 116 111 34 44 49 55 46 57 56 44 34 49 48 .......... <1 200 <1 ((:DATE . "Wed, 14 Oct 2009 05:00:51 GMT") (:P3P . "policyref=\"http://p3p.yahoo.com/w3c/p3p.xml\", CP=\"CAO DSP COR CUR ADM DEV TAI PSA PSD IVAi IVDi CONi TELo OTPi OUR DELi SAMi OTRi UNRi PUBi IND PHY ONL UNI PUR FIN COM NAV INT DEM CNT STA POL HEA PRE GOV\"") (:CACHE-CONTROL . "private") (:CONNECTION . "close") (:TRANSFER-ENCODING . "chunked") (:CONTENT-TYPE . "application/octet-stream")) <1 #<URI http://download.finance.yahoo.com/d/quotes.csv?s=XIC.TO&f=sxl1d1t1&s=XIC.TO…> <1 #<FLEXI-STREAMS:FLEXI-IO-STREAM #x15616B76> <1 T <1 "OK" Is it what may be expected from such usage of :PARAMETERS in GET requests? If not, is it worth fixing it? - Eugene

2 2

[drakma-devel] [patch] timeouts, take two
by Stanislaw Halik 19 Oct '09

19 Oct '09

Hey Edi, I made a new patch that uses TRIVIAL-TIMEOUT instead of usocket WAIT-FOR-INPUT. Results are great, it yields no slowdown at all when downloading a 10MB zero-filled file. Patch follows. I can't attach files due to posting to Gmane through tin(1) so it's included in the message body. If it gets mangled in the transmission, it's alternatively available at http://tehran.lain.pl/stuff/20090827-drakma-timeout-gray-stream-take2.diff diff -rN -u old-drakma/drakma.asd new-drakma/drakma.asd --- old-drakma/drakma.asd 2009-08-27 05:59:20.000000000 +0200 +++ new-drakma/drakma.asd 2009-08-27 05:59:20.000000000 +0200 @@ -53,10 +53,13 @@ (:file "util") (:file "read") (:file "cookies") + #-lispworks (:file "timeouts") (:file "request")) :depends-on (:puri :cl-base64 :chunga :flexi-streams + #-lispworks #:trivial-timeout + #-lispworks #:trivial-gray-streams #-:lispworks :usocket #-(or :lispworks :allegro) :cl+ssl)) diff -rN -u old-drakma/request.lisp new-drakma/request.lisp --- old-drakma/request.lisp 2009-08-27 05:59:20.000000000 +0200 +++ new-drakma/request.lisp 2009-08-27 05:59:20.000000000 +0200 @@ -200,9 +200,11 @@ force-binary want-stream stream - #+:lispworks (connection-timeout 20) - #+:lispworks (read-timeout 20) - #+(and :lispworks (not :lw-does-not-have-write-timeout)) + (connection-timeout 20) + (read-timeout 20) + #+(or (not :lispworks) + (and :lispworks + (not :lw-does-not-have-write-timeout))) (write-timeout 20 write-timeout-provided-p) #+openmcl deadline) @@ -385,6 +387,7 @@ time units. If the server fails to respond until that time, a COMMUNICATION-DEADLINE-EXPIRED condition is signalled. DEADLINE is only available on CCL 1.2 and later." + #-lispworks (declare (ignore write-timeout-provided-p)) (unless (member protocol '(:http/1.0 :http/1.1) :test #'eq) (parameter-error "Don't know how to handle protocol ~S." protocol)) (setq uri (cond ((uri-p uri) (copy-uri uri)) @@ -450,12 +453,16 @@ :write-timeout write-timeout :errorp t) #-:lispworks - (usocket:socket-stream - (usocket:socket-connect host port - :element-type 'octet - #+openmcl #+openmcl - :deadline deadline - :nodelay t)))) + (let ((sock (trivial-timeout:with-timeout (connection-timeout) + (usocket:socket-connect host port + :element-type 'octet + #+openmcl #+openmcl + :deadline deadline + :nodelay t)))) + (if (or read-timeout write-timeout) + (usocket-timeout:timeout-stream-for-socket + sock :read-timeout read-timeout :write-timeout write-timeout) + (usocket:socket-stream sock))))) #+openmcl (when deadline ;; it is correct to set the deadline here even though diff -rN -u old-drakma/timeouts.lisp new-drakma/timeouts.lisp --- old-drakma/timeouts.lisp 1970-01-01 01:00:00.000000000 +0100 +++ new-drakma/timeouts.lisp 2009-08-27 05:59:20.000000000 +0200 @@ -0,0 +1,62 @@ +(defpackage #:usocket-timeout + (:use #:cl #:usocket #:trivial-timeout #:trivial-gray-streams) + (:shadowing-import-from #:trivial-timeout #:timeout-error) + (:export #:timeout-stream-for-socket)) + +(in-package #:usocket-timeout) + + +(defclass timeout-mixin () + ((socket-of :initarg :socket + :reader socket-of))) + +(defclass timeout-input-stream (trivial-gray-stream-mixin + fundamental-binary-input-stream + timeout-mixin) + ((read-timeout :initform nil + :initarg :read-timeout + :reader read-timeout-of))) + +(defmethod stream-read-sequence ((stream timeout-input-stream) + sequence start end &key) + (with-timeout ((read-timeout-of stream)) + (read-sequence sequence (socket-of stream) :start start :end end))) + +(defmethod stream-read-byte ((stream timeout-input-stream)) + (with-timeout ((read-timeout-of stream)) + (read-byte (socket-of stream)))) + + +(defclass timeout-output-stream (trivial-gray-stream-mixin + fundamental-binary-output-stream + timeout-mixin) + ((write-timeout :initform nil + :initarg :write-timeout + :reader write-timeout-of))) + +(defmethod stream-finish-output ((stream timeout-output-stream)) + (finish-output (socket-of stream))) + +(defmethod stream-write-sequence ((stream timeout-output-stream) + sequence start end &key) + (with-timeout ((write-timeout-of stream)) + (write-sequence sequence (socket-of stream) + :start (or start 0) + :end end))) + +(defmethod stream-write-byte ((stream timeout-output-stream) integer) + (with-timeout ((write-timeout-of stream)) + (write-byte integer (socket-of stream)))) + + +(defclass usocket-timeout-stream (timeout-input-stream timeout-output-stream) + ()) + +(defun timeout-stream-for-socket (socket &key timeout + (read-timeout timeout) + (write-timeout timeout)) + (make-instance 'usocket-timeout-stream + :socket (socket-stream socket) + :read-timeout read-timeout + :write-timeout write-timeout)) +

2 2

[drakma-devel] NTLM auth headers?
by Carlos Konstanski 14 Oct '09

14 Oct '09

Is there any way to get drakma to write an NTLM auth header? I am unfortunately trying to communicate with a .NET web service in an Active Directory environment. If not, I suppose I could use Ã`additional-headers' to write one myself. Carlos Konstanski

2 1

← Newer
1
2
3
4
5
6
7
8
9
...
19
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

drakma-devel