From neuss at math.uni-karlsruhe.de Mon Oct 1 08:47:30 2007 From: neuss at math.uni-karlsruhe.de (Nicolas Neuss) Date: Mon, 01 Oct 2007 10:47:30 +0200 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> (Ram Krishnan's message of "Sat\, 29 Sep 2007 07\:56\:41 -0700") References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> Message-ID: <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> "Ram Krishnan" writes: > Ah, you're absolutely right. > > Simply binding using LET, instead of WITH-OPEN-STREAM would do the trick then. > > Thanks. > > -ram OK, thanks to everyone who responded (and especially to Edi for Hunchentoot and all the other nice software). I have now used something like the following: (define-easy-handler (csv-export-page :uri "/csv-export") ((csv :parameter-type 'boolean) ...) (when csv (setf (content-type) "text/comma-separated-values" (header-out "Last-Modified") (rfc-1123-date (get-universal-time)) (content-length) nil) (let ((out (send-headers))) (csv-output :stream out ...))) (with-output-to-string (out) (fill-and-print-template "csv-export.html" values :stream out))) As much as I can see, this works. Yours, Nicolas From neuss at math.uni-karlsruhe.de Mon Oct 1 09:13:06 2007 From: neuss at math.uni-karlsruhe.de (Nicolas Neuss) Date: Mon, 01 Oct 2007 11:13:06 +0200 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> (Nicolas Neuss's message of "Mon\, 01 Oct 2007 10\:47\:30 +0200") References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> Message-ID: <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> Nicolas Neuss writes: > (define-easy-handler (csv-export-page :uri "/csv-export") > ((csv :parameter-type 'boolean) ...) > (when csv > (setf (content-type) "text/comma-separated-values" > (header-out "Last-Modified") (rfc-1123-date (get-universal-time)) > (content-length) nil) > (let ((out (send-headers))) > (csv-output :stream out ...))) > (with-output-to-string (out) > (fill-and-print-template "csv-export.html" values :stream out))) > > As much as I can see, this works. In the code, ... denote some further parameters which can be configured via the csv-export page. Nicolas From varuzza at gmail.com Mon Oct 1 12:13:47 2007 From: varuzza at gmail.com (Leonardo Varuzza) Date: Mon, 1 Oct 2007 09:13:47 -0300 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> Message-ID: <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> Hello everbody, Coincidently playing with hunchentoot yesterday I encountered the same problem with streams. Based in the code of HANDLE-STATIC-FILE a made a path with 2 functions and one macro: make-http-output-stream witch send the appropriate headers and return a stream. This function deal with content-type and content-disposition headers. with-http-output-stream Create an http-output-stream, run the code and call finish-output handle-stream Copy the stream to the http-output I hope to contribute to the hunchetoot development. Att. Leonardo Varuzza. USP - Brazil. PS: A interesting use of this patch is to send images created by Zach Brene's Vecto library. I just send another patch to him to enable saving the result png to a stream. On 10/1/07, Nicolas Neuss wrote: > Nicolas Neuss writes: > > > (define-easy-handler (csv-export-page :uri "/csv-export") > > ((csv :parameter-type 'boolean) ...) > > (when csv > > (setf (content-type) "text/comma-separated-values" > > (header-out "Last-Modified") (rfc-1123-date (get-universal-time)) > > (content-length) nil) > > (let ((out (send-headers))) > > (csv-output :stream out ...))) > > (with-output-to-string (out) > > (fill-and-print-template "csv-export.html" values :stream out))) > > > > As much as I can see, this works. > > In the code, ... denote some further parameters which can be configured via > the csv-export page. > > Nicolas > _______________________________________________ > tbnl-devel site list > tbnl-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/tbnl-devel > -- The most fundamental particles in this product are held together by a "gluing" force about which little is currently known and whose adhesive power can therefore not be permanently guaranteed. -------------- next part -------------- A non-text attachment was scrubbed... Name: with-http-output-stream.patch Type: text/x-patch Size: 2744 bytes Desc: not available URL: From edi at agharta.de Tue Oct 2 06:14:20 2007 From: edi at agharta.de (Edi Weitz) Date: Tue, 02 Oct 2007 08:14:20 +0200 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> (Leonardo Varuzza's message of "Mon, 1 Oct 2007 09:13:47 -0300") References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> Message-ID: On Mon, 1 Oct 2007 09:13:47 -0300, "Leonardo Varuzza" wrote: > Coincidently playing with hunchentoot yesterday I encountered the > same problem with streams. Based in the code of HANDLE-STATIC-FILE a > made a path with 2 functions and one macro: > > make-http-output-stream witch send the appropriate headers and > return a stream. This function deal with content-type and > content-disposition headers. > > with-http-output-stream Create an http-output-stream, run the > code and call finish-output > > handle-stream Copy the stream to the http-output > > I hope to contribute to the hunchetoot development. Thanks for the patch, but I hope you don't mind if I think that this is in the realm of things that can and should be done by Hunchentoot users and not be provided by the library. Cheers, Edi. From nedenis at gmail.com Tue Oct 2 17:27:14 2007 From: nedenis at gmail.com (Denis Mashkevich) Date: Tue, 2 Oct 2007 19:27:14 +0200 Subject: [hunchentoot-devel] Problem with flexi-stream-bound on openmcl In-Reply-To: References: <2B7F55EC-DBCF-4E1F-8BC7-C671BF1BB5FF@gmail.com> Message-ID: On Sep 30, 2007, at 8:45 PM, Edi Weitz wrote: > Hmm, did you notice that it thinks the package is HUNCHENTOOT and not > FLEXI-STREAMS? My /guess/ is that you didn't recompile Hunchentoot > after upgrading FLEXI-STREAMS. Could you try again by deleting /all/ > FASL files (Hunchentoot and dependencies) and starting from scratch? That probably was the case, although I was pretty sure I removed all dfsls before restarting openmcl and reloading the system. > If this doesn't resolve the issue, let me know. (Right now, I don't > have a means to test on OpenMCL, but I just bought a new laptop with a > Core 2 Duo processor, so I should have it installed in a few days.) Now I have another problem. In a clean image, I load hunchentoot and start a server. For a while, requests are served - sometimes a few seconds, sometimes a minute or more. Then this is signaled: (this is under slime) Stream # is private to # [Condition of type SIMPLE-ERROR] Restarts: 0: [ABORT-BREAK] Reset this process 1: [ABORT] Kill this process Backtrace: 0: (# 0 20) 1: (SWANK:BACKTRACE 1 41) 2: (CCL::CALL-CHECK-REGS 'SWANK:BACKTRACE) 3: (# '(SWANK:BACKTRACE 1 41) 48 'NIL) 4: (# # #) 5: (FUNCALL 'SWANK::EVAL-FOR-EMACS) 6: (SWANK::SLDB-LOOP #) 7: (SWANK::CALL-WITH-BINDINGS '((*PRINT-PRETTY* . T) (*PRINT- LEVEL* . 4) (*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*) (*PRINT-PPRINT-DISPATCH* . #S(CCL::PPRINT-DISPATCH-TABLE :CONSES-WITH-CARS # :STRUCTURES NIL :OTHERS (# #))) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*) (*PRINT-ARRAY* . T) ...) #) 8: (# #) 9: (SWANK::DEBUG-IN-EMACS #) 10: (# # #) 11: (SWANK::CALL-WITH-REDIRECTED-IO # #) 12: (SWANK::CALL-WITH-CONNECTION # #) 13: (INVOKE-DEBUGGER #) 14: (SIGNAL #) 15: (CCL::%ERROR '(# #) -66477372) 16: (# # #(91 0 0 0 0 0 0 0 0 0 ...) 0 1) 17: (WRITE-SEQUENCE #(91) #) 18: (# # "[~A~@[ [~A] ~]] ~A~%" 0 1) 19: (WRITE-STRING "[~A~@[ [~A]~]] ~A~%" #) 20: (CCL::SUB-FORMAT # 0 19) 21: (CCL::DO-SUB-FORMAT #) 22: (FORMAT # "[~A~@[ [~A]~]] ~A~%") 23: (# ':INFO "Default handler called for script ~A") 24: (HUNCHENTOOT::DEFAULT-HANDLER) 25: (HUNCHENTOOT::PROCESS-REQUEST '((:HOST . "localhost:3003") (:USER-AGENT . "Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv: 1.8.1.7) Gecko/20070914 Firefox/2.0.0.7") (:ACCEPT . "text/xml,application/xml,application/xhtml+xml,text/ html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5") (:ACCEPT-LANGUAGE . "en-us,en;q=0.5") (:ACCEPT-ENCODING . "gzip,deflate") (:ACCEPT-CHARSET . "UTF-8,*") (:KEEP-ALIVE . "300") (:CONNECTION . "keep-alive") (:CACHE-CONTROL . "max-age=0")) # ':GET "/" ':HTTP/1.1) 26: (HUNCHENTOOT::PROCESS-CONNECTION # #) 27: (CCL::RUN-PROCESS-INITIAL-FORM '(#) #) 28: (# '(#) 0) 29: (# 788896 #) Without slime, it takes longer, and the thread state is different (waiting for lock, not exhausted). > Debug: Stream # is private to # > While executing: SIGNAL, in process hunchentoot-worker-2(4). ;;; ;;; # requires access to Shared Terminal Input ;;; Do you have any ideas about this? From edi at agharta.de Tue Oct 2 22:07:54 2007 From: edi at agharta.de (Edi Weitz) Date: Wed, 03 Oct 2007 00:07:54 +0200 Subject: [hunchentoot-devel] Problem with flexi-stream-bound on openmcl In-Reply-To: (Denis Mashkevich's message of "Tue, 2 Oct 2007 19:27:14 +0200") References: <2B7F55EC-DBCF-4E1F-8BC7-C671BF1BB5FF@gmail.com> Message-ID: On Tue, 2 Oct 2007 19:27:14 +0200, Denis Mashkevich wrote: > Now I have another problem. In a clean image, I load hunchentoot and > start a server. For a while, requests are served - sometimes a few > seconds, sometimes a minute or more. Then this is signaled: > > (this is under slime) > > Stream # ISO-8859-1) #x929C1E6> is private to # (11) [Exhausted] #x929ADB6> > [Condition of type SIMPLE-ERROR] You might want to search the mailing list archives, specifically May and maybe June 2007, for example this one: http://common-lisp.net/pipermail/tbnl-devel/2007-May/001319.html The upshot is that up until now I couldn't test on OpenMCL myself. My new laptop can run it, but I haven't yet found the time to install it. You'll either have to wait or you'll have to dig into it yourself. From eadmund42 at gmail.com Thu Oct 4 16:17:31 2007 From: eadmund42 at gmail.com (Robert Uhl) Date: Thu, 04 Oct 2007 10:17:31 -0600 Subject: [hunchentoot-devel] Publishing stream data? References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> Message-ID: Edi Weitz writes: > >> Coincidently playing with hunchentoot yesterday I encountered the >> same problem with streams. Based in the code of HANDLE-STATIC-FILE a >> made a path with 2 functions and one macro: >> >> make-http-output-stream witch send the appropriate headers and >> return a stream. This function deal with content-type and >> content-disposition headers. >> >> with-http-output-stream Create an http-output-stream, run the >> code and call finish-output >> >> handle-stream Copy the stream to the http-output >> >> I hope to contribute to the hunchetoot development. > > Thanks for the patch, but I hope you don't mind if I think that this > is in the realm of things that can and should be done by Hunchentoot > users and not be provided by the library. Perhaps there should be a HUNCHENTOOT-CONTRIB which contains such things, and if they become standard/useful enough then they might be merged into Hunchentoot proper? -- Robert Uhl Whoa, there. Those are some strong words for somebody who doesn't even own a machine gun. --Milkman Dan From edi at agharta.de Thu Oct 4 19:26:00 2007 From: edi at agharta.de (Edi Weitz) Date: Thu, 04 Oct 2007 21:26:00 +0200 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: (Robert Uhl's message of "Thu, 04 Oct 2007 10:17:31 -0600") References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> Message-ID: On Thu, 04 Oct 2007 10:17:31 -0600, Robert Uhl wrote: > Perhaps there should be a HUNCHENTOOT-CONTRIB which contains such > things, and if they become standard/useful enough then they might be > merged into Hunchentoot proper? In TBNL we had a contrib repository included with the distribution, but I deemed it too much work (for me) to maintain it together with the main source. We now have this http://common-lisp.net/websvn/listing.php?repname=tbnl&path=%2Fcontrib%2Flsp%2F which currently contains code by Mac Chan. If someone else wants access to this repository for add-ons for Hunchentoot, let me know. Cheers, Edi. From ch-tbnl at bobobeach.com Thu Oct 4 20:30:18 2007 From: ch-tbnl at bobobeach.com (Cyrus Harmon) Date: Thu, 4 Oct 2007 13:30:18 -0700 Subject: [hunchentoot-devel] Publishing stream data? In-Reply-To: References: <87odfmah2u.fsf@ma-patru.mathematik.uni-karlsruhe.de> <6e7f31bf0709280940m43ee4436s72b265b5f8080f6f@mail.gmail.com> <6e7f31bf0709290756w57c0b365vd59a5e4fea70fc91@mail.gmail.com> <87ejgf9q71.fsf@ma-patru.mathematik.uni-karlsruhe.de> <873awv9p0d.fsf@ma-patru.mathematik.uni-karlsruhe.de> <5d9c2e640710010513o2760c4b5gab2639e3ebd3a236@mail.gmail.com> Message-ID: <54048368-09A0-4270-B4DD-B9023E7D9650@bobobeach.com> On a related note, in the absence of a hucnhentoot-contrib, I'll take a moment to plug some of my own hunchentoot-using packages. Two are general purpose libraries that could, in theory, be used by any hunchentoot application, hunchentoot-vhost for using a single hunchentoot server to serve multiple "hosts", and hunchentoot-auth for managing users. The other package is nuclblog, which is a relatively simple blog that runs on top of hunchentoot. They can be found at: http://cyrusharmon.org/projects Comments/suggestions/patches-to-implement-all-the-neat-features-I- never-got-around-to-implementing gladly accepted. In particular, I'd be curious as to how others handle user authentication and what not. It starts off trivial, but, at least from my experience, got to be a bit more complicated than I would have thought at the beginning. Things like redirecting non-authorized users, html pages for logging in vs. browser authentication, password salts, persistent storage for users, an interface for adding new users, etc... all make this a harder problem than one might think at first glance. Perhaps it's just "trivial" and everyone should roll their own solution here, but, comments on the hunchentoot-auth approach would be greatly appreciated. Thanks, Cyrus On Oct 4, 2007, at 12:26 PM, Edi Weitz wrote: > On Thu, 04 Oct 2007 10:17:31 -0600, Robert Uhl > wrote: > >> Perhaps there should be a HUNCHENTOOT-CONTRIB which contains such >> things, and if they become standard/useful enough then they might be >> merged into Hunchentoot proper? > > In TBNL we had a contrib repository included with the distribution, > but I deemed it too much work (for me) to maintain it together with > the main source. We now have this > > http://common-lisp.net/websvn/listing.php?repname=tbnl&path=% > 2Fcontrib%2Flsp%2F > > which currently contains code by Mac Chan. If someone else wants > access to this repository for add-ons for Hunchentoot, let me know. > > Cheers, > Edi. > _______________________________________________ > tbnl-devel site list > tbnl-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/tbnl-devel From wouter at csl.sony.fr Fri Oct 5 15:50:59 2007 From: wouter at csl.sony.fr (Wouter Van den Broeck) Date: Fri, 5 Oct 2007 17:50:59 +0200 Subject: [hunchentoot-devel] Full-duplex socket connections in Hunchentoot Message-ID: <168C6C3E-D445-4664-869A-903263A1D2CD@csl.sony.fr> Hello, I'm considering using Hunchentoot as a server for client applications that connect with the server over full-duplex http socket connections (such as Flash applications using XMLSocket). The general idea is that the server can send messages to the client at its own initiative and not only as a reply to a request. Having looked at the documentation and some aspects of the code, I gather that there is no such functionality by default, but that it should not be too difficult to integrate in Hunchentoot. If possible could you please roughly outline what you would consider to be a 'good' approach to attain this objective such that it would nicely fit in the existing design philosophy? Thanks in advance regards Wouter Van den Broeck From nablaone at gmail.com Fri Oct 5 16:32:11 2007 From: nablaone at gmail.com (=?UTF-8?Q?Rafa=C5=82_Strzali=C5=84ski?=) Date: Fri, 5 Oct 2007 18:32:11 +0200 Subject: [hunchentoot-devel] Full-duplex socket connections in Hunchentoot In-Reply-To: <168C6C3E-D445-4664-869A-903263A1D2CD@csl.sony.fr> References: <168C6C3E-D445-4664-869A-903263A1D2CD@csl.sony.fr> Message-ID: On 10/5/07, Wouter Van den Broeck wrote: > Hello, > > I'm considering using Hunchentoot as a server for client applications > that connect with the server over full-duplex http socket connections > (such as Flash applications using XMLSocket). The general idea is > that the server can send messages to the client at its own initiative > and not only as a reply to a request. > > Having looked at the documentation and some aspects of the code, I > gather that there is no such functionality by default, but that it > should not be too difficult to integrate in Hunchentoot. > > If possible could you please roughly outline what you would consider > to be a 'good' approach to attain this objective such that it would > nicely fit in the existing design philosophy? According to this documentation: http://www.adobe.com/support/flash/action_scripts/actionscript_dictionary/actionscript_dictionary860.htm XMLSocket is not a HTTP based protocol. So instead using Hunchentoot , you should consider writing your own socket server. XMLSocket protocol is quite simple, it should be hard. -- Best regards, Rafal Strzalinski (nabla) http://nablaone.net From austin at pettomato.com Fri Oct 5 19:08:15 2007 From: austin at pettomato.com (Austin Haas) Date: Fri, 5 Oct 2007 15:08:15 -0400 Subject: [hunchentoot-devel] Full-duplex socket connections in Hunchentoot In-Reply-To: References: <168C6C3E-D445-4664-869A-903263A1D2CD@csl.sony.fr> Message-ID: <20071005190815.GA11270@bean.chicago> Hi, I use both Hunchentoot regularly and a custom Lisp server for communicating with Flash via the XMLSocket. They really don't have very much in common, and I can confirm that what Rafal said is true; it is pretty simple to write the XMLSocket server from scratch. My most basic one is about 150 lines of code, and it was one of my first Lisp projects. -austin -- Austin Haas Pet Tomato, Inc. http://pettomato.com On Fri Oct 05 18:32 , Rafa? Strzali?ski wrote: > On 10/5/07, Wouter Van den Broeck wrote: > > Hello, > > > > I'm considering using Hunchentoot as a server for client applications > > that connect with the server over full-duplex http socket connections > > (such as Flash applications using XMLSocket). The general idea is > > that the server can send messages to the client at its own initiative > > and not only as a reply to a request. > > > > Having looked at the documentation and some aspects of the code, I > > gather that there is no such functionality by default, but that it > > should not be too difficult to integrate in Hunchentoot. > > > > If possible could you please roughly outline what you would consider > > to be a 'good' approach to attain this objective such that it would > > nicely fit in the existing design philosophy? > > According to this documentation: > http://www.adobe.com/support/flash/action_scripts/actionscript_dictionary/actionscript_dictionary860.htm > XMLSocket is not a HTTP based protocol. So instead using Hunchentoot , > you should consider writing your own socket server. XMLSocket protocol > is quite simple, it should be hard. > > > > -- > Best regards, > Rafal Strzalinski (nabla) > http://nablaone.net > _______________________________________________ > tbnl-devel site list > tbnl-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/tbnl-devel > From ch-tbnl at bobobeach.com Sat Oct 6 17:19:05 2007 From: ch-tbnl at bobobeach.com (Cyrus Harmon) Date: Sat, 6 Oct 2007 10:19:05 -0700 Subject: [hunchentoot-devel] get-gid-from-name support for SBCL Message-ID: <8E603415-893E-40FE-ABD8-5F336418D945@bobobeach.com> The following patch implements get-gid-from-name for SBCL. Unfortunately, it requires SBCL 1.0.10.31 or later for sb-posix support for getgrnam. Cyrus --- ../ORIGINAL/hunchentoot-0.14.2/unix-sbcl.lisp 2007-01-01 15:50:30.000000000 -0800 +++ unix-sbcl.lisp 2007-10-04 17:16:01.000000000 -0700 @@ -45,5 +45,4 @@ (defun get-gid-from-name (name) "Returns the GID for the group named NAME." - (declare (ignore name)) - (error "GET-GID-FROM-NAME not yet implemented for SBCL. Please send patches...")) + (sb-posix:group-gid (sb-posix:getgrnam name))) From edi at agharta.de Sat Oct 6 22:54:19 2007 From: edi at agharta.de (Edi Weitz) Date: Sun, 07 Oct 2007 00:54:19 +0200 Subject: [hunchentoot-devel] New release 0.14.3 (Was: get-gid-from-name support for SBCL) In-Reply-To: <8E603415-893E-40FE-ABD8-5F336418D945@bobobeach.com> (Cyrus Harmon's message of "Sat, 6 Oct 2007 10:19:05 -0700") References: <8E603415-893E-40FE-ABD8-5F336418D945@bobobeach.com> Message-ID: On Sat, 6 Oct 2007 10:19:05 -0700, Cyrus Harmon wrote: > The following patch implements get-gid-from-name for SBCL. > Unfortunately, it requires SBCL 1.0.10.31 or later for sb-posix > support for getgrnam. Thanks. That it's in the new release now. From sergio.garcia at gmail.com Sun Oct 7 17:10:46 2007 From: sergio.garcia at gmail.com (Sergio Garcia) Date: Sun, 7 Oct 2007 19:10:46 +0200 Subject: [hunchentoot-devel] shared sessions Message-ID: <34ec25750710071010g14f3819ei187a062217be00ed@mail.gmail.com> Hello everyone, I have a hunchentoot instance serving several applications, where each application is accesed using a different subdomain (i.e. app1.my-domain.net, app2.my-domain.net...etc). Is it possible to make sessions global to all the applications? Thanks for any hint, Sergio -------------- next part -------------- An HTML attachment was scrubbed... URL: From edi at agharta.de Sun Oct 7 17:31:36 2007 From: edi at agharta.de (Edi Weitz) Date: Sun, 07 Oct 2007 19:31:36 +0200 Subject: [hunchentoot-devel] shared sessions In-Reply-To: <34ec25750710071010g14f3819ei187a062217be00ed@mail.gmail.com> (Sergio Garcia's message of "Sun, 7 Oct 2007 19:10:46 +0200") References: <34ec25750710071010g14f3819ei187a062217be00ed@mail.gmail.com> Message-ID: On Sun, 7 Oct 2007 19:10:46 +0200, "Sergio Garcia" wrote: > I have a hunchentoot instance serving several applications, where > each application is accesed using a different subdomain > (i.e. app1.my-domain.net, app2.my-domain.net...etc). Is it possible > to make sessions global to all the applications? Not without patching Hunchentoot. For session support using only cookies you'd have to provide the corresponding "domain" parameter to the session cookie, and for cookie-less session support you'd have to modify the optional TEST-FN argument for REWRITE-URLS. From edi at agharta.de Fri Oct 19 23:54:11 2007 From: edi at agharta.de (Edi Weitz) Date: Sat, 20 Oct 2007 01:54:11 +0200 Subject: [hunchentoot-devel] New release 0.14.4 Message-ID: ChangeLog: Version 0.14.4 2007-10-20 Made log stream shared on OpenMCL (thanks to Gary Byers) Download: http://weitz.de/files/hunchentoot.tar.gz From edi at agharta.de Sat Oct 20 00:05:28 2007 From: edi at agharta.de (Edi Weitz) Date: Sat, 20 Oct 2007 02:05:28 +0200 Subject: [hunchentoot-devel] Hunchentoot not working on OpenMCL 1.1pre In-Reply-To: (Tiarnan O'Corrain's message of "Sun, 20 May 2007 12:32:53 +0100") References: Message-ID: On Sun, 20 May 2007 12:32:53 +0100, "Tiarnan O'Corrain" wrote: > Stream # ISO-8859-1) #x93BC97E> is private to # hunchentoot-worker-1(48) [Exhausted] #x93BBC1E> > [Condition of type SIMPLE-ERROR] This should hopefully be fixed with release 0.14.4 - finally... Please try. Cheers, Edi. From edi at agharta.de Sat Oct 20 00:07:58 2007 From: edi at agharta.de (Edi Weitz) Date: Sat, 20 Oct 2007 02:07:58 +0200 Subject: [hunchentoot-devel] Problem with flexi-stream-bound on openmcl In-Reply-To: (Denis Mashkevich's message of "Tue, 2 Oct 2007 19:27:14 +0200") References: <2B7F55EC-DBCF-4E1F-8BC7-C671BF1BB5FF@gmail.com> Message-ID: On Tue, 2 Oct 2007 19:27:14 +0200, Denis Mashkevich wrote: > Stream # ISO-8859-1) #x929C1E6> is private to # (11) [Exhausted] #x929ADB6> > [Condition of type SIMPLE-ERROR] See my recent reply to Tiarnan O'Corrain's email - I hope this is fixed with the 0.14.4 release. It'd be nice if OpenMCL users could try this out. For an explanation see here: http://clozure.com/pipermail/openmcl-devel/2007-October/003953.html Edi. From david at lichteblau.com Sun Oct 21 18:59:40 2007 From: david at lichteblau.com (David Lichteblau) Date: Sun, 21 Oct 2007 20:59:40 +0200 Subject: [hunchentoot-devel] cl-webdav patch for current cxml Message-ID: <20071021185940.GF10680@radon> Hi, the attached patch updates cl-webdav's use of SAX for the current cxml release. User code is now strongly encouraged to implement SAX handlers as subclasses of sax:default-handler or sax:abstract-handler. Without the patch, cl-webdav will still work (at least for a transition period), but issue warnings at run time. Thanks, David -------------- next part -------------- A non-text attachment was scrubbed... Name: cl-webdav-sax.diff Type: text/x-diff Size: 494 bytes Desc: not available URL: From edi at agharta.de Sun Oct 21 21:23:43 2007 From: edi at agharta.de (Edi Weitz) Date: Sun, 21 Oct 2007 23:23:43 +0200 Subject: [hunchentoot-devel] New cl-webdav release 0.1.2 (Was: cl-webdav patch for current cxml) In-Reply-To: <20071021185940.GF10680@radon> (David Lichteblau's message of "Sun, 21 Oct 2007 20:59:40 +0200") References: <20071021185940.GF10680@radon> Message-ID: On Sun, 21 Oct 2007 20:59:40 +0200, David Lichteblau wrote: > the attached patch updates cl-webdav's use of SAX for the current > cxml release. > > User code is now strongly encouraged to implement SAX handlers as > subclasses of sax:default-handler or sax:abstract-handler. > > Without the patch, cl-webdav will still work (at least for a > transition period), but issue warnings at run time. Thanks, the patch is in the new release. Edi.