- tbnl-devel@common-lisp.net - mailman3.common-lisp.net

[hunchentoot-devel] gzip-encoding
by Mac Chan 28 May '07

28 May '07

With this web 2.0 thing, nowadays it's not uncommon to have javascript library weighting over 500k (with extreme case like dojo which adds up to several megabytes). If you google a bit you'll find that this doesn't concern the toolkit authors because they'll say that if you minimize (eliminate whitespaces) the javascript file and turn on gzip encoding it's a non-issue. It was a low-hanging fruit to add gzip-encoding to hunchentoot since the hard work has already been done by Edi's (flexi-stream), Zach (Salza 1) and Sean Ross (gzip-stream 2). In fact it was only a few lines of changes. Using the network monitor in FireBug , my initial testing (with a 790 KB javascript file) shows that the response time is slower with gzip turned on (I'm guessing that it might be related to flexi-stream.) 262 KB 828ms 875ms 735ms 703ms 719ms 593ms 734ms 790 KB 625ms 359ms 672ms 328ms 610ms 641ms 594ms Since I'm testing with the loopback interface, there's virtually no network overhead. So I launch another firefox in vmware and limit the download rate of the network interface to 150k (the most common DSL speed offered in the states) 262 KB 1.72s 1.72s 1.64s 1.79s 790 KB 5.16s 5.21s 5.11s 5.21s Now we're talking. And we can't ignore the 56k line users, can't we? 262 KB 8.39s 5.07s 4.97s 5.07s 790 KB 15.36s … point taken, I don't want to repeat this :-) So now, the questions remain: Should we add this to hunchentoot or make it an extension module? (this will introduce two new dependencies to hunchentoot) If we add this to hunchentoot, should we add a flag to turn on/off gzip encoding? (probably yes) And again, since there's flexi-stream overhead (my guess), should we add a flag to control what content-type we should disable gzip encoding? (probably wasting cpu cycle to gzip jpeg, mp3 and png, etc). Suggestions? -- Mac 1: http://www.cliki.net/Salza 2: http://common-lisp.net/project/gzip-stream/ (defun handle-static-file (path &optional content-type) "A function which acts like a Hunchentoot handler for the file denoted by PATH. Send a content type header corresponding to CONTENT-TYPE or \(if that is NIL) tries to determine the content type via the file's suffix." (unless (or (pathname-name path) (pathname-type path)) ;; not a file (setf (return-code) +http-bad-request+) (throw 'handler-done nil)) (unless (probe-file path) ;; does not exist (setf (return-code) +http-not-found+) (throw 'handler-done nil)) (let ((time (or (file-write-date path) (get-universal-time))) (accept-gzip-p (search "gzip" (header-in :accept-encoding)))) (setf (content-type) (or content-type (mime-type path) "application/octet-stream")) (handle-if-modified-since time) (with-open-file (file path :direction :input :element-type '(unsigned-byte 8) :if-does-not-exist nil) (setf (header-out "Last-Modified") (rfc-1123-date time)) (if accept-gzip-p (setf (header-out "Content-Encoding") "gzip") (setf (content-length) (file-length file))) (let ((out (send-headers))) (when accept-gzip-p (setq out (gzip-stream:make-gzip-output-stream out))) (loop with buf = (make-array +buffer-length+ :element-type '(unsigned-byte 8)) for pos = (read-sequence buf file) until (zerop pos) do (write-sequence buf out :end pos) (finish-output out))))))

2 2

[hunchentoot-devel] Hunchentoot and OpenMCL
by Lennart Staflin 26 May '07

26 May '07

Hi, I have been using OpenMCL and portable aserve. I ran into some trouble with character encoding, and after looking at the code decided that it would be difficult to fix. I am now considering alternatives. I read somewhere that Hunchentoot runs on OpenMCL so I'm going to try it out. Any experience with running Hunchentoot on OpenMCL, specially the bleeding edge OpenMCL? Any experience converting a project from aserve to Hunchentoot? Also I ran into a compile problem: ;Compiler warnings : ; Undefined function CCL::DEQUEUE-TIMER-REQUEST, in INVOKE-WITH-TIMEOUT. ; Undefined function CCL::ENQUEUE-TIMER-REQUEST, in INVOKE-WITH-TIMEOUT. ; Undefined function CCL::MAKE-TIMER-REQUEST, in INVOKE-WITH-TIMEOUT. from port-mcl.lisp: (defun invoke-with-timeout (seconds bodyfn timeoutfn) "Executes the function \(with no arguments) BODY-FN and returns its results but stops execution after DURATION seconds and then instead calls TIMEOUT-FN and returns its values." ;; from Portable AllegroServe (block timeout (let* ((timer (ccl::make-timer-request seconds #'(lambda () (return-from timeout (funcall timeoutfn)))))) (ccl::enqueue-timer-request timer) (unwind-protect (funcall bodyfn) (ccl::dequeue-timer-request timer))))) Probably the timer functions have been removed in some of the later versions of OpenMCL. Unless someone has a fix already I will dig into this. -- //Lennart Staflin

3 5

[hunchentoot-devel] Lisp server page for hunchentoot
by Mac Chan 25 May '07

25 May '07

I ported John Wiseman's lsp for use by hunchentoot. You can check out the source from the contrib directory : svn checkout svn://common-lisp.net/project/tbnl/svn http://common-lisp.net/websvn/listing.php?repname=tbnl&path=%2Fcontrib%2Fls… Comments welcomed. Cheers, -- Mac The following are copied verbatim from the comment (asdf:oos 'asdf:load-op :cl-ppcre) (asdf:oos 'asdf:load-op :hunchentoot-test) (hunchentoot:start-server :port 8080) ;; verify that hunchentoot is running ;; http://localhost:8080/hunchentoot/test/ (asdf:oos 'asdf:load-op :lsp) (in-package :lsp) ;; set debug on so we can examine the generated code (set-lsp-debug t) ;; use cl-user package for all unqualified symbols (set-lsp-package :cl-user) ;; compile the test file (get-lsp-function (merge-pathnames "test.lsp" *this-dir*)) ;; take a look at the generated code (show-lsp-form (merge-pathnames "test.lsp" *this-dir*)) ;; print the output to REPL (do-lsp-request (merge-pathnames "test.lsp" *this-dir*)) ;; install the dispatcher function (push (create-lsp-folder-dispatcher-and-handler ;; uri "/hunchentoot/lsp/" ;; directory *this-dir* ;; valid lsp page suffixes '("lsp" "lhtml" "lhtm" "asp") ;; default content-type for non lsp files "text/plain") hunchentoot:*dispatch-table*) ;; try it out (click refresh a few times to verify that it is _dynamic_) ;; http://localhost:8080/hunchentoot/lsp/test.lsp ;; take a look at the source code too ;; http://localhost:8080/hunchentoot/lsp/lsp.lisp ;; alternatively, you can also create separate handler for each lsp file (push (hunchentoot:create-prefix-dispatcher "/hunchentoot/lsp/test.lsp" (get-lsp-function (merge-pathnames "test.lsp" *this-dir*))) hunchentoot:*dispatch-table*) ;; finally, you can also reuse lsp function within a custom dispatch handler (hunchentoot:define-easy-handler (lsp-demo :uri "/hunchentoot/test/lsp-demo.html") () (do-lsp-request (merge-pathnames "test.lsp" *this-dir*))) ;; http://localhost:8080/hunchentoot/test/lsp-demo.html LAQ: * Why another template module? There are indeed a lot of template libraries including CL-EMB, Edi's own HTML-TEMPLATE, etc. But these libraries are obstinate in that they require you to construct and pass the `environment' variable in order to generate the dynamic content. Secondly, I found that the @if @else template extensions/add-ons are often too limited in expressive power. You are programming lisp in the backend, but all of a sudden you need to drop to a very dumb language when you need to deal with the template code, which is kind of frustrating. John Wiseman's code is very close to what I need, but it is designed to work with Franz's allegoserver. So I port it to work with my preferred web server (hunchentoot) instead. * So should I use this for all my dynamic content generation? No. Absolutely not. If you take a look at the very simple test.lsp, it screams like spaghetti code! In the ideal world, if the graphics designer can deliver properly formatted html in cl-who sexp, I'm all set. But that's not gonna happen. So for pages that I have total control, I prefer to write it using sexp. However, there are times where the designer need to frequently update a messy html template and I cannot afford to sync up with it given the time constraint, using lsp is a good compromise. * Any tips & tricks? Lisp is very hard to write (and even harder to read) without a good editor (like emacs) to help indenting your code and provide structured editing. So unlike other dumber languages (php, perl...), intermixed lisp code with content is extremely hard to maintain and read. In this regard, I would recommend you to use Franz's if-star package ( http://www.franz.com/~jkf/ifstar.txt ) in your template file because it helps you to match the open and close parenthesis for if-then-else conditional expression. Also, unlike other dumber languages where if you have a syntax error in the template file, the compiler can point you to the exact line where it occurs. With lsp, the line number is lost when we use the built-in lisp reader to read the form. However you can use `show-lsp-form' in the REPL to examine the code when you encounter an error.

3 2

[hunchentoot-devel] New release 0.11.0
by Edi Weitz 25 May '07

25 May '07

ChangeLog Version 0.11.0 2007-05-25 Added server names and coupled them with easy handlers (suggested by Mac Chan) Exported SESSION-COOKIE-VALUE instead of SESSION-STRING (suggested by Slava Akhmechet) Documentation fixes (thanks to Victor Kryukov and Igor Plekhov) Download http://weitz.de/files/hunchentoot.tar.gz Cheers, Edi.

1 0

[hunchentoot-devel] (session-cookie-value) should be exported?
by Slava Akhmechet 25 May '07

25 May '07

I'm writing code to rewrite URLs in JavaScript to support AJAX requests for clients that have cookies disabled. It appears that the right thing to send would be the value returned by (session-cookie-value), not the value returned by the exported (session-string). Should 'session-cookie-value' be exported from the 'hunchentoot' package? -- Regards, Slava Akhmechet.

2 1

[hunchentoot-devel] Uploaded files are sometimes incomplete
by Viktor Leis 24 May '07

24 May '07

Hi, when using Hunchentoot with SBCL on debian behind mod_lisp and Apache 1 for video file uploading, the uploads randomly fail (the file is incomplete). Warning on the server: WARNING: 28369838 octets of stray data after form-data sent by client. This happens most of the time if the server is on the same machine as the uploading browser, less often if the server is on a different machine in the LAN and only sometimes if the upload goes through the internet. Using Apache2 seems to fix the problem. Any ideas what the problem may be? Thanks, Viktor

2 2

[hunchentoot-devel] [CL-WEBDAV] Property behaviour
by Ralf Mattes 24 May '07

24 May '07

Ok, i just had a closer look at the property code - too me it seems as if there's a lot of unnecessary redundance in the code. For example: A PROPGET for all properties of a resource will result in a call to 'collect-all-properties' that will call 'all-property-designators' which will result (at the end of a long call chain) in a call to 'get-dead-properties'. After that the same 'get-dead-properties' is called for each property again. Now, according to the documentation 'get-dead-properties' is supposed to return _all_ dead properties so it looks as if we do some extra work here ;-) Cheers, RalfD

2 4

[hunchentoot-devel] Re: Getting MD5
by Gary King 23 May '07

23 May '07

Hmm, It does now. I didn't this morning... sorry for overreacting! (and thanks) On May 23, 2007, at 12:29 PM, Kevin Rosenberg wrote: > Gary King wrote: >> I'm looking for MD5 source. Kevin Rosenberg's link is currently >> inaccessible (I've asked him too and I'm sure he'll fix it soon but >> I'm sometimes impatient! >smile<). > > Hi Gary, > > So http://files.b9.com/md5/ doesn't work for you? > > -- > Kevin Rosenberg > kevin(a)hypershots.com > -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[hunchentoot-devel] Getting MD5
by Gary King 23 May '07

23 May '07

Hi y'all, I'm looking for MD5 source. Kevin Rosenberg's link is currently inaccessible (I've asked him too and I'm sure he'll fix it soon but I'm sometimes impatient! >smile<). thanks, -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

2 1

[hunchentoot-devel] Re: [flexi-streams-devel] Flexi-streams test suite failing on OpenMCL
by Edi Weitz 23 May '07

23 May '07

On Mon, 21 May 2007 21:32:45 +0100, "Tiarnan O'Corrain" <ocorrain(a)gmail.com> wrote: > The flexi-streams test suite is failing on OpenMCL "Version 1.0 > (DarwinPPC32)" on > > Test "hebrew_utf8_lf.txt" (:UTF-8 :EOL-STYLE :LF) [INPUT] > --> (:UTF-8 :EOL-STYLE :LF) [OUTPUT]. FWIW, I just borrowed my wife's iBook (PPC, OS X 10.3) for a couple of minutes and tried with OpenMCL Version 1.1 Prerelease 070512 (PPC-32). The test suite ran without problems for me. Could you check if it works for you as well? > I think this is also preventing Hunchentoot from working on recent > OpenMCL versions In the meantime, I've come to think this is another issue. I also had the problems others reported and the error message was something about some stream being "private" to some process. But I don't have enough time to dig into this, so if nobody steps up to resolve the issues with OpenMCL, I'll remove OpenMCL support from Hunchentoot for now.

3 5