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

[bknr-devel] Re: [bknr-cvs] r2376 - branches/bos/projects/bos/m2
by Hans Hübner 21 Jan '08

21 Jan '08

http://random-state.net/log/3390120648.html On Jan 21, 2008 3:03 PM, <ksprotte(a)common-lisp.net> wrote: > Author: ksprotte > Date: Mon Jan 21 09:03:34 2008 > New Revision: 2376 > > Removed: > branches/bos/projects/bos/m2/allocation-cache-test.lisp > Modified: > branches/bos/projects/bos/m2/bos.test.asd > Log: > no need to have separate allocation-cache-test.lisp > > > Modified: branches/bos/projects/bos/m2/bos.test.asd > ============================================================================== > --- branches/bos/projects/bos/m2/bos.test.asd (original) > +++ branches/bos/projects/bos/m2/bos.test.asd Mon Jan 21 09:03:34 2008 > @@ -6,7 +6,6 @@ > :components ((:file "packages-test") > (:file "test-suites" :depends-on ("packages-test")) > (:file "test-fixtures" :depends-on ("packages-test")) > - (:file "allocation-test" :depends-on ("test-suites" "test-fixtures")) > - (:file "allocation-cache-test" :depends-on ("test-suites" "test-fixtures")) > + (:file "allocation-test" :depends-on ("test-suites" "test-fixtures")) > ;; (:file "utils" :depends-on ("config")) > )) > _______________________________________________ > bknr-cvs mailing list > bknr-cvs(a)common-lisp.net > http://common-lisp.net/cgi-bin/mailman/listinfo/bknr-cvs >

1 0

[bknr-devel] Recursive locks problem on sbcl
by Denis Mashkevich 02 Oct '07

02 Oct '07

Hi Hans, the problem we've spoken of yesterday on irc (recursive lock attempt warning on sbcl using bknr datastore) proved to be trivial to solve - there is a primitive in sb-thread for exactly this case. I have not tested it extensively yet, but it appears to work. The patch is attached. FWIW, being unable to load some third-party libraries supplied with bknr (mainly cxml and unit-test), I ditched thirdparty altogether, installed all the newest versions of whatever bknr-indices and bknr-datastore depend on, and both seem to work. Thanks, Denis (oudeis).

3 5

[bknr-devel] the datastore and multiprocessing
by Kamen TOMOV 19 Sep '07

19 Sep '07

Hi, I am building an application that uses BKNR-datastore back-end and the portable AllegroServe web server. They run on CMUCL. I was wondering about how to optimally tune AllegroServe so that the system runs reliably and be responsive enough. What will be the optimal numbers of the AllegroServe listeners? When does BKNR-datastore yield to other processes, e.g. if there are several listeners and several web requests that access the datastore how will the system act? Thanks in advance, -- Kamen

1 0

[bknr-devel] compilation with cmucl-19d
by Kamen TOMOV 12 Jul '07

12 Jul '07

Hi, Have you tried running the latest BKNR with cmucl-19d? I had an issue compiling the CXML package. I had to isolate the file character.lisp from the package to compile it (I don't use CXML anyway). The rest seems to work. -- Kamen

2 3

[bknr-devel] objects with transient slots
by Kamen TOMOV 07 Jul '07

07 Jul '07

Hi, It is said in the datastore-manual: "... you can specify that a certain slot is transient, which means that it will not be snapshotted and that its value can be changed outside of a transaction." In my application the value of a transient slot is specified on object creation. When reloading the transaction log (server restart) the transient slot is bound and its value is correct. Is this intended behaviour or not? How is snapshotting the database different in that context? Thanks in advance, -- Kamen

2 5

Re: [bknr-devel] iconv download?
by Hans Hübner 08 May '07

08 May '07

2007/5/8, Kevin Raison <raison(a)chatsubo.net>: > I would be happy to help out with the sbcl port. I have already started > playing around with the grin-neu branch and it seems to work with sbcl > 1.0.5. Of course, I have not yet done anything more than experiment. > Are there known issues with sbcl that I could concentrate on? I have not tried BKNR on SBCL yet, but I expect problems in the multiprocessing code. BKNR makes use of the acl-compat package, so the system dependencies would need to be fixed there. Also, BKNR does not directly support Unicode, but I have used unicode characters in a number of projects by just putting them into strings. This is not the right way to do it, and it may be a stumbling block. Generally, I would like to keep all system dependencies out of the baseline code and see them being put into seperate files (like cmu.lisp, sbcl.lisp etc.). Also, it needs to be made sure that the serialization mechanism works on sbcl. All transactions to the store need to be serialized, and BKNR has a mutex that guards all destructive operations. It needs to be verified that this works on sbcl, too. It may be best to write some tests, and this would be on my todo list. The present tests are too informal to be really useful, and I have concentrated on the application side of things. -Hans

1 0

[bknr-devel] iconv download?
by Kevin Raison 08 May '07

08 May '07

Does anyone on this list have a link to the iconv package that bknr requires? The link from cliki.net (http://www.swj.co.jp/~tahara/lisp/lib/iconv_0.2.tar.gz) is broken, and google has not turned up anything. Thanks, Kevin

2 1

[bknr-devel] Re: Security issue
by Kamen TOMOV 23 Feb '07

23 Feb '07

On петък, Февруари 23 2007, Stoyan Dimov wrote: > Това засега е нвъзможно, защото нямаме контролвърху рутера на > БТК.След като се превключим на линията на Интерпред Наско (Живко > няма достъп до рутера) ще сложи верните ограничения. Портовете така > или иначе са органичени. Остава да ограничим и IP-тата до тези на > Агенцията. Добре, но на тази машина за която става дума не може ли да се сложи Firewall? -- Камен

1 1

[bknr-devel] Can't even install bknr
by Marcin Tustin 19 Feb '07

19 Feb '07

I'm trying to install bknr, but I can't get my lisp (sbcl) to find it. In brief, I've put bknr-svn in $SBCL_HOME, and I have the file $SBCL_HOME/bknr-svn/bknr/init.lisp looking as below. The problem is that once I load the file, (asdf:oos 'asdf:load-op :bknr) just drops me into the debugger with the message "debugger invoked on a ASDF:MISSING-COMPONENT in thread #<THREAD "initial thread" {A744531}>: component "bknr" not found." ------------------ init.lisp: (in-package :cl-user) (require :asdf) ;;;;;;;;;;;;; ;; Tweak this (eval-when (:execute :compile-toplevel :load-toplevel) #+allegro (setf (logical-pathname-translations "bknr") `(("**;*.*.*" "bknr/**/")) (logical-pathname-translations "bknr-thirdparty") `(("**;*.*.*" "thirdparty/**/")) (logical-pathname-translations "eboy") `(("**;*.*.*" "eboy/**/"))) #+cmu (setf (logical-pathname-translations "bknr") `(("**;*.*.*" "home:bknr-sputnik/bknr/**/")) (logical-pathname-translations "bknr-thirdparty") `(("**;*.*.*" "home:bknr-sputnik/thirdparty/**/")) (logical-pathname-translations "eboy") `(("**;*.*.*" "home:bknr-sputnik/eboy/**/"))) #+sbcl (setf (logical-pathname-translations "bknr") `(("**;*.*.*" ,(merge-pathnames (make-pathname :directory '(:relative "bknr-svn" "bknr" :wild-inferiors) :name :wild :type :wild :version :wild) (POSIX-GETENV "SBCL_HOME")))) (logical-pathname-translations "bknr-thirdparty") `(("**;*.*.*" ,(merge-pathnames (make-pathname :directory '(:relative "bknr-svn" "thirdparty" :wild-inferiors) :name :wild :type :wild :version :wild) (POSIX-GETENV "SBCL_HOME")))) (logical-pathname-translations "eboy") `(("**;*.*.*" ,(merge-pathnames (make-pathname :directory '(:relative "bknr-svn" "eboy" :wild-inferiors) :name :wild :type :wild :version :wild) (POSIX-GETENV "SBCL_HOME")))))) (eval-when (:execute :compile-toplevel :load-toplevel) #-sbcl (load #p"bknr-thirdparty:asdf;asdf") #+sbcl (require :asdf)) (pushnew (translate-logical-pathname #p"bknr:src;") asdf:*central-registry* :test #'equal) (pushnew (translate-logical-pathname #p"eboy:src;") asdf:*central-registry* :test #'equal) (defparameter *patch-directory* "bknr:patches;") (defun load-patches (&optional (directory *patch-directory*)) (dolist (file (directory (merge-pathnames directory #p"patch-*.lisp"))) (warn "; Loading patch from file ~A~%" file) (load file))) (defun fix-dpd () #+cmu ;; Die Sache mit dem aktuellen Verzeichnis hat CMUCL noch immer nicht im ;; Griff. Nachbessern! (setf *default-pathname-defaults* (pathname (concatenate 'string (nth-value 1 (unix:unix-current-directory)) "/")))) (defun make-wild-pathname (type directory) (merge-pathnames (make-pathname :type type :name :wild :directory '(:relative :wild-inferiors)) directory)) (defun setup-registry () (mapc #'(lambda (asd-pathname) (pushnew (make-pathname :directory (pathname-directory asd-pathname)) asdf:*central-registry* :test #'equal)) (append (directory #p"bknr-thirdparty:**;*.asd") (directory #p"bknr:**;*.asd")))) (defun clean-registry (&optional (dirs asdf:*central-registry*)) (let ((files (mapcan #'directory (mapcan #'(lambda (dir) (when (pathnamep dir) (list (make-wild-pathname "fas" dir) (make-wild-pathname "lib" dir) (make-wild-pathname "x86f" dir) (make-wild-pathname "fasl" dir)))) dirs)))) (dolist (file files) (when (probe-file file) (format t "Deleting binary file ~S~%" file) (delete-file file))))) #+cmu (load-patches) (setup-registry) (fix-dpd) (pushnew :cl-gd-gif *features*)

3 4

Fwd: [bknr-devel] :read :update keywords meaning
by Hans Hübner 06 Nov '06

06 Nov '06

(forgot to Cc to the list) ---------- Forwarded message ---------- From: Hans Hübner <hans(a)huebner.org> Date: 06.11.2006 14:24 Subject: Re: [bknr-devel] :read :update keywords meaning To: Kamen TOMOV <kamen(a)cybuild.com> Hi Kamen, :read means that a slot reader method is automatically defined, :update means that a reader and a writer method is defined. If you have a class (define-persistent-class foo () ((slot-a :read) (slot-b :update))) then macroexpanding this will yield: (defclass foo (store-object) ((slot-a :reader foo-slot-a :initarg :slot-a) (slot-b :accessor foo-slot-b :initarg :slot-b)) (:metaclass persistent-class)) Thus, :read gives you a :reader method whereas :update yields an :accessor. Both create a :initarg so that you can initialize the slots in a make-object. Note that nothing prevents you from writing to a :read slot using change-slot-values or setf. C-c C-m in slime macroexpands a s-expression, which is pretty handy. Cheers, -Hans 2006/11/6, Kamen TOMOV <kamen(a)cybuild.com>: > Hi, > > I tried to find in the documentation the meaning of the :read and > :update slot options in define-persistent-class and I couldn't. Then I > tried in the source and got lost :-) > > Could you please help? Thanks! > > -- > Kamen > _______________________________________________ > bknr-devel mailing list > bknr-devel(a)common-lisp.net > http://common-lisp.net/cgi-bin/mailman/listinfo/bknr-devel >

1 0