From nablaone at o2.pl Fri Sep 3 07:30:57 2004 From: nablaone at o2.pl (=?iso-8859-2?Q?Rafa=B3_Strzali=F1ski?=) Date: Fri, 3 Sep 2004 09:30:57 +0200 Subject: [elephant-devel] Double closing causes SIGSEGV Message-ID: <20040903073057.D172E630C7@rekin7.go2.pl> There is a bug in elephant. --- test case --- (open-store "/path/to/db") (close-store) (close-store) ----- Result: unhandled SIGSEGV CMUCL 19a OS: Linux 2.6 My solution (rather workaround) is in patch. -- Pozdrawiam, Rafal Strzalinski (nabla) http://nablaone.blogspot.com -------------- next part -------------- A non-text attachment was scrubbed... Name: elephant.patch Type: application/octet-stream Size: 1049 bytes Desc: not available URL: From ben at medianstrip.net Sat Sep 4 09:19:26 2004 From: ben at medianstrip.net (Ben) Date: Sat, 4 Sep 2004 05:19:26 -0400 (EDT) Subject: [elephant-devel] bugfixes Message-ID: <20040903020018.N84030@contarex.medianstrip.net> The bad news is: there was a show-stopper bug in the alpha, having to do with OID generation and transaction locks. (I guess that's why they call it "alpha" -- drat.) The good news is: I've fixed it and a ton of other bugs, and added support for OpenMCL! (Thanks to Rafal Strzalinski for finding and fixing one.) The bad news is: there are still some issues with MOP. Mainly, initforms are being evaluated at class definition time. The good news is: those fixes will come soon too -- we're redesigning the slot allocation stuff, and we have to do some CMUCL-wrangling, probably. In the meantime I've repackaged the 0.1 tarball with the fixes I've done. I've also started making a bunch of regression tests (which require the RT asdf package.) I'll make a more formal announcement (and bump the version number) once I finish writing the regression tests, testing OpenMCL et al and produce more docs. The impatient are welcome to try out the tarball of course. The MOP fixes might take some more time, but will happen soon. Thanks for your interest, B From lam at tuxfamily.org Mon Sep 6 16:17:03 2004 From: lam at tuxfamily.org (Nicolas Lamirault) Date: Mon, 06 Sep 2004 18:17:03 +0200 Subject: [elephant-devel] map ? Message-ID: <87u0ubfk3k.fsf@tuxfamily.org> hi, i read the elephant's tutorial, and i would like to know if it is possible to have a map function with btree ? an example : CL-USER> (use-package "ELE") T CL-USER> (defvar friends-birthdays (make-instance 'btree)) FRIENDS-BIRTHDAYS CL-USER> (add-to-root "friends-birthdays" friends-birthdays) # CL-USER> (setf (get-value "Andrew" friends-birthdays) "12/22/1976") "12/22/1976" CL-USER> (get-value "Andrew" friends-birthdays) "12/22/1976" T CL-USER> (setf (get-value "lam" friends-birthdays) "01/01/1976") "01/01/1976" CL-USER> (get-value "lam" friends-birthdays) "01/01/1976" T and a mapbtree function : (mapbtree #'(lambda (key value) (format t "~A : ~A~%" key value)) friends-birthdays) -- Nicolas Lamirault From ben at medianstrip.net Mon Sep 6 17:18:41 2004 From: ben at medianstrip.net (Ben) Date: Mon, 6 Sep 2004 13:18:41 -0400 (EDT) Subject: [elephant-devel] map ? In-Reply-To: <87u0ubfk3k.fsf@tuxfamily.org> References: <87u0ubfk3k.fsf@tuxfamily.org> Message-ID: <20040906130408.A92934@contarex.medianstrip.net> not yet. i plan on filling out the btree functionality with cursors, secondary indices and higher-order functions soon -- it's next on my list after making more tests. if people have ideas on what they'd like this interface to be like, let me know. i'm leaning towards a "maphash" type interface. i guess "with-hash-table-iterator" can be made on top of that with symbol-macrolet. i don't want to get into java-esque iterator protocols unless people really want it. right now i'm testing whether or not it is worth it (performance-wise) to try to implement secondary indices "in Sleepycat", using their "DB->associate" interface. the hard part about this is it requires some finesse with callback functions. the docs seem to indicate that there is no benefit, but it's worth testing. in other news andrew says he is working on revising how the MOP stuff works. it shouldn't affect most people, but the syntax for declaring slots transient might change. i'll keep people updated with info as i get it. thanks for your interest. B On Mon, 6 Sep 2004, Nicolas Lamirault wrote: > > hi, > i read the elephant's tutorial, and i would like to know if it is > possible to have a map function with btree ? > an example : > > CL-USER> (use-package "ELE") > T > CL-USER> (defvar friends-birthdays (make-instance 'btree)) > FRIENDS-BIRTHDAYS > CL-USER> (add-to-root "friends-birthdays" friends-birthdays) > # > CL-USER> (setf (get-value "Andrew" friends-birthdays) "12/22/1976") > "12/22/1976" > CL-USER> (get-value "Andrew" friends-birthdays) > "12/22/1976" > T > CL-USER> (setf (get-value "lam" friends-birthdays) "01/01/1976") > "01/01/1976" > CL-USER> (get-value "lam" friends-birthdays) > "01/01/1976" > T > > and a mapbtree function : > > (mapbtree #'(lambda (key value) > (format t "~A : ~A~%" key value)) > friends-birthdays) > > -- > Nicolas Lamirault > > _______________________________________________ > elephant-devel site list > elephant-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/elephant-devel > From ben at medianstrip.net Sun Sep 19 18:49:18 2004 From: ben at medianstrip.net (Ben) Date: Sun, 19 Sep 2004 14:49:18 -0400 (EDT) Subject: [elephant-devel] Elephant 0.2 Message-ID: <20040919144407.M53520@contarex.medianstrip.net> Elephant 0.2 was released on September 19th, 2004. This is an BETA release. New features: * Secondary indices and cursors * PPC Darwin OpenMCL / SBCL * Doc strings and improved documentation * An RT-based test suite * many bugfixes This release has been tested on CMUCL 19a, SBCL 0.8.14 and Allegro 6.2 on x86 Linux and FreeBSD, and OpenMCL 0.14.2-p1 and SBCL 0.8.14 on PPC Darwin. Tarballs (and now HTML docs) can be found on the website: http://www.common-lisp.net/project/elephant This release should be a lot more stable than the last. Enjoy! Ben and Andrew PS Nicholas, this release has a map-btree function, in addition to more general cursors. From xach at xach.com Fri Sep 24 20:16:29 2004 From: xach at xach.com (Zach Beane) Date: Fri, 24 Sep 2004 16:16:29 -0400 Subject: [elephant-devel] can't get started: open-store fails with "Invalid argument" Message-ID: <20040924201629.GW23441@xach.com> I downloaded UFFI 1.4.26 and Elephant 0.2. I changed the path to libdb to point to my /lib/libdb-4.2.so, which is part of the db4 package on my system. When I try (open-store "xach"), I get: debugger invoked on a SLEEPYCAT:DB-ERROR in thread 22150: Berkeley DB error: Invalid argument restarts (invokable by number or by possibly-abbreviated name): 0: [ABORT ] Abort handling SLIME request. 1: Reduce debugger level (leaving debugger, returning to toplevel). 2: [TOPLEVEL] Restart at toplevel READ/EVAL/PRINT loop. ("varargs entry for SLEEPYCAT:DB-ENV-OPEN" # "xach" :JOINENV NIL :INIT-CDB NIL :INIT-LOCK T :INIT-LOG T :INIT-MPOOL T :INIT-REP NIL :INIT-TXN T :RECOVER NIL :RECOVER-FATAL NIL :CREATE T :LOCKDOWN NIL :PRIVATE NIL :SYSTEM-MEM NIL :THREAD T :MODE 416) 0] What should I do? Zach From ben at medianstrip.net Sat Sep 25 01:17:55 2004 From: ben at medianstrip.net (Ben) Date: Fri, 24 Sep 2004 21:17:55 -0400 (EDT) Subject: [elephant-devel] can't get started: open-store fails with "Invalid argument" In-Reply-To: <20040924201629.GW23441@xach.com> References: <20040924201629.GW23441@xach.com> Message-ID: <20040924210557.M88492@contarex.medianstrip.net> sorry i've been out of town. also for some reason the elephant mailing list rejected my last message. that non-illuminating error (which Berkeley DB produces) indicates that one of the arguments passed to berkeley makes it unhappy. unfortunately as is common with C libraries, it doesn't tell you which one. what OS / DB package are you using? the only way i know how to test this is to use the underlying sleepycat function. try opening a new environment directly using the "db-env-open" function -- the code for "open-controller" in "controller.lisp" should be easy to follow, but flip some of the switches. if i had to make a guess, perhaps your libdb was compiled without thread support. i'm assuming that you have a clean and writable directory to work with. i don't think any of the header constants have changed within the 4.2 release of berkeley db, but worse come to worse you might try compiling berkeley db yourself. sorry i can't be more help. let me know what you find out -- it would be important to know for me what common berkeley DB packages are like. take care, B On Fri, 24 Sep 2004, Zach Beane wrote: > I downloaded UFFI 1.4.26 and Elephant 0.2. I changed the path to libdb > to point to my /lib/libdb-4.2.so, which is part of the db4 package on > my system. > > When I try (open-store "xach"), I get: > > debugger invoked on a SLEEPYCAT:DB-ERROR in thread 22150: > Berkeley DB error: Invalid argument > restarts (invokable by number or by possibly-abbreviated name): > 0: [ABORT ] Abort handling SLIME request. > 1: Reduce debugger level (leaving debugger, returning to > toplevel). > 2: [TOPLEVEL] Restart at toplevel READ/EVAL/PRINT loop. > ("varargs entry for SLEEPYCAT:DB-ENV-OPEN" > # > "xach" > :JOINENV > NIL > :INIT-CDB > NIL > :INIT-LOCK > T > :INIT-LOG > T > :INIT-MPOOL > T > :INIT-REP > NIL > :INIT-TXN > T > :RECOVER > NIL > :RECOVER-FATAL > NIL > :CREATE > T > :LOCKDOWN > NIL > :PRIVATE > NIL > :SYSTEM-MEM > NIL > :THREAD > T > :MODE > 416) > 0] > > What should I do? > > Zach > > _______________________________________________ > elephant-devel site list > elephant-devel at common-lisp.net > http://common-lisp.net/mailman/listinfo/elephant-devel > From ben at medianstrip.net Mon Sep 27 19:19:35 2004 From: ben at medianstrip.net (Ben) Date: Mon, 27 Sep 2004 15:19:35 -0400 (EDT) Subject: [elephant-devel] Elephant Tutorial Message-ID: <20040927151659.L16179@contarex.medianstrip.net> Hi folks, I've put a tutorial on the homepage on how to use Elephant with Araneida to make a simple Blog. http://www.common-lisp.net/project/elephant/with-araneida Hopefully this will be helpful. In bad news I found some silly typos in the release I cut, it shouldn't affect most people but I've updated the tarball. Please refresh. Take care, B