- armedbear-devel - mailman3.common-lisp.net

[armedbear-devel] Release branch (0.25) created: please test
by Erik Huelsmann 04 Mar '11

04 Mar '11

After the merge of the UNSAFE-P-REMOVAL branch to trunk, Mark Evenson discovered a regression: we failed to compile Montezuma. With some help of Alessio Stalla, I've been able to fix this regression last night, clearing up the last blocking issue for branching for a release. In order to make sure our release works with your software, please test the release branch with it. You can check out the branch using Subversion, from svn://common-lisp.net/project/armedbear/svn/branches/0.25.x Any findings can be reported to this list, or, if you have Trac access, directly as a Trac ticket. Thanks in advance for your efforts! Bye, Erik.

1 0

[armedbear-devel] Getting directory listing with symlinks included
by Zach Beane 02 Mar '11

02 Mar '11

If I have a directory containing a file named a.txt and a symlink b.txt that points to a.txt, is there any ABCL function I can call to get both directory entries as separate objects, i.e. the list (#p"a.txt" #p"b.txt")? The semantics of CL:DIRECTORY seem to require resolving the b.txt symlink, so I'm trying to figure out an extra-standard way to do it in ABCL, if possible. Thanks, Zach

2 7

Re: [armedbear-devel] Bordeaux Threads and ABCL
by Mark Evenson 01 Mar '11

01 Mar '11

On Feb 28, 2011, at 08:05 , Andrew Pennebaker wrote: > I'm having trouble getting threads to work on ABCL. I've filed [ticket #135][1] to track the current status of BORDEAUX-THREADS on ABCL. Apparently, among other things that affect trunk is that we have mistakenly left references to THREAD-LOCK and THREAD-UNLOCK in our autoloads which I will remove as soon as a track down the svn commit that removed ThreadLock.java. [1]: http://trac.common-lisp.net/armedbear/ticket/135 -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."

1 1

[armedbear-devel] translate-logical-pathname downcasing paths inconsistently
by Jonathan P. Bona 27 Feb '11

27 Feb '11

As the following code and transcript demonstrate, translate-logical-pathname appears to be downcasing paths inconsistently: ;;; test.cl (setf (logical-pathname-translations "foo") '(("foo:bar;*.lisp" "/FooRoot/bar/*.lisp") ("foo:*.lisp" "/FooRoot/*.lisp"))) ;;; In the following pathname translation, the entire path is downcased (pprint (translate-logical-pathname "foo:bar;foobar.lisp")) ;;; In each of the next two, the entire path EXCEPT for the host is downcased -- ;;; the host is unchanged (pprint (translate-logical-pathname "foo:foobar.lisp")) (pprint (translate-logical-pathname "foo:Foobar.lisp")) --- Armed Bear Common Lisp 0.24.0 Java 1.6.0_22 Apple Inc. Java HotSpot(TM) Client VM Low-level initialization completed in 0.478 seconds. Startup completed in 1.386 seconds. Type ":help" for a list of available commands. CL-USER(1): (load "test.cl") #P"/fooroot/bar/foobar.lisp" #P"/FooRoot/foobar.lisp" #P"/FooRoot/foobar.lisp" T CL-USER(2): --- SBCL and CLISP both produce the following output for the above code: #P"/FooRoot/bar/foobar.lisp" #P"/FooRoot/foobar.lisp" #P"/FooRoot/foobar.lisp" Allegro's ALISP and MLISP both produce the following: #P"/FooRoot/bar/foobar.lisp" #P"/FooRoot/foobar.lisp" #P"/FooRoot/Foobar.lisp"

2 1

[armedbear-devel] :case :common
by pjb＠informatimago.com 27 Feb '11

27 Feb '11

CLHS says: 19.2.2.1.2.2 Common Case in Pathname Components For the functions in Figure 19-2, a value of :common for the :case argument that these functions should receive and yield strings in component values according to the following conventions: * All uppercase means to use a file system's customary case. * All lowercase means to use the opposite of the customary case. * Mixed case represents itself. Note that these conventions have been chosen in such a way that translation from :local to :common and back to :local is information-preserving. [pjb@kuiper :0.0 lisp]$ abcl Armed Bear Common Lisp 0.20.0 Java 1.6.0_22 Sun Microsystems Inc. Java HotSpot(TM) Server VM Low-level initialization completed in 0.265 seconds. Startup completed in 0.718 seconds. Type ":help" for a list of available commands. CL-USER(1): (make-pathname :name "TEST" :type "LISP" :case :common) #P"TEST.LISP" CL-USER(2): (namestring (make-pathname :name "TEST" :type "LISP" :case :common)) "TEST.LISP" I'm on a linux system on a ext3 file system. The customary case is lower case (case significant on this particular file system, but 99.999% of the files on unix are lower case). Therefore I would expect to get #P"test.lisp" Notice that: (make-pathname :name "test" :type "lisp" :case :common) should produce #P"TEST.LISP" and that: (make-pathname :name "Test" :type "Lisp" :case :common) should produce #P"Test.Lisp" And that I understand "yeld strings in component values" to mean that the processing should be done on each parameter independently. -- __Pascal Bourguignon__ http://www.informatimago.com/ A bad day in () is better than a good day in {}.

3 4

[armedbear-devel] abcl and maxima
by Reinhard Oldenburg 27 Feb '11

27 Feb '11

Dear All, after quite a while I came back to running maxima on abcl becaus there is an increasing interest in incorporationg maxima in Java based projects. Some time ago Mark explained that in principal loadinf form jars can be done, but it seems that it ned some time to twea maxima-load.lisp so that it loadsfrom a jar containing all the abcl files. Has anything happend on this in the meantime? Reinhard -- Prof. Dr. Reinhard Oldenburg Goethe-Universität Frankfurt Institut für Didaktik der Mathematik und Informatik Senckenberganlage 9 (Sekretariat) Robert-Mayer-Str. 10 (Dienstzimmer) D-60325 Frankfurt/Main Telefon +49 (0)69 798 23770 Telefax +49 (0)69 798 22553 E-Mail oldenbur(a)math.uni-frankfurt.de WWW www.math.uni-frankfurt.de/~oldenbur

2 1

[armedbear-devel] Support for -- in parameter lists
by Ville Voutilainen 20 Feb '11

20 Feb '11

It's common that a double-dash (--) alone can be used to tell a command line parser to stop interpreting further arguments. So, I implemented that very feature for ABCL in r13226. Lo and behold: given a test snippet, saved in cmdline.lisp: (loop for item in *command-line-argument-list* do (format t "got arg ~a~%" item)) we can do ./abcl --batch --eval '(load "cmdlinetest.lisp")' -- hops hups jee jee --eval '(format t "hah~%")' -- -- -- Armed Bear Common Lisp 0.25.0-dev-svn-13225M Java 1.6.0_22 Sun Microsystems Inc. Java HotSpot(TM) Server VM Low-level initialization completed in 0.626 seconds. Startup completed in 1.778 seconds. got arg hops got arg hups got arg jee got arg jee got arg --eval got arg (format t "hah~%") got arg -- got arg -- got arg --

1 0

Re: [armedbear-devel] newbie question
by Lukas Georgieff 17 Feb '11

17 Feb '11

Hi "all", thanks for the great amount of answers. Seems that the best way to handle my problem is to pack the necessary file within the jar file. Best regards Lukas -----Original Message----- From: Alessio Stalla Sent: Wednesday, February 16, 2011 5:35 PM To: Mark Evenson Cc: armedbear-devel(a)common-lisp.net Subject: Re: [armedbear-devel] newbie question On Wed, Feb 16, 2011 at 5:23 PM, Mark Evenson <evenson(a)panix.com> wrote: > To use more CL-like language: a file "test.lisp" has a FASL "test.abcl". > Our FASL is a zip of the initialization code ('%%%._'), the classes > our bytecode compiler emits ('%%%-nnn.cls') for each top-level form in > the corresponding ('%%%.lisp') source file. The contents of the > bootstraip ('%%%._') are executed when this FASL is (re)loaded. Replace "for each top-level form" with "for each function definition" and it's 100% correct ;) A. _______________________________________________ armedbear-devel mailing list armedbear-devel(a)common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/armedbear-devel

1 0

[armedbear-devel] newbie question
by Lukas Georgieff 16 Feb '11

16 Feb '11

Hi, I am new to ABCL and have a question to it's capabilities:My goal is to load an exsting part of a system coded in lisp into Java and to use the lisp code as basic library for a java application. Currently my approach is to load the lisp code with:Interpreter interpreter = Interpreter.createInstance();interpreter.eval("(load \"my-lisp-code.lisp\")");... So the lisp code must be present in the project every time it is running.Is there any chance to translate the lisp code to a Java source file by using ABCL? So it would not be necessary to provide the lisp code in the java application after it is translated once. Best regards Lukas

4 6

[armedbear-devel] 'unsafe-p-removal' branch integrated in trunk
by Erik Huelsmann 15 Feb '11

15 Feb '11

Just a second ago, I've merged the unsafe-p-removal branch to trunk. It's been sitting in its own branch, completed, for about 4 weeks, waiting for us to release 0.24.0 - which cleared trunk for destabilizing changes. The purpose of the change is to push the responsibility of maintaining stack consistency (ie. prevent 'stack inconsistency' class verification exceptions) to pass2. Before, this responsibility was a joint effort between pass1 and pass2, with a heavy focus on pass1. However since we started inlining local functions, not enough information is available in pass1 anymore to execute its responsibilities correctly. In pass2 all information is available to take on this responsibility and with some additional infrastructure, that's what this change achieves. Due to the fact that quite a bit of pass2 code has been refactored, the change could have a slightly destabilizing effect on trunk. Please help test your software on ABCL to make sure we release - when we do - code as stable as ever! Bye, Erik.

1 0