Tue Aug 2 17:05:00 CEST 2005 pc@p-cos.net * Initial revision (v0.3). Tue Aug 16 17:49:56 CEST 2005 pc@p-cos.net * New known but untested problems. Sat Sep 10 12:35:08 CEST 2005 pc@p-cos.net * Minor edit in functionally irrelevant file. Sat Oct 29 03:35:42 CEST 2005 pc@p-cos.net * Changes triggered by Christophe Rhodes / SBCL. Sun Oct 30 00:39:27 CEST 2005 pc@p-cos.net * Checked against OpenMCL 1.0 - no changes necessary. Sat Nov 5 16:02:17 CET 2005 pc@p-cos.net * Added a test for extensible :allocation types. Because of this new test, the feature tests for LispWorks 4.3 didn't work anymore. Instead of fixing this, I have dropped support for LispWorks 4.3. Sat Nov 5 16:05:52 CET 2005 pc@p-cos.net * Added the actual test for extensible :allocation types. Sat Nov 5 22:55:44 CET 2005 pc@p-cos.net * Removed an unnecessary slot definition from a test case. Thu Nov 17 20:59:57 CET 2005 pc@p-cos.net * Known and untested problems turned into test cases. Thu Nov 17 21:01:49 CET 2005 pc@p-cos.net * Added a test that checks whether empty generic functions can be called. Thu Nov 17 21:04:20 CET 2005 pc@p-cos.net * Added a test that checks whether a defmethod form accepts multiple qualifiers. Thu Nov 17 21:05:51 CET 2005 pc@p-cos.net * Made the test whether slot-xxx-using-class methods are specialized on slot definition metaobjects more fine-grained. Thu Nov 17 21:07:25 CET 2005 pc@p-cos.net * Added release notes to reflect recent changes. Wed Nov 30 21:21:36 CET 2005 pc@p-cos.net * Checked against SBCL 0.9.7. Mon Dec 5 14:40:24 CET 2005 pc@p-cos.net * Checked againts clisp 2.36. Tue Dec 20 17:19:13 CET 2005 pc@p-cos.net * Switched to symbols in asdf definition. Wed Dec 28 16:01:56 CET 2005 pc@p-cos.net * Checked against SBCL 0.9.8. Wed Dec 28 22:24:04 CET 2005 pc@p-cos.net * Added a test whether reinitialize-instance calls finalize-inheritance. Tue Jan 3 12:28:10 CET 2006 pc@p-cos.net * Checked against clisp 2.37. Thu Jan 19 14:22:09 CET 2006 pc@p-cos.net * Allegro 8.0 added, and fixed a few typos in the release notes. Thu Jan 19 21:21:51 CET 2006 pc@p-cos.net tagged 0.4 Fri Jan 27 15:24:02 CET 2006 pc@p-cos.net * Checked against clisp 2.38 and SBCL 0.9.9. Wed Feb 1 00:54:59 CET 2006 pc@p-cos.net * Added more detailed checks for metaobject readers. Wed Feb 1 00:56:43 CET 2006 pc@p-cos.net * Finalized the separate listing of release notes. Wed Feb 1 00:59:53 CET 2006 pc@p-cos.net * Conditionalization for LispWorks 4.3 (thanks to Jeff Caldwell). Fri Feb 3 13:10:03 CET 2006 pc@p-cos.net tagged 0.41 Sat Feb 11 20:28:18 CET 2006 pc@p-cos.net * Added a test for slot inheritance in subclasses of specified metaobject classes. The CLOS MOP specification requires that subclasses of specified metaobject classes do not inherit slots whose names are exported from any package and/or accessible in the package common-lisp-user. I have added a test that checks whether this requirement is fulfilled. Unfortunately, most CLOS implementations break this requirement a lot. See Closer to MOP for a detailed report on violations. Tue Feb 14 17:32:13 CET 2006 pc@p-cos.net * Fixed the test for slot inheritance in subclasses of specified metaobject classes. The previously added test was too strict. I have tested whether slot names were exported by any package, while the AMOP specification only requires that they are not exported by the standard packages as defined by ANSI Common Lisp. I have still made it a little bit stricter than that by including the package with the CLOS MOP definitions as well because it is very likely that one uses it for MOP programming. Tue Feb 14 19:36:25 CET 2006 pc@p-cos.net * Added rudimentary support for ECL. Wed Feb 15 14:18:45 CET 2006 pc@p-cos.net * Added rudimentary support for GNU Common Lisp. Wed Feb 15 17:01:05 CET 2006 pc@p-cos.net * Minor change to previous slot inheritance check. In the previous version of the test for inheritance of exported slot names from specified metaclasses, I have checked the direct slots of the metaclasses themselves. This seems somewhat fishy - theoretically, a CLOS MOP could have special cases for slot inheritance for the system-provided metaclasses. Therefore, I have reverted to tests on actual subclasses of the specified metaclasses. Fri Feb 24 22:45:10 CET 2006 pc@p-cos.net * Updated code because MCL is disappearing from OpenMCL's features list Wed Mar 1 21:09:12 CET 2006 pc@p-cos.net * Updated the version number and copyright information in the system definition. Thu Mar 9 15:31:01 CET 2006 pc@p-cos.net * Added "real" support for ECL. The test now run through, but take the reporst with a grain of salt at the moment. Mon Mar 13 23:51:46 CET 2006 pc@p-cos.net * Improved support for ECL. Sat Mar 25 15:53:14 CET 2006 pc@p-cos.net * ECL now supports slot-definition-location, so some tests that were commented out for ECL can now be activated. Sat Mar 25 18:08:49 CET 2006 pc@p-cos.net * More adaptations for ECL. Mon May 1 10:26:08 CEST 2006 pc@p-cos.net * locations are passed to effective slot definitions in ecl. Mon May 1 10:28:06 CEST 2006 pc@p-cos.net * Removed some old commented code. Mon May 1 10:28:26 CEST 2006 pc@p-cos.net * Removed some print statements for debugging purposes. Tue Jul 11 14:14:50 CEST 2006 pc@p-cos.net * Added a test for lambda-list / argument-precedence-order dependencies in re/initialization of generic functions. Tue Jul 11 14:16:53 CEST 2006 pc@p-cos.net * Removed a test whether (setf find-class) calls reinitialize-instance. This was a misunderstanding on my side. Fri Jul 14 21:46:14 CEST 2006 pc@p-cos.net * Added the metaobject readers tests. Thu Jul 20 10:43:14 CEST 2006 pc@p-cos.net * Guarded calles to finalize-inheritance by checks for class-finalized-p. Thu Jul 20 10:45:08 CEST 2006 pc@p-cos.net * Fixed the test for :reinitialize-instance-calls-finalize-inheritance. The test case was wrong because it called finalize-inheritance explicitly, which means that it always succeeded unconditionally. Now, the call to finalize-inheritance is guarded by a check for class-finalized-p. Additionally, the call to reinitialize-instance is more 'sophisticated' to avoid interference of potentially acceptable optimization. The new test reveals that CMUCL, MCL, OpenMCL and SBCL don't implement this feature correctly. Many thanks to Christophe Rhodes for spotting the error in the test case and helping to find a better formulation. Thu Jul 20 16:33:35 CEST 2006 pc@p-cos.net * Updated list of of known missing features. Fri Jul 21 14:54:31 CEST 2006 pc@p-cos.net * Tighter check for :reinitialize-instance-calls-finalize-inheritance. The previous version of this check was a bit too liberal. The new version is tighter, but means that clisp has to be considered not implementing this feature. Sat Jul 29 14:53:22 CEST 2006 pc@p-cos.net tagged 0.42 Mon Aug 21 22:37:32 CEST 2006 pc@p-cos.net * Removed a spurious in-package declaration in the .asd file. Sat Aug 26 12:37:58 CEST 2006 pc@p-cos.net * Checked against SBCL 0.9.16 - added two new extra features. Sat Aug 26 14:46:42 CEST 2006 pc@p-cos.net tagged 0.43 Sat Aug 26 14:58:52 CEST 2006 pc@p-cos.net * Updated version number in the .asd file. Thu Nov 30 21:27:03 CET 2006 pc@p-cos.net * Checked against SBCL 1.0, which introduces new extra features. Thu Dec 28 01:19:27 CET 2006 pc@p-cos.net * Added new tests for standard metaobject classes. Thu Dec 28 01:21:31 CET 2006 pc@p-cos.net * Simplified the output for feature reports. Thu Dec 28 01:22:33 CET 2006 pc@p-cos.net * Added a report function for quick comparison between two versions of a given CL implementation. Thu Dec 28 01:23:23 CET 2006 pc@p-cos.net * Siginificantly simplified the test for extensible allocations. Thu Dec 28 16:08:52 CET 2006 pc@p-cos.net * Updated version number in the .asd file. Thu Dec 28 16:09:27 CET 2006 pc@p-cos.net tagged 0.44