Return-Path: Delivered-To: apmail-apr-dev-archive@apr.apache.org Received: (qmail 11448 invoked by uid 500); 29 Dec 2002 22:57:19 -0000 Mailing-List: contact dev-help@apr.apache.org; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: Delivered-To: mailing list dev@apr.apache.org Received: (qmail 11437 invoked from network); 29 Dec 2002 22:57:19 -0000 Date: Sun, 29 Dec 2002 15:12:40 -0800 (PST) From: X-X-Sender: To: Subject: testdso.c: library_not_impl Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N This function is currently being mis-used. On most platforms (most Unixes, HP, Windows), libraries and modules are identical, on Darwin they seem to not be equivalent. For that reason, we have two sets of tests, one for modules and another for libraries. Currently on Windows, we are reporting that the library tests aren't implemented because they are identical to modules. The goal for the test suite is to remove as many Failures and NotImpl results as possible. To further that goal, we have two options. 1) Just run the tests twice (once for modules and once for libraries) even though it is testing the same thing. 2) Don't run the tests, but don't report that they weren't run. Unix is currently doing option 1. Windows is doing a third option, not running them and reporting that they aren't being run. I would like to come up with a single solution that all platforms follow. If a platform doesn't have separate libraries and modules, should the library tests be run or not? Ryan