- From: EspeonEefi <eefi@MIT.EDU>
- Date: Tue, 12 Jul 2005 06:01:06 -0400
- To: public-cwm-bugs@w3.org
- Message-Id: <1121162466.28762.12.camel@localhost.localdomain>
On Thu, 2005-06-16 at 17:51 -0400, EspeonEefi wrote: > On Mon, 2005-06-13 at 17:43 -0400, EspeonEefi wrote: > > The documentation for the verbose command line option does not match the > > current state of the code. In addition, the handling of the granularity > > option also seems a bit buggy. > > The attached patch has been checked into CVS. I don't think we can > really test the output of the usage. However, I'm still looking for > examples of when the output from -g 0 differs from that of -g 1; I'm > leaving the closing of this bug until I have a test case that > demonstrates the granularity functionality. Test added. Like the delta return codes test, this test is a bit hacky, though I guess it's a bit less so. Note that after poking some through the code, the lumping code that gets triggered by default and gets disabled by setting -g 0 makes a difference only when the left-hand side is an empty formula. In other instances, the way delta constructs patches seems to automatically lump things that are related to the same left-hand side together. Once again, if you know of more cases in which -g 0 should make a difference, please reply to this thread! Test added: http://www.w3.org/2000/10/swap/test/delta/detailed.tests#t4 -- eefi
Received on Tuesday, 12 July 2005 10:01:21 UTC