From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cagney To: Eli Zaretskii Cc: gdb-patches@sources.redhat.com Subject: Re: [rfa] Fix DOS clash with mi0-var-*.exp Date: Fri, 29 Jun 2001 08:31:00 -0000 Message-id: <3B3C9491.5020500@cygnus.com> References: <3B3B8433.7020300@cygnus.com> <3791-Fri29Jun2001102809+0300-eliz@is.elta.co.il> X-SW-Source: 2001-06/msg00521.html > Date: Thu, 28 Jun 2001 15:23:31 -0400 >> From: Andrew Cagney >> >> The attatched patch addresses a DOS file name clash caused by all the >> mi0-var-*.exp files I recently created. >> >> Is this the correct way to do this? The files aren't used in a DOS build. > > > Yes, this is the way. However: > > >> + @V@/gdb/testsuite/gdb.mi/mi0-var-cmd.exp @V@/gdb/testsuite/gdb.mi/mi0-varcmd.exp >> + @V@/gdb/testsuite/gdb.mi/mi0-var-child.exp @V@/gdb/testsuite/gdb.mi/mi0-varchild.exp > > > These two still clash: truncation to 8+3 yields "mi0-varc.exp" for > both of them. Will fix. Hmm, that would mean that mi-var-cmd.exp and mi-var-child.exp also clash. Ah, yes found them hidden in fnchange.lst, would you object to me (separatly) sorting that file? > Actually, I wonder why do we have to keep those mi0- prefixes. This > is all in gdb.mi subdirectory, so there should be no need to stress > the fact we are talking about mi. If the zero is significant, we can > keep it, but "mi" can go, I think. If my memory is correct, it is a namespace proof things. While it looks like a two level structure dejagnu ignores this and just worries about the basename. If I tell dejagnu to run the test ``var-cmd.exp'' I get all of them. Andrew