From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cagney To: Joel Brobecker Cc: Daniel Jacobowitz , Paul Hilfinger , jimb@redhat.com, gdb-patches@sources.redhat.com Subject: Re: [RFA] Introduce notion of "search name" Date: Thu, 13 May 2004 14:29:00 -0000 Message-id: <40A38639.2010207@gnu.org> References: <20040430084538.ECDE1F2E1C@nile.gnat.com> <20040430134955.GA15786@nevyn.them.org> <20040503084937.439F4F2C0A@nile.gnat.com> <20040511194843.GA15952@nevyn.them.org> <20040512105959.806E6F2DE4@nile.gnat.com> <20040512132708.GA25122@nevyn.them.org> <40A2313E.1080100@gnu.org> <20040512142254.GA26822@nevyn.them.org> <40A23E97.1020906@gnu.org> <20040512165907.GF16083@gnat.com> X-SW-Source: 2004-05/msg00401.html I.e., through out GDB there is the assumption that symtab internally uses simple strings. We shouldn't. Recalling that one of the underlying problems here was the need to construct search names on the fly using more complext debug info, we should instead have interfaces such as: This is a worthy goal, but I see it as a goal separate from what the patch is trying to achieve. This is in part true. The mistake is to seperate cleanups (or refactorings) from enhancements. Enhancement and fixes needs to go hand-in-hand with a corresponding cleanups. If we don't do this, we quickly find that those long over-due changes never occure. > From what Paul is saying, the change you > are objecting to introduces a check that is correct within the current > implementation, and makes the kind of assumption that's already used > everywhere. Why can't we use the current framework as is until the > cleanup is done (BTW: who's going to do it, and when?). Admitedly, we > have to document a temporary assumption, but I don't think this will > affect the symbol interface cleanup you're suggesting all that much. If this, as an implementation detail, were to be strictly hidden behind the symtab interface, it wouldn't be a problem. Unfortunatly, as I illustrated, the SYMBOL_*_NAME macros permieate GDB, so hand in hand with this change, should be changes to screw down that aspect of the symtab interface, and then have Ada use that. enjoy, Andew