From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 12582 invoked by alias); 13 Jul 2006 19:25:41 -0000 Received: (qmail 12570 invoked by uid 22791); 13 Jul 2006 19:25:40 -0000 X-Spam-Check-By: sourceware.org Received: from nitzan.inter.net.il (HELO nitzan.inter.net.il) (192.114.186.20) by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 13 Jul 2006 19:25:38 +0000 Received: from HOME-C4E4A596F7 (IGLD-80-230-11-74.inter.net.il [80.230.11.74]) by nitzan.inter.net.il (MOS 3.7.3-GA) with ESMTP id EDJ75671 (AUTH halo1); Thu, 13 Jul 2006 22:24:39 +0300 (IDT) Date: Thu, 13 Jul 2006 19:25:00 -0000 Message-Id: From: Eli Zaretskii To: Nathan Sidwell CC: mark.kettenis@xs4all.nl, gdb-patches@sourceware.org, dan@codesourcery.com In-reply-to: <44B690FA.6000006@codesourcery.com> (message from Nathan Sidwell on Thu, 13 Jul 2006 19:29:14 +0100) Subject: Re: add vector container type Reply-to: Eli Zaretskii References: <44B2273D.9030906@codesourcery.com> <24259.192.87.1.22.1152532444.squirrel@webmail.xs4all.nl> <44B690FA.6000006@codesourcery.com> X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2006-07/txt/msg00158.txt.bz2 > Date: Thu, 13 Jul 2006 19:29:14 +0100 > From: Nathan Sidwell > CC: Mark Kettenis , gdb-patches@sourceware.org, > dan@codesourcery.com > > Eli Zaretskii wrote: > >> Date: Mon, 10 Jul 2006 13:54:04 +0200 (CEST) > >> From: "Mark Kettenis" > >> Cc: gdb-patches@sourceware.org, "Daniel Jacobowitz" > >> > >> Oh, and Eli will want you to write a paragraph on how to use this in > >> gdbint.tex. > > > > (You meant gdbint.texinfo.) Yep, it would be nice. > > Is this version ok? Thanks. Yes, it's okay, but please fix the following gotchas: > + pointers to objects around -- in the former case the pointers are Please use "---" -- 3 dashes in a row -- to produce an em-dash. "--" gets typeset as a shorter dash, almost as a hyphen (and makeinfo simply removes one dash, so we will get "-" in the Info output, which is not what you wanted). There are other instances of "--", please fix them all. > + dereferenced and the objects copied into the vector. The scalar > + object variant is suitable for int-like objects, and the vector "int" is a C symbol, so please use a @code markup for it: ... is suitable for @code{int}-like objects ... > + There are both 'index' and 'iterate' accessors. The iterator returns If the '...' quotes here were meant to emphasize the terminology, @dfn{iterate} etc. will do it better. If the quotes were just quotes, please use the ``...'' style, it looks better in print. Please also fix the other uses of '...' > + use the embedded_size & embedded_init calls to create such objects, ^^^^^^^^^^^^^ ^^^^^^^^^^^^^ These are symbols, so please use @code > + pointer to an array of data), because, if we allow NULL to also NULL is also a symbol. > + Vector types are defined using a DEF_VEC_@{O,P,I@}(TYPEDEF) macro. Please give all the macros the @code markup. > + Variables of vector type are declared using a VEC(TYPEDEF) macro. I suspect that TYPEDEF is a placebo: it stands for something else, like int or pointer. If so, please use @code{VEC(@var{typedef})} instead. @var will do what is necessary to typeset meta-syntactic variables such as this one. Use @var{typedef} in every reference to that argument in the following text as well. Finally, perhaps you could precede this description with a short explanation when would a GDB hacker use this facility, and also add at least a simple list of the functions/macros provided by vec.c.