From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gerrit Renker Date: Mon, 22 Oct 2007 14:50:50 +0000 Subject: Re: [RFC]: field name identifier conventions Message-Id: <200710221550.50989@strip-the-willow> List-Id: References: <200710201654.35326@strip-the-willow> In-Reply-To: <200710201654.35326@strip-the-willow> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: dccp@vger.kernel.org Thanks everyone for replying, below is the summary of earlier emails. 1) Ian wrote: | I think the best way to do this is once Arnaldo agrees and then | prepare a big patch to do this. If we don't get Arnaldo's buy in then | it will be hell to maintain the patch set. Fully agree - Arnaldo is the maintainer and as such it is up to him to decide on such things affecting global consistency. If there is a new naming scheme which finds his approval, I am more than happy to update the entire test tree, but if not I wouldn't like to change the patches: these should reflect the existing DCCP style/conventions. 2) All agreed that the naming scheme could be changed. 3) There are the following suggestions: * Arnaldo: "->c2tx_" * Leandro: "tx->c4" If we combine these suggestions, considering that for the CCID sockets this will be likely `hcrx' for RX sockets and `hctx' for TX sockets (like referring to struct sk_buff as skb), then Leandro's scheme is nice and small: * hcrx->c3_last_counter * hcrx->c4_last_counter * hctx->c3_last_win_count * hctx->c4_no_feedback_timer In conclusion, what about the following naming scheme:? * hctx->c"N"_FIELDNAME /* N = CCID number, 0..255 */ * hcrx->c"N"_FIELDNAME * avr->avr_FIELDNAME /* r for record ? */ * av->av_FIELDNAME /* or drop the `av/avr' ? */ The actual patch will probably cost less work than making this scheme consistent.