RE: [NeoStats-Devel] [Commits] r2545 - in trunk: . include src/protocol
"M" <[email protected]>
| Newsgroups | gmane.comp.neostats.devel |
|---|---|
| Message-ID | <[email protected]> |
Reed Loden: > > Author: Mark > > Date: Tue Apr 19 05:53:21 2005 > > New Revision: 2545 > > > > Log: > > split ircu support into seperate protocol files for each supported > > form > > no, no, no, and no. > > PLEASE do NOT do that. > Why in the world are you doing it in the first place? > You're adding a ton of problems for yourself that you will > not want to deal with in the first place. > > There is no need to redo code that is pretty much the same > except for a couple of things. Exactly. And there is no reason to allow a change for one variant of IRCu to corrupt code for all other variants. If you have followed the updates to neostats protocol systems and my emails on it you will have noticed a great effort has been made to remove redundancy across all ircds. IRCu was exempted from this since you were making changes (which as noted in my email last night broke vanilla IRCu) and ircu as a protocol requires specific changes. I am also looking at this from a user perspective where a user will not go in and change a define in code so we end up wasting time telling them they need to modify code for there ircd to work. With the split system, they just select the appropriate ircd as for every other protocol supported. The protocol options of PROTOCOL_B64SERVER and PROTOCOL_B64NICK were provided originally to support the new protocol system with IRCu in mind. Unreal 3.2 can also use some b64 for servers and other ircds may need similar support in the future. At the moment the split represents no more work than updating say Ultimate and all variants of it. Future work will reduce the code in each modules in a way that does not compromise the core or the three variants now specifically supported. You should know by now that many of my changes in 3.0 reduce the work and maintainance of the code. Mark.