Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!crdgw1!crdgw1.ge.com!barnett From: barnett@crdgw1.crd.ge.com (Bruce Barnett) Newsgroups: comp.protocols.appletalk Subject: Sparc versions of atalkad Message-ID: <3255@crdgw1.crd.ge.com> Date: 16 Oct 89 15:27:15 GMT Sender: news@crdgw1.crd.ge.com Reply-To: barnett@crdgw1.ge.com (Bruce Barnett) Organization: GE Corp. R & D, Schenectady, NY Lines: 51 Who is maintaining atalkad, especially the sparc version? I have been making changes to it, and I have seen patches from two other people for the sparc version of atalkad-align.c Peter Seuffert has posted the most recent patches, but I don't think all of his patches should be installed. (Peter, you "fixed" the calls to log the wrong way, I believe.) I have been making some changes myself, and there is a desperate need for someone to "own" atalkad.c, acting as a coordinator for changes. Can anyone act as a central coordinator? I would do so myself, except for: a) I don't understand all of the code. b) I can't get ANY version of atalkad to work on my system. I have made several changes which improve the -c reporting. I have also made several changes in readtab that eliminates such awful code as: #ifdef sparc /* gak. */ if ((int)cp & (sizeof(short) - 1)) #else if ((int)cp & 1) #endif To be specific, I have inverted the logic in the portion of code that reads the ASCII file so that instead of accepting ANY field and force fitting the value into the proper location, which hopefully won't cause a pointer problem on the sparc, the new version has a very rigid parser that expects a value of the proper field length for that field. The down side is that you can't specify an inet address with shorts, but must use either an L or I field. I also didn't implement character or string fields. The up side is that if you specify the wrong type of field size, you get an error Expected Long or IPaddress in IPFILE field on line 83 It seems to me that this is preferable to the current version, where ANYTHING is valid. However, I don't really know if my changes work. I repeat my question: Who is the coordinator for the sources of atalkad? -- Bruce G. Barnett uunet!crdgw1!barnett