dcw@SUN-BEAR.LCS.MIT.EDU (Dave Whitney) (04/11/89)
(you'll note this is going to the whole net, so bear with me if some of this doesn't appropriate for you specifically). Alright. Here's the deal: (note: "kermit" is any modem program claiming to support the Kermit file transfer protocol. It seems that the only program that does this properly is Kermit-65, available for free from Apple2-l and Kermsrv and perhaps n other different sources.) BinSCII v1.0.2 doesn't care what (if any) the line terminator is. If Kermit is messing that up, then it is a very bad bug, but irrelevant to BinSCII. BinSCII 1.01 required that the line terminator be just 1 character, but it didn't matter what that one char was. BinSCII doesn't use any "odd" characters. The characters appearing in BinSCII files are the uppercase and lowercase letters and the ten digits and open and close paren. The only characters which *might* be considered tricky are the parens, but I think they are safe enough. If Kermit is choking on BinSCII files, then it is most likely choking on any ordinary text files also. If you intend on uploading newly created BinSCII files by using Kermit, then you should create them with the CR line terminator and NOT the LF terminator - no matter what system you are sending the files to. Kermit is supposed to translate line terminators for you. Since the CR is the standard on Apples, Apple Kermit will be looking for CRs and nothing else. If you intend on uploading newly created BinSCII files by any other "straight" binary method (such as Xmodem), then create the file with LFs if you are sending to a Unix machine. If you're sending to something that needs CRLF, then you're (momentarily) out of luck. I'll work on this. For those who haven't heard, BinSCII does NOT require you to chop mail headers/trailers from the text files. You may also have more than one BinSCII "file" (I call them each segments) in a text file. BinSCII just keeps chugging until there are no more segments to chug. It doens't matter what order you process the segments - as long as you do all of them, then everything will be happy. A later version will have a way of telling you if segments are missing. BinSCII 1.0.3 will be posted in BinSCII and Executioner formats (shortly). It tells the user what's going on in a slightly more clear way. It also starts numbering created segment files at number 01. On top of that, it numbers the files in HEX, so now 256 segment files can be created (that puts the upper translatable file size at a shade over 3 MB). I'm hoping to have a nice user interface attached soon. Andy Nicholas and I are talking about it. I hope this clarifies any problems people are having. Dave Whitney dcw@sun-bear.lcs.mit.edu dcw@goldilocks.mit.edu ...!mit-eddie!sun-bear!dcw My employer pays me well. This, however, does not mean he knows what I say. I wrote Z-Link & BinSCII. Send me bug reports. I use a //GS. Send me Tech Info.