NETWORK@FRSAC11.BITNET (11/18/87)
About all this unnecessary complex thing: >Date: 16 Nov 87 03:53:15 GMT >From: ucsdhub!jack!man!crash!cacilj!paul@sdcsvax.ucsd.edu (Paul Close) >Subject: Uudecoding on UN*X (was Re: GULAM PROBLEMS) >To: info-atari16@score.stanford.edu > >In article <4433@well.UUCP> dhawk@well.UUCP (David Hawkins) writes: >>Yup. I saved all three parts to one file. Edited it with vi >>and removed the header of the first part. Searched for the >>line that started with include >>and deleted it down to the table line of the second part and All slow, useless and harmful practice, prone to editing error. >>make it seamless, i.e. all lines starting with X >>and did the same connecting the second part to the third part. >>so when I finished the whole file looked the same (after the >>first table header) with each line starting with the same character. >>Ran uudecode on it. >>-- >>David Hawkins {ptsfa,hplabs,ucbvax}!well!dhawk >>It is a luxury to be understood. - Ralph Waldo Emerson - > >I like to uudecode the files on the host UN*X machine, and then transfer >the arc files directly (takes less time...). Since each part is May be true only iff: your link is 8bit from your ST to UN*X. And you lose all checking upon what you transfert. (I have experimented here a case where Kermit is transfering garbage and not telling it!!!. On Un*x the latest CKermit 4E(067) on the ST the Uniterm 1.8a, in at least one case of 'set terminal parity' the transfert is wrong, and Kermit does not know about it!!!) >surrounded by News garbage, they have to be fitted together as described All true uudecode just dont care about mail/news garbage, it is therefor useless to remove it. (This is in uuencode/uudecode specifications!) >above. The editor method gets pretty tiring when you get 11 part archivs!! >Here's what I do: First save each part in it's own file such that * will >reference the files in correct order (I use the part number 01, 02,...). >Then run this shell script with the files as arguments: >csh: > #!/bin/csh ..... >now run uudecode on ``x.uue''. This works like a charm, and saves much >editing!! >- >Paul Close paul@cacilj.CTS.COM ...!{uunet, crash}!cacilj!paul What if you do not have a Un*x ? The 'table' 'begin' 'include' ... lines are precisely lines to allows you to avoid all unecessary editing, with the correct uudecode you will just double click / invoke uudecode on the first file, and all chaining and checking will be automatically done. No sweat. (If it was uuencoded with uue.ttp, not some archeological item) BTW the character translation is not done on BITNET, it is some kind of brain-damaged gateways that are the problem, and uud.ttp is taking care of most of it anyway. (I said uud.ttp, not uudecode.* ...) uud.ttp and uue.ttp are available at all good places. (arced/uuencoded or .hex if you are still naked) Jean-Pierre H. Dumas network@frsac11 (bitnet) network%frsac11.bitnet@mitvma.mit.edu (arpanet) dumas@sumex-aim.stanford.edu (arpanet)