lee@chsun1.uchicago.edu (dwight lee) (09/25/90)
A BITNET site sent me a uuencoded binary, and it showed up in my mailbox with blanks at the ends of lines stripped, and some lines concatenated. I'd like to use vi to patch the file up so that it can be properly decoded. Most uuencoded lines are 61 characters long. I'd like to be able to search for a line that isn't 61 characters long. Any reasonably automated method will do! Regexps, macros... anything. Also, is there a way for a search in vi to be inverted? ie "search for the next line which does not begin with a capital M". This would also be good for checking on mangled uuencoded files. Feel free to suggest an alternate method of repair, too. Thanks to you all! - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Dwight A Lee / lee@chsun1.uchicago.edu / 815-758-1389 / tCS/BB / Font I speak only for myself. / "I am not the only dust my mother raised" - TMBG
brister@decwrl.dec.com (James Brister) (09/25/90)
On 25 Sep 90 08:50:27 GMT, lee@chsun1.uchicago.edu (dwight lee) said: > I'd like to be able to search > for a line that isn't 61 characters long. Any reasonably automated method > will do! Regexps, macros... anything. This isn't so easy. Searching for a line that's LONGER than 61 characters isn't to bad: /...............................................................* (That's 63 periods, because you want 62 or more characters) I'm not sure of a simple way to look for lines that are less than 61 characters in length. > Also, is there a way for a search in vi to be inverted? ie "search for > the next line which does not begin with a capital M". This would also be > good for checking on mangled uuencoded files. This is easy: /^[^M] James -- James Brister brister@decwrl.dec.com DEC Western Software Lab., Palo Alto, California. .....!decwrl!brister
dylan@ibmpcug.co.uk (Matthew Farwell) (09/25/90)
In article <lee.654252627@chsun1> lee@chsun1.uchicago.edu (dwight lee) writes: >Most uuencoded lines are 61 characters long. I'd like to be able to search >for a line that isn't 61 characters long. Any reasonably automated method >will do! Regexps, macros... anything. I don't think theres any easy way to actually do this within vi. If you want to do this, use something like awk or perl. >Also, is there a way for a search in vi to be inverted? ie "search for the >next line which does not begin with a capital M". This would also be good for >checking on mangled uuencoded files. Using /^[^M] would do this (where ^M is ^ (carat), NOT ctrl M. This means match anything without M as the first character at the beginning of the line. If you want to do an operation on all line which match this regexp, use the ex commands :g or :v :g/^[^M]/<ex command> or :v/^M/<ex command> :v is just the opposite of :g (as in the -v option of grep) I think it would be far easier for you to sort things out in perl or awk. If you have any problems, mail me. Dylan. -- Matthew J Farwell | Email: dylan@ibmpcug.co.uk The IBM PC User Group, PO Box 360,| ...!uunet!ukc!ibmpcug!dylan Harrow HA1 4LQ England | CONNECT - Usenet Access in the UK!! Phone: +44 81-863-1191 | Sun? Don't they make coffee machines?
steinbac@hpl-opus.HP.COM (Gunter Steinbach) (09/25/90)
> / hpl-opus:comp.editors / lee@chsun1.uchicago.edu (dwight lee) / > 1:50 am Sep 25, 1990 / > Most uuencoded lines are 61 characters long. I'd like to be able to search > for a line that isn't 61 characters long. Any reasonably automated method > will do! Regexps, macros... anything. Just look for /...................................................................../ (Make that 61 dots - I didn't count mine.) > Also, is there a way for a search in vi to be inverted? ie "search for the > next line which does not begin with a capital M". This would also be good for > checking on mangled uuencoded files. That one is easy: /^[^M]/ The first caret anchors you at the front of the line, the caret inside square brackets inverts character- set search. You should not have "set ignorecase" on. Good luck! Guenter Steinbach gunter_steinbach@hplabs.hp.com
Dan_Bloch@TRANSARC.COM (09/26/90)
lee@chsun1.uchicago.edu (dwight lee) writes: > A BITNET site sent me a uuencoded binary, and it showed up in my mailbox > with blanks at the ends of lines stripped, and some lines concatenated. > I'd like to use vi to patch the file up so that it can be properly decoded. > > Most uuencoded lines are 61 characters long. I'd like to be able to > search for a line that isn't 61 characters long. Any reasonably automated > method will do! Regexps, macros... anything. Okay, here we go. What you want to do is something like :v/^..............................................$/s/.*/&<TAG>/ That should be 61 dots. I put in fewer in the interest of having it all fit on one line. What this does is finds every line that isn't exactly 61 characters long and put <TAG> on the end of it. Then you can search through the file for normally for <TAG>, take whatever actions you want, and delete all the <TAG>s when you're done. Obviously, you can use any handy marker instead of <TAG>. > Also, is there a way for a search in vi to be inverted? ie "search for > the next line which does not begin with a capital M". This would also be > good for checking on mangled uuencoded files. That's what the :v command above does. It's exactly like :g, only it searches for lines not matching a given pattern. It only works as a line mode command, so if you want to treat each line individually you need some trick like tagging them. > Feel free to suggest an alternate method of repair, too. Thanks to you all! Couldn't you get the guy to retransmit it? Dan Bloch dan@transarc.com
avi@taux01.nsc.com (Avi Bloch) (09/26/90)
In article <62420017@hpl-opus.HP.COM> steinbac@hpl-opus.HP.COM (Gunter Steinbach) writes: >> / hpl-opus:comp.editors / lee@chsun1.uchicago.edu (dwight lee) / >> Also, is there a way for a search in vi to be inverted? ie "search for the >> next line which does not begin with a capital M". This would also be good for >> checking on mangled uuencoded files. > >That one is easy: >/^[^M]/ > This doesn't always work, since an empty line is also a line 'which does not begin with a capital M' and that search pattern will only find lines with at least 1 character on it. I don't of any way that will also catch empty lines. Any takers? -- Avi Bloch National Semiconductor (Israel) 6 Maskit st. P.O.B. 3007, Herzlia 46104, Israel Tel: (972) 52-522263 avi@taux01.nsc.com
dattier@ddsw1.MCS.COM (David W. Tamkin) (09/27/90)
Dan_Bloch@TRANSARC.COM wrote in <gazyNEr0BwwZIof1o_@transarc.com>: | :v/^..............................................$/s/.*/&<TAG>/ | | That should be 61 dots. I put in fewer in the interest of having it all | fit on one line. What this does is finds every line that isn't exactly | 61 characters long and put <TAG> on the end of it. Then you can search | through the file for normally for <TAG>, take whatever actions you want, | and delete all the <TAG>s when you're done. Obviously, you can use any | handy marker instead of <TAG>. Wouldn't :v/^.....{sixty-one periods total}....$/s/$/<TAG>/ make the same substitution? David Tamkin Box 7002 Des Plaines IL 60018-7002 708 518 6769 312 693 0591 MCI Mail: 426-1818 GEnie: D.W.TAMKIN CIS: 73720,1570 dattier@ddsw1.mcs.com
zlsiial@mcc.ac.uk (A.V. Le Blanc) (09/27/90)
>> Couldn't you get the guy to retransmit it?
Perhaps the original submission came from a site which, like mine,
truncates blank characters from the ends of all lines received.
I have never yet received a UUencoded file in the mail or by
ordinary FTP which I have been able to decode without preprocessing,
and that usually produces error messages anyway.
The code used for UUencoding was simply badly chosen, and some
of us hope it will be changed.
Yours,
A. V. Le Blanc
ZLSIIAL@UK.AC.MCC.CMSdylan@ibmpcug.co.uk (Matthew Farwell) (09/27/90)
In article <4769@taux01.nsc.com> avi@taux01.UUCP (Avi Bloch) writes: >This doesn't always work, since an empty line is also a line 'which does not >begin with a capital M' and that search pattern will only find lines with at >least 1 character on it. > >I don't of any way that will also catch empty lines. Offhand, I can't think of anything that will do this. The only thing that comes close is :v/^M/# which prints out lines which don't have an M at the beginning (including empty lines), with their respective line numbers. You can then use the G command to go to those line numbers. Bit of a bodge really. Dylan. -- Matthew J Farwell | Email: dylan@ibmpcug.co.uk The IBM PC User Group, PO Box 360,| ...!uunet!ukc!ibmpcug!dylan Harrow HA1 4LQ England | CONNECT - Usenet Access in the UK!! Phone: +44 81-863-1191 | Sun? Don't they make coffee machines?
new@ee.udel.edu (Darren New) (09/29/90)
In article <1744@m1.cs.man.ac.uk> zlsiial@cms.mcc.ac.uk (A.V. Le Blanc) writes: >The code used for UUencoding was simply badly chosen, and some >of us hope it will be changed. Hmmmm.... I think it has already been changed. My version uses graves instead of spaces. Another alternative would be to use "abe", a clearly superior choice. -- Darren -- --- Darren New --- Grad Student --- CIS --- Univ. of Delaware --- ----- Network Protocols, Graphics, Programming Languages, Formal Description Techniques (esp. Estelle), Coffee -----
dylan@ibmpcug.co.uk (Matthew Farwell) (09/29/90)
In article <1744@m1.cs.man.ac.uk> zlsiial@cms.mcc.ac.uk (A.V. Le Blanc) writes: >>> Couldn't you get the guy to retransmit it? >Perhaps the original submission came from a site which, like mine, >truncates blank characters from the ends of all lines received. >I have never yet received a UUencoded file in the mail or by >ordinary FTP which I have been able to decode without preprocessing, >and that usually produces error messages anyway. UUencode shouldn't even have blanks in it. (see below) >The code used for UUencoding was simply badly chosen, and some >of us hope it will be changed. Tell me if I'm wrong, but this was my understanding of the uuencode algorithim. 3 8 bit bytes are taken from the file. These are mapped onto 4 8 bit bytes, and the top 2 bits of each character are padded with 0's ie: 11110110 11001001 10101001 gets converted to: 111101 101100 100110 101001 which then gets mapped to: 00111101 00101100 00100110 00101001 You then have ascii characters with values ranging between 0 and 63. In ascii, this still includes control characters and spaces, so 33 is added to these values. You now have a file of characters which range between 33 and 96. In ascii, these characters are all printable + therefore can be mailed without mailers barfing. The file is then split into lines about 60 characters long, and each line has a 'M' added to the front to stop mailers taking any ~'s as commands to the mailer. Main point of all this. UUencode adds 33 (not as would be expected 32) so as not to include the complications which come with having spaces in a file, which as seems to have happened in your case might get stripped from the end of a line. UUencode was designed to be as simple + robust as possible, for transferring files between systems. If you have a uuencode which includes spaces, change it. Dylan. -- Matthew J Farwell | Email: dylan@ibmpcug.co.uk The IBM PC User Group, PO Box 360,| ...!uunet!ukc!ibmpcug!dylan Harrow HA1 4LQ England | CONNECT - Usenet Access in the UK!! Phone: +44 81-863-1191 | Sun? Don't they make coffee machines?
ted@isgtec.uucp (Ted Richards) (10/02/90)
In article <1990Sep25.135441.10647@ibmpcug.co.uk> dylan@ibmpcug.CO.UK (Matthew Farwell) writes: : In article <lee.654252627@chsun1> lee@chsun1.uchicago.edu (dwight lee) writes: : >Most uuencoded lines are 61 characters long. I'd like to be able to search : >for a line that isn't 61 characters long. Any reasonably automated method : >will do! Regexps, macros... anything. : I don't think theres any easy way to actually do this within vi. If you : want to do this, use something like awk or perl. How about v/^.............................................................$/<cmd> This executes <cmd> for every line that does not contain exactly 61 characters (provided you type in exactly 61 dots :-). <cmd> could, for example, put a mark at the beginning or end of every line that you want to edit individually. -- Ted Richards ...uunet!utai!lsuc!isgtec!ted ted@isgtec.UUCP ISG Technologies Inc. 3030 Orlando Dr. Mississauga Ont. Canada L4V 1S8