From decwrl!uunet!allbery Thu Mar 29 20:48:10 PST 1990 Article 1464 of comp.sources.misc: Path: decwrl!uunet!allbery From: wyle@b.inf.ethz.ch ("Mitchell F. Wyle") Newsgroups: comp.sources.misc Subject: v11i089: mailsplit 2.7 Message-ID: <82806@uunet.UU.NET> Date: 29 Mar 90 23:47:25 GMT Sender: allbery@uunet.UU.NET Lines: 2099 Approved: allbery@uunet.UU.NET (Brandon S. Allbery - comp.sources.misc) Posting-number: Volume 11, Issue 89 Submitted-by: wyle@b.inf.ethz.ch ("Mitchell F. Wyle") Archive-name: mailsplit2.7/part01 People have posted several utilities similar to mailsplit recently; tarmailchunky, shipit, abe, and ztar come to mind. However, there are a lot of users out there waiting for mailsplit v2.7. I have been promising users a new release for almost a year. Version 2.7 fixes some bugs and adds more system-V compatibility. If someone could test mailsplit 2.7 on RISCOS I'd be much obliged. Please send bug fixes, flames, comments, etc. to me at uunet!ethz!wyle. Thanks, -Mitch #! /bin/sh # This is a shell archive. Remove anything before this line, then unpack # it by saving it into a file and typing "sh file". To overwrite existing # files, type "sh file -c". You can also feed this as standard input via # unshar, or by typing "sh 'Configure' <<'END_OF_FILE' X#! /bin/sh X# X# Configure shell script for mailsplit X# X# @(#)Configure 2.6 88/08/20 M F Wyle X# X# "Yes, you may rip this off to use in other distribution packages." X# -Larry Wall- X# X# Thanks Larry I did! -Mitch X# X# If your sh doesn't like these # comments, nuke them. You must X# also get rid of the # comments in mailsplit and joinr. X# X# $Id: Configure,v 1.5 90/02/07 12:05:00 yamanaka Rel $ X# Xpackage=mailsplit X X: sanity checks XPATH='.:/bin:/usr/bin:/usr/local/bin:/usr/ucb:/usr/local:/usr/lbin:/etc:/usr/new:/usr/new/bin:/usr/nbin:/usr/ucb/bin:/usr/old' Xexport PATH || (echo "OOPS, this isn't sh. Desperation time. I will feed Xmyself to sh."; sh $0; kill $$) X Xif test ! -t 0; then X echo "Say 'sh Configure', not 'sh /dev/null 2>&1 && \ X echo "(I see you are using the Korn shell. Some ksh's blow up on XConfigure," && \ X echo "especially on exotic machines. If yours does, try the Bourne shell Xinstead.)" X Xecho "creating a work directory named UU" Xif test ! -d UU; then X mkdir UU Xfi Xcd UU X X: We must find out about Eunice early Xeunicefix=':' Xif test -f /etc/unixtovms; then X eunicefix=/etc/unixtovms Xfi Xif test -f /etc/unixtovms.exe; then X eunicefix=/etc/unixtovms.exe Xfi X X: some greps do not return status, grrr. Xecho "grimblepritz" >grimble Xif grep blurfldyick grimble >/dev/null 2>&1 ; then X contains=contains Xelif grep grimblepritz grimble >/dev/null 2>&1 ; then X contains=grep Xelse X contains=contains Xfi Xrm -f grimble X: the following should work in any shell Xcase "$contains" in Xcontains*) X echo " " X echo "AGH! Grep doesn't return a status. Attempting remedial action." X cat >contains <<'EOSS' Xgrep "$1" "$2" >.greptmp && cat .greptmp && test -s .greptmp XEOSS Xchmod +x contains Xesac X X: first determine how to suppress newline on echo command Xecho "Checking echo to see how to suppress newlines..." X(echo "hi there\c" ; echo " ") >.echotmp Xif $contains c .echotmp >/dev/null 2>&1 ; then X echo "...using -n." X n='-n' X c='' Xelse X cat <<'EOM' X...using \c XEOM X n='' X c='\c' Xfi Xecho $n "Type carriage return to continue. Your cursor should be here-->$c" Xread ans Xrm -f .echotmp X X: now set up to do reads with possible shell escape and default assignment Xcat <myread Xcase "\$fastread" in Xyes) ans=''; echo " " ;; X*) ans='!';; Xesac Xwhile expr "X\$ans" : "X!" >/dev/null; do X read ans X case "\$ans" in X !) X sh X echo " " X echo $n "\$rp $c" X ;; X !*) X set \`expr "X\$ans" : "X!\(.*\)\$"\` X sh -c "\$*" X echo " " X echo $n "\$rp $c" X ;; X esac Xdone Xrp='Your answer:' Xcase "\$ans" in X'') ans="\$dflt";; Xesac XEOSC X: set up shell script to do ~ expansion Xcat >filexp <&2 X exit 1 X fi X case "\$1" in X */*) X echo \$dir/\`$expr x\$1 : '..[^/]*/\(.*\)'\` X ;; X *) X echo \$dir X ;; X esac X fi X ;; X*) X echo \$1 X ;; Xesac XEOSS Xchmod +x filexp X$eunicefix filexp X X X X: general instructions Xcat <loc X$startsh Xcase \$# in X0) exit 1;; Xesac Xthing=\$1 Xshift Xdflt=\$1 Xshift Xfor dir in \$*; do X case "\$thing" in X .) X if test -d \$dir/\$thing; then X echo \$dir X exit 0 X fi X ;; X *) X if test -f \$dir/\$thing; then X echo \$dir/\$thing X exit 0 X elif test -f \$dir/\$thing.exe; then X : on Eunice apparently X echo \$dir/\$thing X exit 0 X fi X ;; X esac Xdone Xecho \$dflt Xexit 1 XEOSC Xchmod +x loc X$eunicefix loc Xloclist=" Xbasename Xcat Xtar Xrm Xsed Xmv Xawk Xhead Xpwd Xsplit Xcompress XMail Xtouch Xuuencode Xuudecode Xuncompress X" Xpth='. /bin /usr/bin /usr/local/bin /usr/ucb /usr/local /usr/lbin /etc /usr/new /usr/new/bin /usr/nbin /usr/ucb/bin /usr/old' X Xfor file in $loclist; do X xxx=`loc $file $file $pth` X eval $file=$xxx X eval _$file=$xxx X case "$xxx" in X /*) X echo $file is in $xxx. X ;; X *) X echo "I don't know where $file is. I hope it's in everyone's PATH." X ;; X esac Xdone Xcase "$test" in Xtest) X echo "Hopefully test is built into your sh." X ;; X/bin/test) X if sh -c "PATH= test true" >/dev/null 2>&1; then X echo "Using the test built into your sh." X test=test X fi X ;; X*) X test=test X ;; Xesac Xcase "$echo" in Xecho) X echo "Hopefully echo is built into your sh." X ;; X/bin/echo) X echo " " X echo "Checking compatibility between /bin/echo and builtin echo (if Xany)..." $echo $n "hi there$c" >foo1 X echo $n "hi there$c" >foo2 X if cmp foo1 foo2 >/dev/null 2>&1; then X echo "They are compatible. In fact, they may be identical." X else X case "$n" in X '-n') n='' c='\c' ans='\c' ;; X *) n='-n' c='' ans='-n' ;; X esac X cat </dev/null 2>&1 ; then X echo "Your sh handles # comments correctly." X shsharp=true X spitshell=cat X echo " " X echo "Okay, let's see if #! works on this system..." X echo "#!/bin/echo hi" > try X $eunicefix try X chmod +x try X try > today X if $contains hi today >/dev/null 2>&1; then X echo "It does." X sharpbang='#!' X else X echo "#! /bin/echo hi" > try X $eunicefix try X chmod +x try X try > today X if test -s today; then X echo "It does." X sharpbang='#! ' X else X echo "It doesn't." X sharpbang=': use ' X fi X fi Xelse X echo "Your sh doesn't grok # comments--I will strip them later on." X shsharp=false X echo "exec grep -v '^#'" >spitshell X chmod +x spitshell X $eunicefix spitshell X spitshell=`pwd`/spitshell X echo "I presume that if # doesn't work, #! won't work either!" X sharpbang=': use ' Xfi X X X: figure out how to guarantee sh startup Xecho " " Xecho "Checking out how to guarantee sh startup..." Xstartsh=$sharpbang'/bin/sh' Xecho "Let's see if '$startsh' works..." Xcat >try <'README' <<'END_OF_FILE' Xmailsplit sends files and/or directories via e-mail using tar, Xcompress, etc. It sends an awk-script to re-assemble the pieces at the Xreceiving side. It is useful for sending directories and large images Xaround when you don't have ftp, or you have to cross e-mail network Xboundaries. X XTo configure and/or install mailsplit for your site/machine, just run the XConfigure script (i.e.): X X% sh Configure X XIf you're too lazy to read the man pages, here's a quick test. After XConfigure'ing the programs, send yourself a sample directory, for example: X X< Send yourself your bin directory: > X X% mailsplit ./bin X XWatch mailsplit run. In a few minutes, you should receive mail. XCreate a new directory, cd to it, and run joinr on the incoming mail: X X% mkdir delete.me X% cd delete.me X% mail X X X X% joinr p? X X* * * XIt is a Bourne-shell script. Read the over-blown man-page. Send me Xideas for improvements, bug reports. X-Mitch (wyle@ethz.uucp) X X*** Addendum for 2.4 *** XApologies to those of you who received 2.0. I fixed all the bugs which Xcaused it not to work (here). Please be cautious with the enclosed Xversion. I have tested it extensively on a Sun-3/xx under Sun OS 3.4, Xand a vax running Ultrix 2.0, but bugs may have crept in... The Xjoin-script enclosed with each first message now accepts repeats and Xpieces out-of-order. If you don't trust mailsplit 2.4, at least test Xand install this new joiner script. It really helps. X X*** Addendum for 2.5 *** XMore apologies to those of you who received 2.4. I am in the process Xof testing this package along with it's new Configure script (Thanks, XLarry!) on our Cray, a Unisys sys V, Ultrix, Sun OS, a convex, a Xsequent, an alliant, etc. When I'm confident it works on the machines Xhere, I'll release it to usenet. In the meantime, please beat up on it Xand send me bugs, ideas, etc. Configure can configure it to run on XUltrix, Sun OS, Dynix (sequent), convex (alliant), and vanilla sys-V X(Unisys 5000). If it won't run on your machine, please tell me! X XIt now understands "compact" formats, though I think one should stick Xto compress. There are PD implementations of compress. I noticed that Xsome of the machines I tested on lack uuencode/uudecode. There are PD Ximplementations of that as well, but after the Configure, I did NOT test Xmailsplit on those machines which were missing uuencode/decode. X Xcut was replaced by a sed script. I should have known cut was a Sun-ism. X X*** Addendum for 2.6 *** X+ joinr now explicitely exits 0; there was a bug with BSD 4.3. X+ The new, robust joinr script died with old awk (vers 7). I therefore X changed it to be less robust but more portable: pieces after # 10 will X NOT be checked for duplicates and/or out-of-order. I'll probably change it X back after everyone has new awk :-( X+ Added -N flag to NOT send the joinr script (now that it's grown) X+ Added a man page for joinr X+ Added a how-to pargagraph to the front of this file for impatient people :) X+ More support for compact/uncompact X+ Extensive testing on all machines at ETH campus END_OF_FILE if test 3102 -ne `wc -c <'README'`; then echo shar: \"'README'\" unpacked with wrong size! fi # end of 'README' fi if test -f 'README2.7' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'README2.7'\" else echo shar: Extracting \"'README2.7'\" \(2656 characters\) sed "s/^X//" >'README2.7' <<'END_OF_FILE' X XFixes in version 2.7 by HIDEKI YAMANAKA: X X1. Automatic recognition of an uncomprerss program and an decoder. X X Mailsplit stores an archive like X X ... X begin 666 /tmp/1spxxx X ... X X If the file name is prefixed like /tmp/1spxxx.Z or /tmp/1spxxx.C, X we can simply get an appropriate uncompress program. And some people X prefer btoa than uuencode. the header of btoa is like X X ... X xbtoa5 666 /tmp/1spxxx Begin X ... X X So also we can get an appropriate decoder like "awk '/[bB]egin/{print $1}'". X If the result is "begin", then "uudecode'; if it is "xbtoa5", then "btoa -a". X X If the mechanism is included in the joinr, we can change a compress program X and a encoder on the command line but configuration time. X X Lastly, I think the filename "/tmp/1spxxx" is harmful because it is absolute X path name. If some site configures the temporal directory "/usr/tmp", the X site cannot reassembles transmitted files unless /tmp has enough space. X (Even though /usr/tmp is enough space for reassembling the pieces.) X I think the basename is better. X X2. Making more System V compatible. X X Joinr script isn't always compatible with other site. Commands in the script X are specified by its absolute pathname, but there are many System V machines X which are not same the path name. So I think better introduce PATH variable X in the script and strip dirnames of the commands. X X The "tar" program of System V produces verbose messages in the error stream X when specifying -v option and mailsplit is requires another error detection X mechanisms. The "tar" requires -o option when restoring so that the owner and X permission would be appropriate. (Otherwise the owner is others and the X permission is like that. X X System V machines doesn't always have the "compress", instead they have X "pack". (Pack doesn't permit to read data from the stdin, so a slight X change is needed.) X X System V machines doesn't have "Mail", instead they have "mailx". X X3. Abort routine when missing some parts. X X You would think errors will detect later part of the joinr. When joinr X passes an incorrect compressed file, uncompress sometimes produces X huge garbage outputs. Moreover users cannot easily know missing parts X if only an error detection message is shown. X X--- XHideki Yamanaka X X-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= X INTERNATIONAL INSTITUTE FOR X ADVANCED STUDY OF SOCIAL INFORMATION SCIENCE, X FUJITSU LIMITED, JAPAN X X HIDEKI YAMANAKA X XE-MAIL: yamanaka@iias.fujitsu.co.jp or X yamanaka%iias.fujitsu.co.jp@uunet.uu.net (in precise) X-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= END_OF_FILE if test 2656 -ne `wc -c <'README2.7'`; then echo shar: \"'README2.7'\" unpacked with wrong size! fi # end of 'README2.7' fi if test -f 'TODO' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'TODO'\" else echo shar: Extracting \"'TODO'\" \(953 characters\) sed "s/^X//" >'TODO' <<'END_OF_FILE' X X 1. chmod in joinr SHOULD be 777 NOT 0777 X X 2. chmod should NOT have -f flag!! X X 3. change the tar command!!!! don't check for output! X X 4. Send a GENERIC joinr without all the &#@^&^ pathnames!! X X 5. Use compress -d instead of uncompress X X 6. Ask where the mail program is, don't assume /usr/ucb/mail. X X 7. Configure doesn't understand relative directories! Use ~ if available, X cd .. (from UU when asking questions). X X 8. Add the "to-be-installed-dir" to the path, export path! X X 9. Include COMPLETE command used to send the stuff! X X10. When a directory of '.' is provided as the name to send, X perhaps cwd should be in subject line? X X11. Make uuencode BITNET (perhaps use btoa/atob) transparent!!! X***** Add support for NON-uuencode uudecode including atob btoa Xabe!! X***** X X12. Allow user to change the @&!&^# names of mailsplit and joinr!! X X13. Ad wc tests (a la cshar) and crc check counts to pieces and complete X tarchive. END_OF_FILE if test 953 -ne `wc -c <'TODO'`; then echo shar: \"'TODO'\" unpacked with wrong size! fi # end of 'TODO' fi if test -f 'joinr.SH' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'joinr.SH'\" else echo shar: Extracting \"'joinr.SH'\" \(5704 characters\) sed "s/^X//" >'joinr.SH' <<'END_OF_FILE' Xecho "Extracting joinr (with variable substitutions)" X: This section of the file will have variable substitutions done on it. X: Move anything that needs config subs from !NO!SUBS! section to !GROK!THIS!. X: Protect any dollar signs and backticks that you do not want interpreted X: by putting a backslash in front. You may delete these comments. X$spitshell >joinr <>joinr <<'!NO!SUBS!' X############################################################################## X# You might want to change these, but it's not likely. X XME=`$BA $0` # This program's name XVE="2.7 90/03/28" # Version of this program X X############################################################################## X# Hacker city X Xecho " " Xecho "$ME version $VE (C) 1988, 1989, 1990 M F Wyle" 1>&2 Xecho " " X Xecho "...initializing, preparing to extract..." Xif test ! -t 0; then X echo "...found data on standard in; ignoring command line..." Xelse X echo "...standard input empty, looking for files on command line..." X case $# in X 0) case $ME in X joinr) echo " " X echo "Usage: $ME [ < ] file(s)" X echo "Example: $ME < p1of2 p2of2" X echo " " X exit 1 ;; X *) $CAT $0 |/bin/sh $0 X exit 0 ;; X esac ;; X esac X echo "...found $# file(s) on command line, feeding them to $ME" X case $ME in X joinr) $CAT $* | $0 ;; X *) $CAT $0 $* | /bin/sh $0 ;; X esac X exit 0 Xfi X Xcase $ME in X sh) echo " " X echo "Usage: $ME file(s)" X echo "Example: $ME p1of2 p2of2" X echo " " X exit 1 ;; Xesac X X$RM $T1* $T2 $T3* $T4* # get rid of old files X$CAT <<'EOA' > $T4 # create awk extract script XAWK ' XBEGIN { inhex = 0; nparts=0 } X/^---/ { X if ($2 == "start" && $3 == "of" && $5 == "part" && $7 == "of") { X nparts= $8 X fd=proc_files[$6 - 1] X if (fd == 0) fd= $6 X if (fd < 10) file = "T1" fd X else if (fd < 100) file = "T1" "_" fd X else file = "T1" "__" fd X if (proc_files[$6] > 0) { X file = "T2" X } else { X proc_files[$6] = fd X } X inhex = 1 X getline X } X else if ($2 == "end" && $3 == "of" && $5 == "part" && $7 == "of") X inhex = 0 X} X{ if (inhex) print $0 > file } XEND { nerrors=0 X for (i=1; i <= nparts; i++) X if (proc_files[i] == 0) { X print "*** part " i " is missing. ***" X nerrors++ X } X if (nerrors > 0) print "abort" > "T3" X}' XEOA X X# Make awk script use process-specific filenames X$SED -e "s;AWK;$AWK;" -e "s;T1;$T1;" -e "s;T2;$T2;" \ X -e "s;T3;$T3;" < $T4 > $T4"p" X$RM $T4 X$MV $T4"p" $T4 X# Invoke the awk script on std in: Xecho ...extracting data... X. $T4 Xif test -s $T3 ; then X echo "...aborted extraction..." X $RM $T1* $T2 $T3 $T4 X exit 1 Xfi X$CAT $T1* > $T3 # gather pieces Xecho ...finished extraction, starting uudecode... Xheader=`$HEAD $T3 | $AWK '/[bB]egin/{print $1 " " $3}'` Xset $header Xfname=$2 Xcase $1 in X xbtoa5) uudecode=btoa X UUOPT='-a' ;; X begin) uudecode=uudecode X UUOPT= X ;; Xesac XOurDir=`$PWD` Xcd $TD X$touch $T3 Xchmod 777 $T3 X$uudecode $UUOPT $T3 X$RM $T3 Xchmod 777 $fname XIFS=. Xset $fname XIFS=' ' Xcase $2 in X C) uncompress=uncompact X CE=C ;; X z) uncompress=unpack X CE=z ;; X Z) uncompress=uncompress X CE=Z ;; X tar) uncompress=: X $MV $fname $1 ;; X *) $MV $fname $1.$CE ;; Xesac Xfname=$1 X Xecho ...finished uudecode, starting uncompress... X$uncompress $fname.$CE 1>/dev/null 2>&1 X$MV $fname $OurDir Xcd $OurDir Xecho ...finished uncompress, starting tar extraction... X$RM $T1* # tar returns no exit status!! Grrr. X$TA $TAO $fname 2>$T1 # check exit status of tar Xif test -s $T1 ; then Xecho "... >>> Sigh <<< $TA seems to be having problems:" X$CAT $T1 Xecho "...saving $TA archive in $fname." Xelse Xif test ! -f /vmunix ; then # for System V (by yamanaka@iias.fujitsu.co.jp) X $TA tfv $fname Xfi Xecho "...That's all folks!" X$RM $fname Xfi X$RM $T1* $T2* $T3* $T4* Xexit 0 X!NO!SUBS! Xchmod 755 joinr X$eunicefix joinr END_OF_FILE if test 5704 -ne `wc -c <'joinr.SH'`; then echo shar: \"'joinr.SH'\" unpacked with wrong size! fi chmod +x 'joinr.SH' # end of 'joinr.SH' fi if test -f 'mailsplit.SH' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'mailsplit.SH'\" else echo shar: Extracting \"'mailsplit.SH'\" \(10274 characters\) sed "s/^X//" >'mailsplit.SH' <<'END_OF_FILE' Xxfile="mailsplit" Xecho "Extracting $xfile (with variable substitutions)" X: This section of the file will have variable substitutions done on it. X: Move anything that needs config subs from !NO!SUBS! section to !GROK!THIS!. X: Protect any dollar signs and backticks that you do not want interpreted X: by putting a backslash in front. You may delete these comments. X$spitshell >$xfile <>$xfile <<'!NO!SUBS!' X# X# mailsplit: package, transmit, re-assemble unix directories via e-mail X# X# @(#)mailsplit.SH 2.7 90/03/28 M F Wyle, H. Yamanaka X# X# $Id: mailsplit.SH,v 1.8 90/03/19 22:40:14 yamanaka Rel $ X# X# Add -A option and strip dirnames of commands X# 90/01/25 H Yamanaka X X X############################################################################## X# You might want to change these, but it's not likely. X XME=`$BASENAME $0` # Program (my) name XLI=700 # Default number of lines for split-up pieces XCM="$0 $*" # Command line XVE="2.7 90/03/28" # Version of this program XMF="" # Default Mailer flags XIJ="true" # Flag which signals if joinr script should be prepended XTM=false # Default test Mode flag (If set, don't do anything) XT1=$TD/1sp$$ # Default temp file for output of tar, compress XT2=$TD/2sp$$ # Default temp file for uuencoded output XT3=$TD/3sp$$ # Default temp file prefix for split pieces XT4=$TD/4sp$$ # a copy of joinr X############################################################################## X X# You do NOT want to change these variables or their initial values! X XRA=all # Default range of pieces to send XFC=0 # File Count Xne="FL=$FL' '" # Next command in parser XSU="_default_" # Default Subject is first file or dir sent XFL="" # Initialize file list XTO="" # Initialize recipient Xumask 0 # Important for writing reading files created! X X# Human Factors Law Number 4: X# "Upon startup, thou shalt tell the user where he is and how to quit." X# -Ben Schneiderman- X Xecho " " 1>&2 Xecho "$ME version $VE (C) 1988, 1989, 1990 M F Wyle, H. Yamanaka" 1>&2 Xecho " " 1>&2 X X############################################################################## X# Read beyond here at your own risk; you've entered... The hacker zone... X# Parse the command line for recipient, options, flags, files, dirs: X Xfor AR in $* ; do X case $AR in X -C ) ne="CO=" ;; X -D ) ne="TD=" ;; X -M ) ne="MF=" ;; X -N ) IJ="false" ;; X -A ) MS="archive" ;; X -T ) ne="TA=" ;; X -V ) DB=9 ;; X -b ) ne="CB=" ;; X -d ) ne="DE=" ;; X -l ) ne="LI=" ;; X -m ) ne="MA=" ;; X -n ) TM=true ;; X -r ) ne="RA=" ;; X -s ) ne="SU=" ;; X -t ) TM=true ;; X -v ) echo "$ME version $VE by Mitch Wyle" ; exit 0 ;; X -x* ) DB=`echo $AR | cut -c3` ; DB=${DB:-0} ;; X -u ) ne="UU=" ;; X -* ) echo " " ; echo "Bad flag for $ME": $AR X echo "Usage: $ME [] " X echo "Example: $ME werner@utastro.uucp ./casetool" X echo " " X exit 0 ;; X * ) eval $ne"$AR" ; ne='FL=$FL" "' ;; X esac Xdone Xcase $# in X 0 | 1 ) X echo " " X echo "Usage: $ME [] " X echo "Example: $ME werner@utastro.uucp CaseTool" X echo " " X exit 0 ;; Xesac Xset $FL XTO=$1 Xcase $SU in X _default_ ) SU=$2 ;; Xesac X Xcase $DB in X 8|9) set -x ;; Xesac X X# create a copy of joinr in /tmp: Xcat >$T4 <<'JOINR!END' X!NO!SUBS! Xcat ./joinr >>$xfile X$spitshell >>$xfile <<'!NO!SUBS!' XJOINR!END X Xcase $RA in # hacked from usenet commentary; will X all) ;; # be replaced by an awk script RSN X * ) X SIFS="$IFS" X IFS="," X for i in $RA ; do X R=$R" "$i X done X RA=$R X IFS=$SIFS X R="" X for tok in $RA ; do X case $tok in X *-*) X SIFS="$IFS" X IFS='-' X set $tok X IFS="$SIFS" X i=$1 X max=${2:-99} X while test $i -le $max ; do X R=$R" "$i X i=`expr $i + 1` # only exec in the entire parse X done ;; X *) R=$R" "$tok ;; X esac X RA=$R X done ;; Xesac X Xcase $DB in X 5|6|7|8|9) echo "...beginning tar and compress..." ;; Xesac Xset $FL ; shift Xfor i in $* ; do X if test -s $i ; then # Test existence of each file to send X FC=`expr $FC + 1` # Increment file counter X fi Xdone Xif test $FC -gt 0 ; then # Is there something to send? X case $DB in # Yup. continue X 1|2|3|4) X case $TM in X true ) X echo "would have issued command:" X echo "$TA cvf $T1 $* 2> /dev/null" X case $CO in X *compress) X echo "CE=Z; $CO $CF $T1 1>/dev/null 2>&1 " ;; X *compact) X echo "CE=C; $CO $T1 1>/dev/null 2>&1 " ;; X *pack) X echo "CE=z; $CO $CF $T1 1>/dev/null 2>&1 " ;; X *cat) X echo "CE=tar; $MV $T1 $T1.$CE" ;; X *) X echo "$CO $T1 1>/dev/null 2>&1 " ;; X esac ;; X false) $TA cvf $T1 $* 2> /dev/null X case $CO in X *compress) X CE=Z; $CO $CF $T1 1>/dev/null 2>&1 ;; X *compact) X CE=C; $CO $T1 1>/dev/null 2>&1 ;; X *pack) X CE=z; $CO $T1 1>/dev/null 2>&1 ;; X *cat) X CE=tar; $MV $T1 $T1.$CE ;; X *) X $CO $T1 1>/dev/null 2>&1 ;; X esac ;; X * ) echo "Sigh... Programmer error in $ME at line 93." ;; X esac ;; X *) X case $TM in X true ) X echo "would have issued command:" X echo "$TA cvf $T1 $* 2> /dev/null" X case $CO in X *compress) X echo "CE=Z; $CO $CF $T1 1>/dev/null 2>&1 " ;; X *compact) X echo "CE=C; $CO $T1 1>/dev/null 2>&1 " ;; X *pack) X echo "CE=z; $CO $T1 1>/dev/null 2>&1 " ;; X *cat) X echo "CE=tar; $MV $T1 $T1.$CE" ;; X *) X echo "$CO $T1 1>/dev/null 2>&1 " ;; X esac ;; X false) $TA cvf $T1 $* 2> /dev/null X case $CO in X *compress) X CE=Z; $CO $CF $T1 1>/dev/null 2>&1 ;; X *compact) X CE=C; $CO $T1 1>/dev/null 2>&1 ;; X *pack) X CE=z; $CO $T1 1>/dev/null 2>&1 ;; X *cat) X CE=tar; $MV $T1 $T1.$CE ;; X *) X $CO $T1 1>/dev/null 2>&1 ;; X esac ;; X * ) echo "Sigh... Programmer error in $ME at line 153." ;; X esac ;; X esac Xelse # Nope, nothing to send Xcase $DB in X [5-9] ) echo "...whoops! Nothing to send! Punting..." 1>&2 ;; Xesac X exit 1 Xfi Xcase $DB in X 5|6|7|8|9) echo "...finished tar, compress, starting uuencode..." Xesac Xcase $TM in X true ) X echo "would have issued command:" X case $UU in X uuencode|*uuencode) X echo "$UU $T1.$CE `$BASENAME $T1.$CE` > $T2 " ;; X *) X echo "(cd $TD; $UU `$BASENAME $T1.$CE` > $T2) " ;; X esac ;; X false) X case $UU in X uuencode|*uuencode) X $UU $T1.$CE `$BASENAME $T1.$CE` > $T2 ;; X *) X (cd $TD; $UU `$BASENAME $T1.$CE` > $T2) ;; X esac ;; X * ) echo "Sigh... Programmer error in $ME at line 170." ;; Xesac Xcase $DB in X 5|6|7|8|9) echo "...finished uuencode, starting split..." ;; Xesac Xcase $TM in X true ) X echo "would have issued command:" X echo "$SPLIT -$LI $T2 $T3" ;; X false) $SPLIT -$LI $T2 $T3 ; ;; X * ) echo "Sigh... Programmer error in $ME at line 128." ;; Xesac Xcase $TM in X true ) X echo "would have issued commands:" X case $MS in X archive) echo "$CAT > $TO"_n ;; X Subj) echo "$MA -s "$SU - part n of m" $TO" ;; X esac X echo "with n ranging from 1 to m; m cannot be determined." X echo "without running for real." X exit 0 ;; Xesac X Xn=1 Xset $T3* Xfor f do X SE=false X case $RA in X all) SE=true ;; X * ) X for p in $RA ; do X case $p in X $n ) SE=true ;; X esac X done ;; X esac X case $SE in X true) X { X echo " " X case $n in X 1) X if test ! -s $f ; then X case $DB in X [5-9]) echo "...whoops! $SPLIT failed! Punting..." 1>&2 ;; X esac X exit 1 ; X fi X case $IJ in X true) cat $T4 ;; X esac X echo " " ;; X esac X echo " " X echo "This file was packed by mailsplit version:" X echo $VE X echo with command: X echo "$CM" X echo "on: `date`" X echo " " X echo "--- start of $SU part $n of $#" X $CAT $f X echo "--- end of $SU part $n of $#" X echo " " X echo " " X } | X case $MS in X Subj) $MA $MF -s "$SU - part $n of $#" $TO ;; X archive) $CAT > $TO$SPRT$n ;; X *) $MA $MF $TO ;; X esac X case $MS in X archive) echo "...output into $TO$SPRT$n..." ;; X *) case $DB in X [5-9]) echo "...sent part $n of $# to $MA..." X echo "...sleeping $DE seconds (waiting for mailer) ..." ;; X esac X sleep $DE ;; X esac X esac X n=`expr $n + 1` X if test `expr $n % 10` -eq 0; then X SPRT=$SPRT"_" X fi Xdone X$RM -f $T1.$CE $T2 $T3* $T4 X!NO!SUBS! Xchmod 755 $xfile X$eunicefix $xfile END_OF_FILE echo shar: 2 control characters may be missing from \"'mailsplit.SH'\" if test 10274 -ne `wc -c <'mailsplit.SH'`; then echo shar: \"'mailsplit.SH'\" unpacked with wrong size! fi chmod +x 'mailsplit.SH' # end of 'mailsplit.SH' fi if test -f 'undoc' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'undoc'\" else echo shar: Extracting \"'undoc'\" \(6947 characters\) sed "s/^X//" >'undoc' <<'END_OF_FILE' XSince I tore the comments out of the code, and it is not customary to send XSCCS files around the net, I made this auxiliary file of documentation. X XAn alphabetical list of variables, their use and meaning: X XAR = argument XCM = command line XCO = compress program may be arc, etc. XDB = Debug level (Verbosity level) XDE = Delay between pieces XFC = file count XFL = File list (build up file list from params on command line in XFL = any order (see /usr/bin/spell) for a good example of obfuscated, XFL = clever parsing code XLI = the number of lines to split pieces into XMA = mail transport agent XME = the name of the program running XMF = Mail flags to send to transport XR = range of pieces to be re-sent during parse stage XRA = range of pieces to be re-sent if mailers on the way failed XSE = send flag; if true, send the piece. If false, don't XSP = the name of the split program XSU = subject of message (defaults to first file name) XT1 = temporary file containing output of tar and compress. XT2 = temporary file containing uuencoded stuff XT3 = temporary file(s) containing split pieces XTA = tar program (may be shar, tar, cat, etc ;-) XTD = Temporary directory in which temp files should go can be . or /tmp. XTM = Test-mode flag (-n option of make) Don't do anything, just XTM = echo what you would have done if testmode option not on. XTO = recipient(s) of message XUU = the uuencode program (may be replaced with btoa / atob ;-) XVE = version number Xf = file name (piece) during loop which sends pieces Xi = a loop index used during parse of the range of files to be re-sent Xmax= The loop-limit used in parse of range of files to be re-sent Xmax= is assumed to be 99 if no top-of-range is given. Xn = piece number during the loop which sends pieces via mail Xne = next param on command line (after flag is set, next param gets Xne = eval'd to the var set in ne (see above) Xt = a specific token containing a range to be expanded (n-m) Xtok= Token during parse of range of files to be re-sent X X XHistory: X# Coder Date Comments X#------------ -------- --------------------------------------------------- X# M F Wyle 02nov87 Shamelessly stolen from tarmail :-) X# WRU 18feb88 changed variable MAILER, added SPLITTER, X# LINES, USER and a warning about uudecode X# M F Wyle 25mar88 Add tar, extract join-script in 1st msg X# WRU 03apr88 fixed some quoting problems in "echo"-strings X# cosmetic changes: renamed "mailsplit" X# created THISNAME X# fixed some typos, added Email-addresses, example, X# section for suggested improvements X# MFW 05apr88 Fix more quoting problems in echo strings (shell X# variables); added /bin/rm cleanup to join-script X# MFW 21jun88 based on WRU's suggestions: X# 1) add file-name to subject X# 2) verbose explanations X# 3) add parts-order checking! Yippeee! X# MFW 08jul88 Fix uudecode setuid bug X# MFW 12jul88 Fix empty or no files bug, added FILECOUNT X# MFW 17jul88 Added umask 0, delay before mail, check after split X# MFW 19jul88 Major re-write for flags, verbose, sccs control X# MFW 25jul88 Added a range-parser, more flags, some documentation X# Up'd release to 2.0 for distribution X# MFW 29jul88 Fixed 8 separate bugs; upgraded join-script X# to accept pieces out of order and duplicate pieces. X# MFW 04Aug88 Fixed more bugs; added comments, put sys-specific X# stuff at the top for configuring. hacked join X# script because it leaks bugs like an ant-hill. X# X# All the comments from the net about parsing ranges X# won't fix the *@&$#& bugs in join. X# MFW 11Aug88 Fixed joinr. Added -N flag, wrote Configure script, X# tested on a slew of machines X# MFW 18Aug88 made joinr compatible with old awk, added exit X XVerbal description of algorithm: X====== =========== == ========== XWhen the program starts, it initializes some variables, then parses the Xentire command line. It builds up a list of non-flag options in the F Xvariable. The first parameter in F is the e-mail path. All others are Xfiles or directories. X XIf no subject is set via the "-s subject" flag, the subect is assigned Xfrom the first file to send. The three temporary file names (variables XT1, T2, and T3) are then assigned. X XThe range of pieces to be re-sent is parsed. X XDebugging output is echo'd to standard out as appropriate for the Xdebugging level. If the -t or -n option is set, nothing is really Xdone, except printing out what would have happened. X XThe tar, compress, split commands are performed, with appropriate Xverbosity for the debugging level. Certain checks are made along the Xway to verify that the commands did not fail. If one fails, the Xprogram exits with an appropriate failure message. X XThe files created in $T3* are sent via mail one at a time, with Xdelays. The first file sent has a join-script pre-pended. A file Xcounter $n is used to label the subject lines. X XThe temporary files are deleted. X X* * * X XThe join script which comes with each 1st message sent by mailsplit was Xupgraded to accept out-of-order pieces and repeated sequences of Xpieces. It ignores repeats, and glues the pieces together in their Xproper order. These features involve saving each piece in a separate Xfile named by it's part number, and saving it only once. awk(1) can Xaccomplish this task easily. An associative array is formed, Xcontaining all of the pieces seen by the script. If a piece comes Xthrough a second time, the script notices by looking in the associative Xarray. Duplicate pieces are piped into the file /tmp/garbage (perhaps XI should parameterize this name?) and later deleted. Since the first Xpass is now really a sort filter of the pieces, cat(1) is used to push Xthe sorted pieces together into uudecode. X XThe join script in 2.4 sucks rocks. I'm re-writing the @#*@ with Xmore error-checking, comments, configurable variables, and maybe it'll Xwork on somebody else's *@&#&$ machine! Can you say, "Test?" X XNew features in 2.5: X- There is a flag for turning off sending the joinr script. X- The joinr script is a "here" document and is spit into a temp directory X at startup. X- The package comes with a Configure script (a la Larry Wall) which will X configure site-specific commands, parameters and install the sucker! X- sccs blew up. I may move to rcs. X- Expect a better Configure script from metaconfig by 3.0 X XNew features in 2.6: X+ joinr is LESS robust for compatibility with old awk. Expect machine- X specific joinr scripts from an intelligent Config script in 3.0 X XNew features in 2.7: X X-Use a generic subject name "archive" to strip directory names in long X paths. END_OF_FILE if test 6947 -ne `wc -c <'undoc'`; then echo shar: \"'undoc'\" unpacked with wrong size! fi # end of 'undoc' fi if test -f 'joinr.1' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'joinr.1'\" else echo shar: Extracting \"'joinr.1'\" \(1715 characters\) sed "s/^X//" >'joinr.1' <<'END_OF_FILE' X.\" @(#)joinr.1 2.6 88/08/20 M F Wyle X.nr N 24 X.nr D 5 X.TH JOINR 1 88/08/11 X.UC 4 X.SH NAME Xjoinr \- re\-assemble mailsplit pieces into original directory tree X.SH SYNOPSIS X.B joinr X[ \< ] X.B files X.SH DESCRIPTION X.\" --------------------------------------------------------------- X.\" This defines appropriate quote strings for nroff and troff X.ds lq \&" X.ds rq \&" X.if t .ds lq `` X.if t .ds rq '' X.\" Just in case these number registers aren't set yet... X.if \nN==0 .nr N 10 X.if \nD==0 .nr D 5 X.\" --------------------------------------------------------------- X.I Joinr Xis a program to re-build files and or directories which have been sent Xvia mailsplit(1). It uses tar(1), uncompress(1) and awk(1) to Xcombine, extract, and decypher the separate pieces of mail created by Xmailsplit. The script accepts the pieces in X.I any Xorder and allows for duplications. That is, one can send the same pieces Xthrough joinr repeatedly, and send the pieces through in any order. X.LP XIt will accept files on the standard input, so a command of the form: X.br X.sp X\% cat lesson?of? | joinr X.br X.sp Xshould produce the same result as: X.br X.sp X\% joinr lesson?of? X.br X.sp XThe program is configured to run on your system by the Configure script which Xcomes with the mailsplit distribution. However, each mailsplit message sent Xalso prepends this script to the first message sent unless mailsplit is Xexplicitely instructed not to send joinr. X.SH AUTHOR XMitchell F. Wyle (wyle@ethz.uucp) X.SH BUGS X.LP XConfigure doesn't ask about defaults and options X.LP XShould bootstrap itself at the receiving site (as with configure) X.SH SEE ALSO Xsplit(1), tar(1), mail(1), uuencode(1), uudecode(1), compress(1), Xawk(1), sh(1) END_OF_FILE if test 1715 -ne `wc -c <'joinr.1'`; then echo shar: \"'joinr.1'\" unpacked with wrong size! fi # end of 'joinr.1' fi if test -f 'mailsplit.1' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'mailsplit.1'\" else echo shar: Extracting \"'mailsplit.1'\" \(11634 characters\) sed "s/^X//" >'mailsplit.1' <<'END_OF_FILE' X.\" @(#)mailsplit.1 2.6 88/08/20 M F Wyle X.nr N 24 X.nr D 5 X.TH MAILSPLIT 1 88/08/20 X.UC 4 X.SH NAME X.B mailsplit X\- package, split, transmit file directories by electronic mail X.SH SYNOPSIS X.B mailsplit X[ X.B options X] X.B mail-path files X.br X really: X.br X.B mailsplit X[ X.B \-ANVntv X] [ X.B \-C compress program X] [ X.B \-D temporary directory X] [ X.B \-M Mailer options X] [ X.B \-T archive program X] [ X.B \-b compress-bits X] [ X.B \-d delay time X] [ X.B \-l split lines X] [ X.B \-m mail agent X] [ X.B \-r parts X] [ X.B \-s subject X] [ X.B \-xn set debug-level n X] [ X.B \-u ascii encode program X] X.I mail-path[,mailpath] X.I files [files] ... X.SH DESCRIPTION X.\" --------------------------------------------------------------- X.\" This defines appropriate quote strings for nroff and troff X.ds lq \&" X.ds rq \&" X.if t .ds lq `` X.if t .ds rq '' X.\" Just in case these number registers aren't set yet... X.if \nN==0 .nr N 10 X.if \nD==0 .nr D 5 X.\" --------------------------------------------------------------- X.I Mailsplit Xis a program to package, compress, document, transmit, and re-assemble Xmultiple files and directories to other users via electronic mail. XMailsplit uses tar(1), compress(1), split(1) and mail(1) to send the files, Xand awk(1) to re-assemble them at the receiving side. Mailsplit attempts to Xpreserve the integrity, structure, mode, protection, and ownership of the Xtransmitted directories. X.SH OPTIONS XCommand line options can be used to override the default options in the source Xcode of X.I mailsplit. XOptions and flags can be specified in any order on the command line. XThe first non-option or flag token on the command line is interpretted Xto be the mail path whither the data should be sent. All other Xnon-option or flag tokens are considered files or directories to be Xsent. Since the program is a shell-script, users might want to tune Xthe defaults for their sites. All suggestions, improvements, Xbug-reports, etc. would be greatly appreciated. Send them to Xwyle@ethz.uucp. X.IP -N Xtells mailsplit Not to send the joinr script prepended to the first Xmessage. The default is to prepend your version of the joinr script Xwith each mailsplit batch sent. Use this option to suppress sending Xthe joinr script. This option is useful if one is sending directories Xto a recipient who already has the mailsplit package installed on her Xsystem. X.IP -V Xsets the X.I verbose Xoption, and is equivilant to -x9 (maximum debugging). When this option is Xset, X.I mailsplit Xwill echo everything he does while he's doing it. This option will generate Xan awful lot of useless debugging information and is not recommended unless Xyou are debugging. X.IP -n Xputs X.I mailsplit Xinto "test" mode. This mode is analogous to make(1) -n where X.I mailsplit Xtells you which commands he would have issued without actually doing anything. XThis mode is also good for debugging, or when you're curious what mailsplit Xmight do with a complicated path or directory. X.IP -t Xsame as -n X.IP -v Xdisplays version number and exits. This option is useful only for finding out Xwhich version / patch level your mailsplit is. X.IP -A Xwill use the generic name "archive" for the subject instead of a long Xpathname. X.IP -C compress program Xtells X.I mailsplit Xto use the next argument as the program to use instead of compress(1) Xfor compressing the archive stream. In some circumstances, using Xcompress(1) may expand the size of a stream, in which case one would Xmight want to use cat(1) to send the archive through unchanged. The author is Xdubious of this option. X.IP -D directory Xinstructs X.I mailsplit Xto use X.I directory Xinstead of /tmp for temporary files created during the archival, compression, Xand encoding process. This option is necessary if there is very little space Xin /tmp, one is sending very large directories, or the file tree on this Xsystem is bizare. X.IP -M mailer options XWhen this option is used, X.I mailsplit Xpasses the X.I mailer options Xargument(s) to the mail program. Some mailers need options to get the job Xdone (we won't mention an infamous X.400 mailer). Other times, one would Xprefer to set options for efficiency, tell the mailer not to run Xinteractively, etc. To set more than one option, enclose X.I mailer options Xin quotes. Be careful about the shell's eating quotes. One might have to Xexperiment a bit to set all the options needed or desired. X.IP -T archive program XIf this option is set, X.I mailsplit Xuses X.I archive program Xinstead of tar(1) to archive all the files and directories. If the system Xdoes not have tar(1), it is unlikely that X.I mailsplit Xwill work. On the other hand, tar(1) expands binary files enormously, so one Xmight substitute cat(1) instead of tar(1) to reduce the size of the archive. X.IP -b compress-bits Xuses X.I b Xbits for compression. Some e-mail reachable users have bizare XLempel-Ziv implementations or flakey hardware which requires fewer than Xthe BSD default 16 bits for the common compression string code. The Xdefault value of this parameter is set to 12 when mailsplit is Xdistributed, which seems to be pretty universal. If one knows that the Xdata are being sent to a BSD site or a site with 16-bit compression, Xone should change the default in the code to 16. X.IP -d seconds Xtells X.I mailsplit Xto wait X.I seconds Xseconds after each time he invokes the mailer. Some mailers (a certain XX.400 system with "MTA congestion problems" will remain nameless) have Xproblems dealing with sudden, large bursts of messages. One might want Xto set the default in the code to some non-zero value if the mailer on Xhis system has such problems. If X.I mailsplit Xis usually used for small directories, and the default delay is small, Xone should use this flag to set the delay higher if a large number of files or Xlarge directory is sent. X.IP -l lines XThis option sets the maximum number of lines for a single mail Xmessage. The default as distributed is set to keep a message smaller Xthan 60 Kilobytes. See the discussion below. One might want to set Xthe default higher, and use this option to set a smaller number when Xsending mail through mailers which don't like large messages. (A Xcertain X.400 mailer...) Remember that if the archive pieces have to pass X.I through Xa mailer which wants small messages, the lines size must be set smaller. X.IP -m mailer XIf the system on which X.I mailsplit Xis running has more than one mailer (user agent) one can use this flag Xto choose a mailer other than the default (/usr/ucb/Mail as Xdistributed). X.IP -r parts XThe -r(esend) option is used to re-send pieces which became lost in the Xwide-area mail networks. X.I parts Xis a list of pieces to be re-sent to the recipient. X.I Parts Xcan be a comma-separated list, or numbers between dashes, or just one number Xand a dash, or any combination. The formal BNF of X.I parts Xis: X.nf X.br X Xparts :== X | number,parts X | number-number X | number- X Xnumber:== X | X Xdigit :== 0|1|2|3|4|5|6|7|8|9 X X.br X.fi XOnly those pieces in X.I parts Xwill be sent. This option is useful for re-transmitting part 3 of 27 instead Xof having to re-transmit all 27 pieces. One must be sure that the other Xoptions are set X.I exactly the same Xas the first transmission, including line-length, etc. X.IP -s subject Xsets the subject to X.I subject Xinstead of using the first file argument for the subject of each mail message Xgenerated. X.IP -xn Xsets the debugging level to n, where n is a number from 0 to 9. The higher Xthe number, the more debug output. Setting n to 0 causes X.I mailsplit Xto work silently. No output will be sent to standard out, and only failure Xmessages will be sent to standard error. Setting n to 9 is equivilant to -V X(verbose), and gives maximum debug output. X.IP u encoding program XThis option sets the encoding program to something other than Xuuencode(1c). The encoding program must be able to take 8-bit binary Xand produce 7-bit ASCII data. X.SH DISCUSSION X.PP XA few X.I de X.I facto Xstandards have evolved on wide-area (e-mail) networks to guarantee the safe Xtransmission of information: X.sp X.IP - XThe information must be text-only, i.e. 7-bit ASCII X.IP - XLines of text may not exceed 80 characters. A message must therefore Xcontain an end-of-line character at least after every 80 characters. X.IP - XA single message may not exceed 64 Kilobytes. X.IP - XLines of text starting with the word "From" will be translated Xto ">From." X.IP - XBlank lines at the end of a mail message are deleted. X.IP - XA message containing a period "." on a line by itself will be terminated Xabnormally. X.PP XUpon invocation, the program parses all flags and options, then looks Xfor at least two parameters on the command line: The e-mail path(s) of Xthe receiver(s), and the file or directory name(s) to send. If one or Xboth of these parameters is missing, a help message is displayed with a Xshort "usage" description and an example. X.PP XThe directories or files are recursively archived and compressed into a Xtemporary binary file. The binary file is then sent through uuencode Xso that it is encoded into 7-bit ASCII with end-of-lines every 79 Xcharacters. The resulting ASCII file is then split into several Xdifferent files, each of which is smaller than 64 Kilobytes. X.PP XThe first file is prepended with instructions as well as the program Xneeded to unpackage the transmitted message or messages. All messages Xare pre- and post-pended with special start and end text, to assist both Xthe person receiving the message and the unpackaging program. Finally, Xeach message is sent via electronic mail to the recipient, and all Xtemporary files are deleted. X.PP XDuring these operations, status information is displayed after each Xsuccessful step. If there are no files to send, an error message is Xdisplayed. X.PP XAt the receiving end, a first-time user must extract the unpackaging Xprogram from the first e-mail message, and make the program available Xfor his use. A user must perform this step only once, as the packaging Xformat should not change. Depending on the size of the transmitted files Xor directories, the receiver may get one or several mail messages. The Xmessages contain subject headers of the form: X.br X.\" .nf X.sp X"--- start of part of " X.sp X.br Xwhere is the first file or directory name, is the total Xnumber of messages into which the package was split, and is in the Xrange 1 to m. X.PP XThe receiver must then send all of the parts through the unpackaging Xprogram in their proper order. This is accomplished either from within Xthe mail reading program, or by saving the messages into a file and Xinvoking the unpackaging program on the saved file. Because e-mail Xsystems will often deliver messages out of order, the re-assembler Xprogram detects and corrects when pieces are sent through it repeatedly Xor out of order. Only the first part number is used; all others are Xignored. X.PP XThe re-assembler program first filters out repeated pieces and sorts Xthem into their correct order before it joins them together to decode, Xuncompress, and extract the individual files or directories. X.SH EXAMPLES X.LP X% mailsplit wyle@ethz.uucp /etc/hosts X.LP X% mailsplit wyle@ethz.uucp sources -r 5,7,17-21 X.LP X% mailsplit -l 700 -x9 wyle@ethz.uucp goodies -s "examples" X.LP X% mailsplit -l 700 -x9 -r 19- wyle@ethz.uucp goodies X X.SH AUTHOR XMitchell F. Wyle (wyle@ethz.uucp) X.SH BUGS X.LP XBSD specific? X.LP XRe-assembler should accept pieces out of order X.LP XShould be packaged with a configurer, makefile, install script X.SH "SEE ALSO" Xsplit(1), tar(1), mail(1), uuencode(1), uudecode(1), compress(1), Xawk(1), sh(1) END_OF_FILE if test 11634 -ne `wc -c <'mailsplit.1'`; then echo shar: \"'mailsplit.1'\" unpacked with wrong size! fi # end of 'mailsplit.1' fi echo shar: End of shell archive. exit 0