Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!uwmcsd1!lakesys!dougm
From: dougm@lakesys.UUCP (Doug McIntyre)
Newsgroups: comp.emacs
Subject: Problem in Micro Emacs v3.9e
Keywords: MicroEmacs
Message-ID: <408@lakesys.UUCP>
Date: 19 Feb 89 02:49:00 GMT
Organization: Lake Systems - Milwaukee, Wisconsin
Lines: 44


	I am having a problem with MicroEmacs v3.9e, and was wondering what
I'm doing wrong. I am writing an unshar utility in emacs, and ran accross a
new format that I had to redo the one I already have. What I am trying to 
do is pick up the first word from a line of text read from a buffer...
Here is what I have so far.... Basically a lot of simple stuff to make sure
that everything is working correctly..
------------------------------
set %sharsep "SHAR_EOF"
beginning-of-file
search-forward %sharsep
!while &not &seq $status FALSE          ; looking for a line like....
        beginning-of-line
        search-forward ">"              ; sed >command.h <<SHAR_EOF
        set %shbuf $cbufname
        set %nametemp #$cbufname        ; and %name is set to command.h+++
        set %pos &sindex %nametemp " "  ; here we find space after command.h
        set %pos &sub %pos 1            ; adjust to # of chars in filename
        set %name &left %nametemp %pos  ; grab %pos characters from %name
        next-line                       ; go to beginning of shar file..
        beginning-of-line
        set-mark                        ; copy from here...
        search-forward %sharsep
        beginning-of-line
        copy-region                     ; to the char right before SHAR_EOF
        select-buffer %name             ; Then go to buffer command.h
        change-file-name %name          ; set the file name to command.h
        yank                            ; and yank it in..
        select-buffer %shbuf            ; go back to shar file..
        write-message &cat "Done with file " %name
        search-forward %sharsep         ; finds SHAR_EOF of file just copied
        search-forward %sharsep         ; finds begining of next...
!endwhile
---------

Well everything works just fine, but the %name variable turns up garbage,
while I'm expecting to copy the left (%pos)most characters..
and so I get garbage buffer names and files, which really can't be saved off
after editing..
-- 
------------------------------------------------------------------------------
UUCP: uunet!marque!lakesys!dougm			Compuserve: 70611,2215
INET: dougm@lakesys.lakesys.COM					 APLE: DougMac
------------------------------------------------------------------------------