Shell Programming and Scripting

This for instance

Code:

# find /usr/local -type f|while read f;do echo ${f##*/};done | head 
a2p.1
c2ph.1
pl2pm.1
perlthrtut.1
perltooc.1
xsubpp.1
perlintro.1
perlmodlib.1
perlartistic.1
perlfaq.1

works well (but please omit the pipe to head which I only added for brevity's sake)

Ohh, I got it

When I open the log file by windows, it appears like:
aaa.htmlbbb.html

but when I open it by vi
it is like
aa.html
bbb.html

Could it be you have a strange record separator defined,
or do your filenames contain whitespace or similar odd characters?

Oh, sorry I overlooked this
That's natural because Windows text files use the sequence of \r\n
to denote an end of line,
while Unix only uses a single newline \n
For Windows usage you need to convert your file.
Either look for tools like unix2dos or use some Perl like

Code:

$ perl -i.unix -npe 's/\n$/\r\n/' log.txt

Please note, that you may well omit the extension .unix in the above Perl command.
This has nothing to do with our conversion (I admit, badly chosen name).
It will only hold a backup copy of the original file with the extension .unix.
If you omit it no backup copy will be kept.

thanks buffonix

If you don't intend to edit the file on Unix at all
then you can squeeze in the extra carriage returns in situ in your loop like so

Code:

# find /usr/local -type f|while read f;do printf "%s\r\n" ${f##*/};done > log.txt