Yup this works, though you can't save the output to the original file (it blanks it). So I first copy the original to a backup, and then run the backup through sed and overwrite the original (then delete the copy). I can wrap all this in a batch script to fix the entire directory.
I have a book on awk (that was another idea of what to use here) and probably will order that camel book. (Perl is usefull for CGI also, which is something else I want to learn.). --- [EMAIL PROTECTED] wrote: > > I this case the '\'s that appear in #include > statements are the ONLY > > ones that need to be changed, so I can look for > #include. > > Good. So you need to change `\` to `/` in lines > that start with > '#include': > > cat file.c |sed -e '/^[ \t]*#include/ s,\\,/,g' > >outfile.c > > That I think should do it. If it fails, at least > you should be on the > right track. > > (Basically it means "On every line that starts with > optional whitespace > followed by an include statement, substitute all > instances of > backslashes with slashes".) > > And for homework, study regular expressions (man > grep) and sed (man > sed). As someone said, you could also try learning > perl. Actually, vi > or emacs's search and replace facilty can also do > this quite easily. > > Luck porting your program! > === Amateur Radio, when all else fails! http://www.qsl.net/wa2mze Debian Gnu Linux, Live Free or ..... _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com
