I've had some issues with my email so
I'm just now seeing your response.
Thanks for taking the time to write
up the shell script. Looks like
a nice way to do it. I guess it
boils down to whether its possible to
determine the type of a file from its
name. Though as I think about it,
I could replace your if statement
with a call to the UNIX "file" command and then it
wouldn't be name dependent. But
it would be dependent on how
good a job the "file" command
does and I've seen it get things wrong.
Date: Mon, 7 Jun 2004 17:17:10 -0400
Subject: RE: newbie question on initializing repository
In this case, I'd import a blank project, and then use "cvs add"
script to add the
files, using the -kb option accordingly. Here's a c-shell pseudo-code
# Assume *.h and *.cpp files are text files, and the rest are binary
for f in `find .`; then
if [ -f $f ]; then
# found a file, test if binary
if [ $f == *.h || $f == *.cpp ]; then
# add binary file
cvs add -kb $f;
# add text file
cvs add $f;
if [ -d $f ]; then
# add directory
cvs add $f;
You get the idea. Gurus may have a better solution though....
Behalf Of Bret A Cooper
Sent: Monday, June 07, 2004 4:47 PM
Subject: newbie question on initializing repository
This is a first time user question.
We have collected all of our source files within a directory
structure and are ready to initialize the cvs
repository. There are a couple thousand files. Roughly 20%
binary type files and are scattered
around throughout various directories.
My original plan was to run the cvs init command from the
topmost directory and be done. However,
I now believe its not that simple, as the binary files(only)
must be initialized with the -kb option.
Is that correct? If so, any suggestions on the simplest/easiest
to go about this?