2005/11/15, Gabriel Briones <[EMAIL PROTECTED]>: > sed can do the trick > > sed -e 's/foo/bar/g' file.txt > new_file.txt > > the above command will replace all the existence of foo with bar on > file.txt and will redirect the modified output to new_file.txt
Well the OP did say hundreds of files, so a little more work is necessary. For the GNU version of sed (which should be available in all GNU/Linux distributions but probably not in OSX) I'd do something like: find /Directory/Subdirectory -name "*.txt" \ | xargs sed --in-place=.bak 's/foo/bar/g' With the proper options, the "find" command will create a list of files you want to change (e.g. only *.txt or *.html) and pass (or pipe) this list to "sed" via the special utility "xargs". The "--in-place" or "-i" option tells sed to change the file itself rather than just have the changes output to the screen or redirected to a file. The added "=.bak" will create a backup of the changed file(s) with the extension .bak. (Note: the "--in-place" may not be available in all versions of GNU sed. Also, you could replace xargs with the built-in "-exec" or "-execdir" options of find) > On 11/15/05, Che Sosa <[EMAIL PROTECTED]> wrote: > > Mabuhay PLUG members, > > > > I have a problem of parsing text. What I want is to parse certain > > sequences from hundreds of files and replace them with a certain > > pattern of text. > > > > Is this doable with sh? What tools/programming languages is suitable for > > this? _________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List [email protected] (#PLUG @ irc.free.net.ph) Read the Guidelines: http://linux.org.ph/lists Searchable Archives: http://archives.free.net.ph
