removing extension from file without knowing it

I know how to remove the extension of a file, when I know it as:

nameis=$(basename $dataset .csv)

but I want to remove any extension without kno开发者_开发百科wing it beforehand, anyone know how to do this?

Any help appreciated, Ted

In bash you can do the following:

nameis=${dataset%.*}

... e.g.:

$ dataset=foo.txt
$ nameis=${dataset%.*}
$ echo $nameis
foo

This syntax is described in the bash man page as:

${parameter%word}

${parameter%%word}

Remove matching suffix pattern. The word is expanded to produce a pattern just as in pathname expansion. If the pattern matches a trailing portion of the expanded value of parameter, then the result of the expansion is the expanded value of parameter with the shortest matching pattern (the "%" case) or the longest matching pattern (the "%%" case) deleted. If parameter is @ or *, the pattern removal operation is applied to each positional parameter in turn, and the expansion is the resultant list. If parameter is an array variable subscripted with @ or *, the pattern removal operation is applied to each member of the array in turn, and the expansion is the resultant list.

Now, if you want some stylish old-school regexp:

echo "foo.bar.tar.gz" | sed "s/^\(.*\)\..*$/\1/"

-> Shall return: foo.bar.tar

I will break it down: 
s/   Substitute
^      From the beginning
\(     Mark
.*     Everything (greedy way)
\)     Stop Marking (the string marked goes to buffer 1)
\.     until a "." (which will be the last dot, because of the greedy selection)
.*     select everything (this is the extension that will be discarded)
$      until the end
/    With (substitute)
\1      The buffer 1 marked above (which is the filename before the last dot(.)
/    End

Cristiano Savino

You could remove everything following the last dot, if any, with sed:

nameis=$(echo $filename | sed 's/\.[^.]*$//')

But that would not work on files with double extensions such as .tar.gz.

Something like ${dataset%.*} might work; beware of files without extensions, though, as it will look for a dot that's not part of an extension to chop off.

nameis=${dataset%%.*} Assumes none of your basenames have a dot in it. This returns the longest possible dotless string from the left.

I would suggest instead of basename use sed command for that, like this:

echo "$fileName" | sed 's/\..*$//g'