I have a text file which has the following line:
/u/tux/abc/spool/frtbla_mcdetc_0000149099_20101126135009990_1.tifI want to extract frtbla_mcdetc_0000149099_20101126135009990_1.tif;
the word after the last slash (/).
5 Answers
If you are in a Linux-like environment you can use the basename utility:
basename $(<your_file)use this command
cat text_file_name | cut -d '/' -f 6If you know the exact format of the directory structure and it won't alter, you can use cut:
$ cut -f6 -d '/' file.txtHere uses cut to treat the directory separators as a delimiter and extract the 6th one.
If instead all you know is it is the last part of a line but don't know the directory structure, you can use rev as well:
$ rev file.txt | cut -f1 -d '/' | revHere the file is reversed and the first field is extracted, before being reversed again.
The following applies to all strings in a shell, not just filenames, and is far easier than cut because you don't need to know how many fields there are before the one you want:
$ foo=/path/to/file/split/by/slashes.txt
$ echo ${foo##*/}
slashes.txtThis uses the 'greedy trim', i.e. trim everything until the last '/' as described here:
${foo <-- from variable foo ## <-- greedy front trim * <-- matches anything / <-- until the last '/' }I don't know what the standard shell is in Aix, but bash is available and supports edited expansion of variables.
If your full name is in the variable FileName, then ${FileName##*/} displays the name with all leading characters deleted as far as the last /; by contrast ${FileName#*/} deletes up to the first /, while ${FileName%/*} deletes trailing characters from the last / (ie the directory path).
If you generate the file name(s) by a find command, then you need a command like:
find ... | while read FileName; do echo ${FileName##*/}; doneIf in a text file:
while read FileName; do echo ${FileName##*/}; done < FileList.txtReplace the echo command by whatever processing you need to do with the name.
