The name of the INDEX must be specified. A new index will be created if it does not exist already. Note that the index is read into memory for each operation: collate operations on multiple files into one call if possible. No performance tests have been run, so there is no advice as to how large the index can realisticly be. It is currently used to index a collection of 10,000 files.
addan item to the index
removean item from the index
findmatching items in the index
-lEach of the FILEs listed contains a list of files to process.
-qScreen display should be limited to essential information.
$ ./dup index add *.cpp Unable to read the digest index, will create one Added dup.cpp Added test_md5.cpp
$ cp dup.cpp dup.dup
$ ./dup index find dup.dup Found duplicate: *dup.dup dup.cpp
$ ./dup index remove *.cpp Removed dup.cpp Removed test_md5.cpp