CODEWITHSUNDEEP
linuxbashunixrenamebatch-processing
user2643338
user2643338

Reputation: 23

How to batch-rename files by date?

What commands could I use to rename a few hundred files that are currently named 

file.2003-01-02.txt
file.2003-01-04.txt

... and so on? I would like them to be named:

file_y2003m01d02.txt
file_y2003m01d04.txt

... etc.

In other words, file.2007-12-09.txt would become file_y2007m12d09. Is there a simple set of commands to do this?

Similarly I have another problem, where files are named file_y2003m02d01_grid.txt. I would like to know how to remove _grid from each filename, so that it matches the format I proposed above.

Upvotes: 2

Views: 1014

Answers (2)

seanmcl
seanmcl

Reputation: 9946

You can also use sed:

for example:

ls | while read f; do echo "mv $f $(echo $f | sed 's/\./_y/;s/-/m/;s/-/d/')"; done

This will show you the commands that bash will run. To actually do the move, remove the echo and quotes:

ls | while read f; do mv $f $(echo $f | sed 's/\./_y/;s/-/m/;s/-/d/'); done

Upvotes: 0

schesis
schesis

Reputation: 59118

You can use the rename command:

rename 's/^file\.([0-9]{4})-([0-9]{2})-([0-9]{2})\.txt$/file_y$1m$2d$3.txt/' *

This uses Perl regular expression substitution to transform filenames. The command above says:

  1. Find files starting ^ with file. (the . has to be escaped, otherwise it matches any character), followed by the captured () group [0-9]{4} (a digit, 4 times), then -, then another captured group of a digit twice, etc., and ending $ with .txt;

  2. Then, rename those files to file_y followed by the first captured group $1, followed by m, followed by the second captured group $2, etc., and ending with .txt.

You should also be able to work out how to use the same command to solve your second problem, with what you no know about how rename works.

Upvotes: 3

Related Questions