How to run a command only if a specific file has a certain size

Question

How can I execute a command only if a certain file exceeds a defined size? Both should at the end run as a oneliner in crontab.

Pseudocode:

* * * * * find /cache/myfile.csv -size +5G && echo "file is > 5GB"

Answer 1

If you have GNU stat, you can use its --printf option to get its size.

e.g.

size=$(stat --printf '%s' /cache/myfile.csv)
if [ "$size" -gt 5368709120 ] ; then  # 5 GiB = 5 * 1024 * 1024 * 1024
  echo "file is > 5GB"
fi

See man stat for details.

---

BSD's stat (e.g. on FreeBSD and on Mac) has a similar formatting option, -f:

size=$(stat -f '%z' /cache/myfile.csv)

---

Alternatively, you could use perl's built-in stat function, or its -s file test operator (which is similar to bash's -s file test but it returns the file's size rather than just true if it exists and is non-empty). perl's stat function returns a 13-element list (array) of metadata about a file containing the following data (copied from perldoc -f stat):

[...] Not all fields are supported on all filesystem types. Here are
the meanings of the fields: 

  0 dev      device number of filesystem
  1 ino      inode number
  2 mode     file mode  (type and permissions)
  3 nlink    number of (hard) links to the file 
  4 uid      numeric user ID of file's owner
  5 gid      numeric group ID of file's owner
  6 rdev     the device identifier (special files only) 
  7 size     total size of file, in bytes
  8 atime    last access time in seconds since the epoch
  9 mtime    last modify time in seconds since the epoch
 10 ctime    inode change time in seconds since the epoch (*)
 11 blksize  preferred I/O size in bytes for interacting with the
             file (may vary from file to file)
 12 blocks   actual number of system-specific blocks allocated
             on disk (often, but not always, 512 bytes each) 

(The epoch was at 00:00 January 1, 1970 GMT.)

Field 7 is the one we need.

To return the file's size (for later use in a shell command or script) using stat:

# stat
perl -e 'print scalar((stat(shift))[7])' /cache/myfile.csv

# -s
perl -e 'print -s shift' /cache/myfile.csv

Or to do it all in perl:

# stat
perl -e 'print "File is > 5 GiB\n" if (stat(shift))[7] > 5*1024*1024*1024' /cache/myfile.csv

# -s
perl -e 'print "File is > 5 GiB\n" if -s shift > 5*1024*1024*1024' /cache/myfile.csv

See perldoc -f stat and perldoc -f -X (as well as help test in bash).

BTW, perl's shift function removes the first element of an array (by default @ARGV, the array of command line args, if not specified) and returns its value. It's often used in a loop to process all elements of an array, but here we're only interested in the first arg (the filename). See perldoc -f shift for details, including notes on lexical scope and use in a subroutine.

Answer 2

To use the file size as a precondition you can use stat or find:

[ -n "$(find /cache/myfile.csv -prune -size +5G 2>/dev/null)" ] && echo "file is > 5GB"

Or if the target command (echo, here) is short, put it into the exec part of `find

find /cache/myfile.csv -prune -size +5G -exec echo "file is > 5GB" \;

The -prune is in case myfile.csv might be a file of type directory, to prevent find from descending into it.

Answer 3

If you need to treat files in a shell, both version only execute shell's command only if all conditions are met: is a file, is named myfile.csv and is > 5G:

find /cache -name 'myfile.csv' -type f -size +5G -exec bash -c '
    echo "$1 is > 5GB"
' bash {} \;

or

find /cache -name 'myfile.csv' -type f -size +5G -exec bash -c '
    for file; do echo "$file is > 5GB"; done
' bash {} +

Answer 4

Note that some shells have the feature built-in.

SHELL=/bin/tcsh
* * * * * if (-Z /cache/myfile.csv > 5*1024*1024*1024) echo 'file is > 5GiB'

Or with zsh, here using glob qualifiers and an anonymous functions, though zsh also has a stat builtin that predates both GNU and BSD stat:

SHELL=/bin/zsh
* * * * * (){ if (($#)) echo 'file is > 5GiB'; } /cache/myfile.csv(NLG+5)

(note that like for find -size +5G, we're talking of gibibytes (1GiB = 1,073,741,824 bytes) here, not gigabytes (1GB = 1,000,000,000 bytes))

For symlinks, tcsh will get the size of the file it eventually resolved to while zsh's LG+5 qualifier like find's -size will check the size of symlink itself. Change to -LG+5 to check the size after symlink resolution. zsh's stat builtin gives you information after symlink resolution by default, -L to change that. In GNU and BSD stat, that's reversed. Same with find where -L tells it to follow symlinks.

For more ways to get the size of a file, see How can I get the size of a file in a bash script?