File extension statistics ?

[Vilsen]

Is it possible to write something in the terminal in some way

so that you get a list of all different file extensions that a hard drive contains

and how many of each kind ??

Example of possible result:

*.jpg 123456 files total

*.gif 54321 files total

and so on...

[Flemur]

Code: Select all

$ find -iname "*.jpg" |  wc
  15101   15147  696695

15101 is the number of .jpg files. Dunno what the others are, try

Code: Select all

man wc

[MrEen]

Code: Select all

$ find -iname "*.jpg" |  wc
  15101   15147  696695

15101 is the number of .jpg files. Dunno what the others are, try

Code: Select all

man wc

My understanding:
15101 is the number of lines (the total number of matching filenames)
15147 is the number of words (you have a few jpegs with spaces in their names)
696695 is the number of characters (bytes) returned

[Flemur]

My understanding:
15101 is the number of lines (the total number of matching filenames)
15147 is the number of words (you have a few jpegs with spaces in their names)
696695 is the number of characters (bytes) returned

Sounds right - I never knowingly use delimiters in file names, so got curious and it turned out they were in a "Deep Art Effects" directory copied from a phone.

Edit: OP, notes on this comamnd:
find -iname "*.jpg" | wc

"iname" = find by name, i = ignore case (finds JPG, jPG, etc)
"*.jpg" = files which end in .jpg. Use the quotes so "*.jpg" doesn't get expanded, e.g. if you ran

Code: Select all

find -iname *.jpg 

in a directory with fred.jpg, it would expand to:

Code: Select all

find -iname fred.jpg

and find files named fred.jpg.

To see the difference, go to a directory with some jpg files and enter

Code: Select all

echo *.jpg
echo "*.jpg"

[xenopeek]

File extensions don't mean much in Linux, but sure.

Perhaps a command like this:
find . -type f -printf '%f\n' | grep '^[^.].*\.' | awk -F. '{print $NF}' | sort | uniq -c | sort -n

This finds all files in the current directory and prints their name, ignores any files that start with a dot, prints the extensions of the remaining files, sorts those extensions, counts each unique extensions and finally sorts on the results such that the extensions are shown from lease frequent to most frequent.

Running that command from your home directory shouldn't take seconds. Running it on your entire filesystem (from / directory) will take a bit.

[Vilsen]

I am deeply impressed! I come closer and closer to a "salvation"
and I realize with fear that so many others on this earth have been so misled by this troll that Pyttemjuk sold into us.

Perhaps one can now go one step to find a program or otherwise use the time / date

information contained in the EXIF data contained in each image (JPEG) (RAW)

and use this date to sort images into Chronological order ? Someone who knows something about this ???

[Vilsen]

[SOLVED]