Number of files in directory

bullium · 01-06-2005, 12:18 PM

Here is what I have so fare. It works but the only problem is that the count is not correct, it is also counting the directories in the total. I'm trying to get a total count of my MP3 files from My Music and all its subdirectories. Here is the directory structure.

My Music
| |_> Artist
| |_> Album
| |_> songname1.mp3
| |_> songname2.mp3
|__> Artist 2
|_> Album2
|_> songname1.mp3

Code:

ls -1R | wc -l

(note the "-1" option to 'ls' is hyphen-one, NOT lower-case L, which is
what the option to 'wc' is).

All of my music is in MP3 format so a *.mp3 would work I just don't know how to ad it into the command.

Valindar · 01-06-2005, 12:27 PM

Use a regular expression with grep:

Code:

ls -1R | grep .*.mp3 | wc -l

list all recursively one per line, piped through grep which will output only: any amount of characters, then a dot character then 'mp3' then piped through wc -l to count it.

Edit: Just adding, grep is a very, very useful tool, you might want to read the man page and learn to use it (this is presuming you don't know how to). Grep, awk, sed etc. are very powerful allies when you're working in the CLI!
Check out http://tldp.org for the beginner's guide to bash - amongst teaching you bash basics it will also teach you regexps, grep, awk and sed.

bullium · 01-06-2005, 12:37 PM

Quote:

Originally posted by Valindar
Use a regular expression with grep:

Code:

ls -1R | grep .*.mp3 | wc -l

list all recursively one per line, piped through grep which will output only: any amount of characters, then a dot character then 'mp3' then piped through wc -l to count it.

Edit: Just adding, grep is a very, very useful tool, you might want to read the man page and learn to use it (this is presuming you don't know how to). Grep, awk, sed etc. are very powerful allies when you're working in the CLI!
Check out http://tldp.org for the beginner's guide to bash - amongst teaching you bash basics it will also teach you regexps, grep, awk and sed.

Thanks worked like a charm! I'm still working on learning all of the | in and outs. How are you a Newbie

!

Valindar · 01-06-2005, 12:43 PM

I'm not a newbie - I just haven't posted much

bullium · 01-06-2005, 01:26 PM

Quote:

Originally posted by Valindar
I'm not a newbie - I just haven't posted much

For whoever reads this post and it helps them, I noticed that I had a few mp3's that weren't lowercase hints "MP3" "mp3" "Mp3". Any way I added another argument to grep to also grab those. Here it is.

Code:

ls -1R | grep -i .*.mp3 | wc -l

Have fun, and thanks again Valindar

gqchynaboy · 09-10-2009, 12:40 PM

Code:

ls -1R | grep -i .*.mp3 | wc -l

Now how would you store the count in a bash script?

lutusp · 09-10-2009, 12:56 PM

Quote:

Originally Posted by gqchynaboy

Code:

ls -1R | grep -i .*.mp3 | wc -l

Now how would you store the count in a bash script?

It depends on what you mean by "store." Example:

#/bin/sh

total=`ls -1R | grep -i .*.mp3 | wc -l`

echo $total

gqchynaboy · 09-10-2009, 01:02 PM

Quote:

Originally Posted by lutusp

It depends on what you mean by "store." Example:

#/bin/sh

total=`ls -1R | grep -i .*.mp3 | wc -l`

echo $total

I want to use the count in a loop, what you have above seems to store the string comand in $total.

lutusp · 09-10-2009, 01:22 PM

Quote:

Originally Posted by gqchynaboy

I want to use the count in a loop, what you have above seems to store the string comand in $total.

What kind of loop? A loop controlled by the filenames, or by the count? Here's a loop controlled by a number:

Code:

n=15

while [ $n -gt 0 ]
do
   echo $n
   let n--
done

Perhaps a few more details about what you want to do?

gqchynaboy · 09-10-2009, 01:39 PM

Quote:

Originally Posted by lutusp

What kind of loop? A loop controlled by the filenames, or by the count? Here's a loop controlled by a number:

Code:

n=15

while [ $n -gt 0 ]
do
   echo $n
   let n--
done

Perhaps a few more details about what you want to do?

I want to get the count of *.html files in a directory. And then use that count to iterate through all the .html files to perform a sed operation.

Thanks!

lutusp · 09-10-2009, 08:27 PM

Quote:

Originally Posted by gqchynaboy

I want to get the count of *.html files in a directory. And then use that count to iterate through all the .html files to perform a sed operation.

Thanks!

Actually, counting the files, and operating on them, should probably be performed separately -- it's simpler and easier to understand.

You can create a list of .html files and move through the list, performing any operations you want. Also, listing the files located in a particular directory, and listing the files located in an entire directory tree, are different operations.

Here's an example that lists the .html files in the current directory and operates on them:

Code:

ls -1 *.html | while read line
do
   echo "do something to \"$line\" here."
done

This method is written to allow spaces in the file names. Note: the "-1" argument to the "ls" call is the number one, not a lowercase "L".

Now here's an example that lists all the .html files in a directory tree, but is otherwise the same as the first example:

Code:

path="/netbackup/data/Network/arachnoid"

find $path | grep "\.html$" | while read line
do
   echo "do something to \"$line\" here."
done

The path in this example happens to be the location of my local Website archive, just change it to suit your own needs. Unlike the first example, this one doesn't need to be run in the target directory, in fact it can be run from any location.

One more thing. Because of how "find" operates, the generated list will not be in any predictable order. If you need the names to be in alphabetical order, add a "sort" command to the stream:

Code:

path="/netbackup/data/Network/arachnoid"

find $path | grep "\.html$" | sort | while read line
do
   echo "do something to \"$line\" here."
done

HTH

nathanpc · 09-10-2009, 08:30 PM

Huh, all this discussion only about a simple thing, ow!
But i know that this is a simple thing, but when you want to get better of a command it gets more and more difficult to use!

gqchynaboy · 09-11-2009, 12:17 PM

Quote:

Originally Posted by lutusp

Actually, counting the files, and operating on them, should probably be performed separately -- it's simpler and easier to understand.

You can create a list of .html files and move through the list, performing any operations you want. Also, listing the files located in a particular directory, and listing the files located in an entire directory tree, are different operations.

Here's an example that lists the .html files in the current directory and operates on them:

Code:

ls -1 *.html | while read line
do
   echo "do something to \"$line\" here."
done

This method is written to allow spaces in the file names. Note: the "-1" argument to the "ls" call is the number one, not a lowercase "L".

Now here's an example that lists all the .html files in a directory tree, but is otherwise the same as the first example:

Code:

path="/netbackup/data/Network/arachnoid"

find $path | grep "\.html$" | while read line
do
   echo "do something to \"$line\" here."
done

The path in this example happens to be the location of my local Website archive, just change it to suit your own needs. Unlike the first example, this one doesn't need to be run in the target directory, in fact it can be run from any location.

One more thing. Because of how "find" operates, the generated list will not be in any predictable order. If you need the names to be in alphabetical order, add a "sort" command to the stream:

Code:

path="/netbackup/data/Network/arachnoid"

find $path | grep "\.html$" | sort | while read line
do
   echo "do something to \"$line\" here."
done

HTH

The way you explained was so much easier then what I was trying to do, thank you very much!

Phil168 · 08-31-2011, 08:50 AM

I just wanted to point out that the solution (#2) does not quite do what the author says, and will sometimes miscount

Quote:

Use a regular expression with grep:
Code:

Code:

ls -1R | grep .*.mp3 | wc -l

list all recursively one per line, piped through grep which will output only: any amount of characters, then a dot character then 'mp3' then piped through wc -l to count it.

Namely the dot is a wildcard in grep regular expressions (and * is the repetition) so this will count anything containing "mp3", rather than ".mp3". Thus if you store your mp3 files in a subdirectory of your music folder labelled mp3 (like I do), then you will miscount. To list only those with ".mp3" you need to use the escape character before the dot. For example

Code:

ls -1Rl | grep '\.mp3'  | wc -l

will count the files with ".mp3" somewhere (the ".*" is unneeded as grep looks at substrings anyway). Even better we could look for ".mp3" at the end of the line (via $) and ignore case (so .MP3 is counted too)

Code:

ls -1Rl | grep -i '\.mp3$'  | wc -l