Script to unzip and gzip files into directory,

eldaria · 03-15-2007, 06:34 PM

Hello all.
I have a whole bunch of zip files that I would like to convert into individual gzip files.

Basically I want to unzip the zip file into a directory named the same as the zip file, but without the .zip
Then I want to individually gzip each file inside.
I need to have it in a sub directory since files inside the zip file can be named the same as in other zip files.
It is a bit difficult to explain so I will try to illustrate

Documents(Eng).zip
Document1.rtf
Document2.rtf
Documents(Ger).zip
Document1.rtf
Document2.rtf
Documents(Fre).zip
Document1.rtf
Document2.rtf

The part I'm having trouble with is the first step of unzipping into a folder named the same as the zip file.
I painfully did one of my directories with Zip files, one by one.
and then to gzip the files was easy,

In folder with all subfolders.

Code:

find -iname '*.rtf' -execdir gzip -9 {} \;

However I have 34 Directories with each containing some 600 zip files.
so as you can guess it is a pain to do it manually. :-)

I tried to make my script grab the file name and then use sed to strip the .zip, but I can't even get that to work.

To prevent any actual change I used just an echo to generate the output, but it get's messed up.
It correctly removes the .zip, but it cuts the filename on each space and special characters.

Code:

#!/bin/bash

for file in $( find -iname '*.zip')
do
        output=`echo $file | sed 's/.zip//g'`
        echo $output
done

exit 0

Any ideas why or maybe a better way of doing it?

cfaj · 03-15-2007, 07:25 PM

Quote:

Originally Posted by eldaria

Hello all.
I have a whole bunch of zip files that I would like to convert into individual gzip files.

Basically I want to unzip the zip file into a directory named the same as the zip file, but without the .zip
Then I want to individually gzip each file inside.
I need to have it in a sub directory since files inside the zip file can be named the same as in other zip files.
It is a bit difficult to explain so I will try to illustrate

Documents(Eng).zip
Document1.rtf
Document2.rtf
Documents(Ger).zip
Document1.rtf
Document2.rtf
Documents(Fre).zip
Document1.rtf
Document2.rtf

The part I'm having trouble with is the first step of unzipping into a folder named the same as the zip file.
I painfully did one of my directories with Zip files, one by one.
and then to gzip the files was easy,

In folder with all subfolders.

Code:

find -iname '*.rtf' -execdir gzip -9 {} \;

However I have 34 Directories with each containing some 600 zip files.
so as you can guess it is a pain to do it manually. :-)

I tried to make my script grab the file name and then use sed to strip the .zip, but I can't even get that to work.

To prevent any actual change I used just an echo to generate the output, but it get's messed up.
It correctly removes the .zip, but it cuts the filename on each space and special characters.

Code:

#!/bin/bash

for file in $( find -iname '*.zip')
do
        output=`echo $file | sed 's/.zip//g'`
        echo $output
done

exit 0

Any ideas why or maybe a better way of doing it?

Untested:

Code:

find -type f -iname '*.zip' |
 while IFS= read -r file
 do
   dir=${file%.zip}
   if mkdir "$dir"
   then
     cd "$dir" || { echo "Could not cd to $dir"; continue; }
     unzip "$file"
     gzip * ## If there are subdirs: find . -type f -exec gzip * \;
   fi
 done

eldaria · 03-16-2007, 04:56 AM

Thank you very much with some modifications it worked,

It failed on 2 points.
It creates all the subfolders within eachother.

Code:

./Documents(Eng)/Documents(Ger)/Documents(Fre)

I solved that by adding a cd .. just after the line of gzip

Unzip also failed with error that it can't find the file.
I first thought it was related to spaces in the name, but later found that it looks in current directory for the file.
So added the full path to the file, and now it works.

Code:

#!/bin/bash

find -type f -iname '*.zip' |
 while IFS= read -r file
 do
   dir=${file%.zip}
   if mkdir "$dir"
   then
     curdir=`pwd`
     cd "$dir" || { echo "Could not cd to $dir"; continue; }
     unzip "$curdir/$file"
     gzip -v9 *
     cd ..
   fi
 done
exit 0

For the learning experience could you explain a bit?

I don't understand what happens with these bits:

Code:

while IFS= read -r file

I know the 'while' command from Visual basic, but not the 'IFS' and I have no man page for the 'read' or the 'while' command, so can't see the syntax.

Code:

dir=${file%.zip}

I understand it strips of the '.zip' but how, I was fighting with sed, but this seems easier.

Code:

|| { echo "Could not cd to $dir"; continue; }

Not sure how this part works, is it if cd fails, it will break the loop?

Hope you have patience to explain. ;-)

cfaj · 03-16-2007, 10:59 AM

Quote:

Originally Posted by eldaria

Thank you very much with some modifications it worked,

It failed on 2 points.
It creates all the subfolders within eachother.

I should have executed cd in a subshell:

Code:

    (
     cd "$dir" || { echo "Could not cd to $dir"; continue; }
     unzip "$file"
     gzip * ## If there are subdirs: find . -type f -exec gzip * \;
    )

Quote:

Code:

./Documents(Eng)/Documents(Ger)/Documents(Fre)

I solved that by adding a cd .. just after the line of gzip

Unzip also failed with error that it can't find the file.
I first thought it was related to spaces in the name, but later found that it looks in current directory for the file.
So added the full path to the file, and now it works.

Code:

#!/bin/bash

find -type f -iname '*.zip' |
 while IFS= read -r file
 do
   dir=${file%.zip}
   if mkdir "$dir"
   then
     curdir=`pwd`

You don't need pwd; the current directory is stored in $PWD.

Quote:

Code:

     cd "$dir" || { echo "Could not cd to $dir"; continue; }
     unzip "$curdir/$file"
     gzip -v9 *
     cd ..
   fi
 done
exit 0

For the learning experience could you explain a bit?

I don't understand what happens with these bits:

Code:

while IFS= read -r file

I know the 'while' command from Visual basic, but not the 'IFS' and I have no man page for the 'read' or the 'while' command, so can't see the syntax.

Read and while are built into the shell; read the shell's man page.

Quote:

Code:

dir=${file%.zip}

I understand it strips of the '.zip' but how,

Read the "Parameter Expansion" section of the shell's man page.

Quote:

I was fighting with sed, but this seems easier.

Code:

|| { echo "Could not cd to $dir"; continue; }

Not sure how this part works, is it if cd fails, it will break the loop?

Hope you have patience to explain. ;-)