Unzip problem please help

[gussy82]

Hello,

I have created a directory and populated it with a zip file from a ftp server.
The issue im having is if i unzip the file the given name of the unzipped file is not the same a the zipped file.

for example (AllData_20110221_163857.zip) becomes (AllData_TMacID_sTime_eTime.csv)

How can i name to unzipped file the same as the zipped file, so that i would get AllData_20110221_163857.csv?

[Irithori]

The zip name is independent of the file and directorynames it contains.

I would look at the process that creates those zips -or probably even before that- and enforce a naming convention by code and/or policy.

[gussy82]

The zip files are created on a daily basis by an external server. my initial idea was to change the name of the file before it is zipped, but this cant be done so i need to find a workround.

Can the unzipped file be renamed with the zip file name and the original zip file deleted?

***I should say that this zip file is one of many that are generated on a daily basis and always contain only 1 csv file*** I can elaborate more if needed.

[Irithori]

Yes, e.g. with a wrapper script, that saves the zip filename in a variable and uses it for a mv.
But this is just a hack/workaround and also requires handling of various errorconditions.

What if the zip is named incorrectly,
if it contains multiple files,
if the zip is corrupt,
etc, etc

It might be possible to omit the zip by using a ftp server/client with compression.

Or dont use ftp in the first place in favour of e.g. rsync or scp.

[gussy82]

How would i phrase the wrapper script just for one unzip-rename-delete original zip file sequence?

I assume i can then create a repeat loop that can work through multple zipped files in the directory, unzipping, naming and deleting the original?

thanks

[Ignotum]

Hi.

Saw your post & rustled up the following:

#!/bin/sh

for FILE in *.zip;                      # Do the following for all zip files
do 
  mkdir .$FILE.tmp                      # Make a temporary hidden directory
  cd .$FILE.tmp                         # ...So that there are no mix-ups
  unzip ../$FILE                                # Unzip the file into it
  NEWFILE=`echo $FILE | sed s/.zip$/.csv/`      # Work out the new filename 
  mv *.csv ../$NEWFILE                          # Move & rename the file
  cd ..                                 # Leave the temporary directory 
  rmdir .$FILE.tmp                      # Remove the temporary directory
done

...But make sure you thoroughly check it works on something non-crucial first --- it's late & my brain's probably not firing on all cylinders...

[Ignotum]

...Any joy with that script, Gussy, old chap?

(Now I've had some sleep and looked over it again, I think it should indeed do what you want to do: just make sure you test it first in a temporary directory on some expendable copies of your zip files.)

[gussy82]

Hello Ignotum....

The script worked perfectly. thankyou for your help! Its much appreciated.

[Ignotum]

Jolly good: glad to have helped!

[gussy82]

Hello,

I might of found a bug in the script. There seems to be an issue with the unzip command. It seems to be partially unzipping some of the files + populating some with random data.
I have tested this againt my standard front end unzip package and traced the fault to the script. I should say that im doing this on a intel mac... :-/.. i hope that doesnt upset anyone, since a lot of the developer guys i work with think im a moron for using one.

These files unzip to 26 meg so there not massive, the script is pretty much as above.

Any ideas? could i try gunzip instead?

cd ./desktop/raptor
for FILE in *.zip; # Do the following for all zip files
do
mkdir .$FILE.tmp # Make a temporary hidden directory
cd .$FILE.tmp # ...So that there are no mix-ups
unzip ../$FILE # Unzip the file into it
NEWFILE=`echo $FILE | sed s/.zip$/.csv/` # Work out the new filename
mv *.csv ../$NEWFILE # Move & rename the file
cd .. # Leave the temporary directory
rmdir .$FILE.tmp # Remove the temporary directory
done
rm *.zip