Hi,

I am using unzip utility to unzip .zip file.
If unzip finds the file while it is downloading, it gives an error something like

End-of-central-directory signature not found. Either this file is not
a zipfile, or it constitutes one disk of a multi-part archive. In the
latter case the central directory and zipfile comment will be found on
the last disk(s) of this archive.

Is there a way by which we can ensure that the .zip file has been properly downloaded before we give it to unzip utility.

Kind regards,

indiancosmonaut

What are you using to download the file?

Some downloaders will put a <filename>.zip.part file out there, along with just <filename>.zip. The .zip file will be zero bytes, while the .part file will contain the data. As soon as the download is complete, the .part file is moved to .zip, and removed, leaving you with one complete file.

Wget may act differently, as so some download utilities used through your web browser. You don't say what you're using, so it's hard to help.

If this is a program you're writing yourself, you should be able to easily monitor the download by looking for the thread in the process table, or checking the return status of the code....

Hi TBOne,

Thanks for replying,

I am using unzip utility to unzip my zip file.
I get the .zip file using secure ftp. And as soon as the file arrives, the utility tries to unzip it and fails giving the error mentioned below.

I am looking for a standard procedure (maybe) to ensure that the file is downloaded 100% and then go to the next line of code where the unzip utility is waiting to unzip it.

I tried,

cp /working_dir/file.zip to /tmp/file.zip
mv /tmp/file.zip to /working_dir/file.zip

but this didn't work.

Kind regards,

indiancosmonaut

Yes, I know...I had asked what you were using to DOWNLOAD the file. Please re-read my post.

What you're using to download the file, will have some sort of rules you can go by.

I got it.

Actually I run a script on my IBM AIX to unzip the file, which continuously checks for the zip file. Meanwhile at any point in time I ftp the file to the AIX box using secureshell (sftp). I am just transferring the file from my local machine to the UNIX box.

I hope I am able to explain

Kind regards,

indiancosmonaut

What you're describing can't be done. It's like asking "How high is up?"; the file(s) you're transferring show up at "any point in time", and can be any size. Since you don't know when they'll show up, or how big they can be, there's no way to know what's going on.

There are some klunky ways of doing things, though. You might be able to transmit a second "confirmation" file, such as a <filename>.txt file, containing the file size of the .zip file. Your script on the 'receiving' box could look for that file, and if it finds it, begins processing. Further error checking could be done by opening the confirmation file, and comparing the size in it, to the actual size of the file, to make sure there were no transmission errors.

Ok.

I think one more thing that could be done is that we check for the file size at regular intervals of time. After two consecutive checks, if the file size is unchanged then we can safely assume that the file is copied 100% and so we can go ahead with the unzipping.

Kind regards,

indiancosmonaut

fixed.

Maybe another way would be to use lsof to check whether the file is opened by any programs. If it's not then it must be done downloading.

Ok. The problem is that I do not have lsof installed. Thats the reason I was considering an alternative. But thanks anyway.

Kind regards,

indiancosmonaut

A simpler way may be to download it as another name, then rename it when it's complete (assuming you're using a script to download it here, otherwise you can do this by hand also. rename/mv takes (almost) no time.

how about every second checking for the existance of:when it dissappears from the process list it should be done.


edit:
of course 'grep sftp <filename>' might show up in the process list so some magic has to be done to combat that.