Hi, I'm using info-zip to create zip archives. When i add files to an existing archive, the entire archive is re-written to a new temporary file and later is renamed. This means that if i have a big file and i want to add a small file to it, it will take a long time since the entire big file will be re-written. Is there any way to avoid this? like an option to append the new files to the existing archive? I've read a little about the zip archive format and it seems possible, since the zip central directory is stored at the end of the file. So theoretically you can append the last file and re-write only the zip central directory. I didn't find this option in info-zip, maybe i missed it? or maybe anyone knows another utility that can do this?

Thanks.

The zip utility, which should be in the repository of almost any distribution can do that with its add-option.

The "zip" and "unzip" utilities that come with Linux distributions are info-zip, and when you add files to an existing file, it re-writes the entire file to a temporary file, and finally renames it.

Indeed info-zip is the owner of the copyright of the zip utility. You can see it by:
To add a file to an existing archive no special option is needed. Here is an example on my system zipping a directory of about 2 Gb:
As you can see from the time required to create and then to add a single file, it could have not been recreated.

I think it has to create a new file and just can't add to the old one. Where it is doing it would be my guess memory unless you don't have enough. It isn't exactly like a tar process. A zip will recompute checksum. Pretty sure pk-zip did it that way a long time ago.

From man.

"When changing an existing zip archive, zip will write a temporary file with the new contents, and only replace the old one when the process of creating the new version has been completed without error."

So it's not possible with info-zip. Thanks.

Keep in mind that when recreating the file, zip does not need to re-compress the original files, just copy the already-compressed data. That is likely to take far less time than it took to create the original archive. Unlike gzip, which compresses the entire content as a single unit, zip independently compresses each file in the archive, and indeed the files in a single archive could use different compression methods.

1 members found this post helpful.