Option to completely skip items with already uploaded files.

[CrimsonGlory]

Currently, default options update the metadata. If you do -keep-metadata, it still make a request to archive.org (i don't know why).

I'm uploading lots of files. 1 item per file. If I cancel the script ^C and then start it, I got for every item already uploaded

409 Conflict

BucketAlreadyExistsThe requested bucket name is not available. The bucket namespace is shared by all users of the system. Please select a different name and try again.d1099353-dbd1-4f55-b7cd-d325aef280bf
0 file(s) to upload in...

It skip them quickly (about 1 per second), but when you have lots of items to upload, 1 wasted second per item can be lot of hours.

There is not an option to just skip those items, without quering archive.org or removing them manually from the csv file.

[kngenie]

(uh, sorry to leave this without a response so long. I haven't paying enough attention to the issues section.)

ias3upload.pl makes item-creation request for updating its metadata even if there's no files to upload, unless --ignore-nofile is specified. but yeah, I think ias3upload should make no request at all for an item if (1) --keep-metadata is specified, (2) it knows item exists, and (3) there's no files to upload. It's an oversight.
Unless I'm missing any use case conflicting with this behavior, we can solve the problem without adding new option, I suppose.