Closed GoogleCodeExporter closed 8 years ago
The uploader.py script seems to only create new items, not check if the item
exists beforehand.
Nonetheless, its a problem on Archive.org itself regarding the change in the
last updated date. It is currently not possible to change the metadata of an
item (probably even with the destroy existing metadata header).
If this problem persists, can we decide on whether we want to drop that line of
metadata regarding the last updated date? IMO downloaders know that the later
the date is in the filename, the newer the dump is.
Original comment by ad...@alphacorp.tk
on 9 Nov 2012 at 3:35
The label is not for downloaders, it's for scripts and querying.
It's not true that items are recreated. Destroying and regenerating metadata is
probably the way to go.
Original comment by nemow...@gmail.com
on 9 Nov 2012 at 4:41
Original comment by nemow...@gmail.com
on 25 Oct 2013 at 12:44
Allow me to handle this (and fix the biggest major issue with uploader.py: the
identifiers).
Archive.org released a metadata API for use in updating the items and adding
new metadata (see http://blog.archive.org/2013/07/04/metadata-api/). I will try
to use that in the script when it finds that there already is an item sitting
in the desired identifier.
About the identifiers, the domain looks too weird to be pleasing the eye. I
will try to get the identifier fixed so that it can be compatible with the
format that was originally desired (and to make it usable for humans and
machines alike)
Original comment by ad...@alphacorp.tk
on 12 Nov 2013 at 6:28
Fixed in rr934
Original comment by nemow...@gmail.com
on 29 Jan 2014 at 4:16
Original issue reported on code.google.com by
nemow...@gmail.com
on 9 Nov 2012 at 12:10