Hi all,

I have a client/server app I'm building. To avoid bundling the AWS secret keys with my little client app, I'm using S3's signed URL feature.

My server (running python) is successfully generating signed URL's (signed for 'PUT', not 'GET'... I'm uploading) that look like this:

https://myBucket.s3.amazonaws.com:443/somename%2Ftesting.jpg?Signature=YuyOUOuho87858gjguy65865kgh%3D&Expires=1216301872&AWSAccessKeyId=U878JAN7HUYFOU76HGBKMHF

Can S3cmd use that signed URL to upload to S3?

hmmm ..I just saw  "# get_torrent - added by dominic@gmail.com"  in the code ... but how do I get at it from the command line?

S3 doest support resume upload does it?

bummer that. ... I bet lots of people want this. I guess I should put this on the S3 forums.

Getting number of downloads is pretty important for me. I'm assuming this is not a trivial task ... but some kind of call to do a date range based number of downloads would be excellent. Something like:

s3cmd.rb  countdownloads  <bucket>:key [startDate] [endDate]

where starDate and endDate are optional ... specifying neither means retrieve all downloads for all time. Specifying only the startDate woud mean from startDate to now. Specifying the end date would mean from all-time start to that date.

its really easy to get downloads from a torrent tracker (garanteed to count only full-downloads)... but I'm not sure how the logging on Amazon S3 works ... just giving you the http requests is not necessaily an accurate refelction of full-downloads (especially for files 200Meg and up).... lots of people would start the download and then quit it when they lose patience ...

Etienne

A cmd to get the torrent would be nice ... I know that all you need to do is add '.torrent' to the URL and you get the torrent, but I have a use case where I dont want the end user to go fetch the torrent themselves. I want to disable the torrent (publically), but get it privately and offer the torrent privately. This is to minimise my bandwidth useage, and essentially just use Amason S3 as a seeder and a tracker.

--progress option is very nice. This kind of thing is missing in a lot of the other S3 libs ... good one

I was saying this on the other thread ... but I'll repeat it here just in case no one reads teh other thread.

been playing around with aws-s3 and I'm really disapointed with it, it has major problems on windows ... s3sync seems to be more mature .... would _love_ an API to use from within RUBY apps.

been trying to use aws-s3 gem ... but it has failings under windows (and the mail list is kind of slow and quiet)

s3sync seems more mature ....... an API to access from with Ruby would be much appreciated