I’ve been playing with the Ruby lib aws-s3… but turns out the library is really not very mature. I have not managed to store _any_ files onto Amazon S3. I spent a couple of days batling with this, thinking I was using the lib wrongly.
But it turned out those days were wasted. My problems are a known bug.
Silence on the aws-s3 list is only broken by an other developper’s frustration about the same bug. Well, I guess this is what open-source software is all about. If there’s a bug, its nto garanteed to get fixed any time soon …and if you really need it fixed then fix it yourself and contribute to the cause.
I spent a day digging around the aws-s3 code to try and find the bug and squish it. Alas, my ruby newbie status failed me, and I just ended up more frsutrated. Its too early for me to contribute to Ruby development. Im still trying to surgically extract Java from my brain.
So I’ve moved on from aws-s3 to s3sync.rb… which seems much more mature. It should be said that the aws-s3 API _is_ well designed… it reflects very well the simplicity and power of Amazon S3.
But s3sync has one really nice feature, which is it can give the progress status of an upload… this is especially usefull for large files (like I have on ambisonicbootlegs)… because when someone uploads, they will need to know whether there is 10 minutes to go or 40 minutes to go.
If only the s3sync people could get together with he aws-s3 people… and have a love in… everyone would win!…. well, perhaps except for the pride of the founders of both those projects. Everyone has to have their fair share of that old open-source gold … “respect-currency“!