On Tue, Jan 12, 2016 at 7:20 PM, Adam Lamar <adamond...@gmail.com> wrote: > Russell, > > Sorry - I meant that I use the FetchS3Object processor, not ListS3.
Ah, that makes more sense, thanks. > You can > set the Object Key property with the name of the key you want to download. > This property supports the expression language, so the object key can be > sourced from each flowfile. > > I'm unsure if there is a good template around, but I'd be willing to share > my flow if that would be helpful. > That would be enormously useful, yes, thanks, please. Russell > Cheers, > Adam > > > On 1/12/16 8:14 PM, Russell Whitaker wrote: >> >> On Tue, Jan 12, 2016 at 7:12 PM, Adam Lamar <adamond...@gmail.com> wrote: >>> >>> On 1/12/16 8:04 PM, Bryan Bende wrote: >>>> >>>> In the case of FetchS3, there is definitely intent to have a ListS3, but >>>> there are still ways to use it with out that... >>> >>> >>> To add to Bryan's comment, I use ListS3 in combination with GetSQS to >>> fetch >>> objects as they are newly placed into an s3 bucket. For my use case, it >>> works great. >>> >> What "ListS3"? There's no ListS3* processor in the v0.4.1 distribution I >> have. >> >> R >> >>> There is also more discussion here: >>> https://issues.apache.org/jira/browse/NIFI-25 . In particular, see Mark >>> Payne's first comment. It sounds like the problem isn't so much >>> implementing >>> the functionality, but some of the semantics around exactly what a ListS3 >>> processor should do. >>> >>> Cheers, >>> Adam > > -- Russell Whitaker http://twitter.com/OrthoNormalRuss http://www.linkedin.com/pub/russell-whitaker/0/b86/329
