As Dan mentioned, the read path hasn't been implemented yet. We are in the process of building out the read path and will open a PR as soon as it's ready. We primarily use parquet and that will definitely be included.
For now, the library's main functionality is reading the metadata. I'll put together an example script on how metadata reads work in the current implementation. -Ted On Fri, Mar 29, 2019, 3:56 PM Anuj Shah <[email protected]> wrote: > Hi Daniel, > > > > I am trying to read parquet files. > > > > Thanks, > > Anuj > > > > *From: *Daniel Weeks <[email protected]> > *Reply-To: *"[email protected]" <[email protected]> > *Date: *Friday, March 29, 2019 at 3:38 PM > *To: *"[email protected]" <[email protected]> > *Subject: *Re: Examples on the Python support? > > > > Anuj, > > > > The current implementation of the python library supports accessing > iceberg metadata (schema, etc) and scanning manifests, but does not include > a read path. > > > > What types of files are you looking to read (parquet, avro, etc.)? > > > > -Dan > > > > On Fri, Mar 29, 2019, 10:38 AM Anuj Shah <[email protected]> > wrote: > > Hi All, > > > > Any thoughts on the query below? > 1. Would I be able to read the iceberg dataset? If not, what contributions > need to be done? > > > > I am happy to contribute. > > > > Thanks, > > Anuj > > > > *From: *Anuj Shah <[email protected]> > *Reply-To: *"[email protected]" <[email protected]> > *Date: *Monday, March 25, 2019 at 4:39 PM > *To: *"[email protected]" <[email protected]> > *Subject: *Examples on the Python support? > > > > Hi All, > > > > I am interested in Ted Gooch (TGooch44) contributed python support for > iceberg. I was looking at the code and wanted to try it out myself, for > starter do you have any examples for it? Like reading and writing iceberg > dataset? > > > > Thanks, > > Anuj > >
