Agreed. It seems like feeds are still just not in share-able shape, and
we should probably cut them from this release - and save them for some
future point when they are. UDFs need to work, 2nd-ary feeds need to
work, the doc needs to be there and follow-able - we just aren't there -
sigh. Let's set them aside for now and ship the release with the rest
of our new goodies - e.g., related to the many bug fixes (esp. for open
data), indexing of nested/open data, and indexing of HDFS external data.
On 8/28/15 7:32 PM, Chen Li wrote:
Heri was trying to make this part work. Any update from him?
In general, if we cannot make secondary feeds work and documented quickly,
I don't think it's a good idea to include it in this release.
Chen
On Fri, Aug 28, 2015 at 8:03 PM, Ian Maxon <[email protected]> wrote:
What's in master (in the docs) seems to work OK, but that's just
connecting a source to a dataset directly. There's some issue with
printing some tweets (e.g. for $x in dataset Tweets return $x) that
contain special characters, but I'm not sure where that comes from.
However I haven't had much luck getting the secondary feeds/functions
to work. The original documentation seemed to refer to a function that
was renamed/moved, but the renamed function (addHashTags) didn't work
right for me either ( in https://asterix-gerrit.ics.uci.edu/#/c/359/).
If there was a function (that will work reliably) we could use to
create a secondary feed that's already in the code, I think that would
let us include that feature in the docs. Without that though, I don't
know what else can be done besides leaving it hidden for the immediate
future.
- Ian
On Fri, Aug 28, 2015 at 4:56 PM, Michael Carey <[email protected]>
wrote:
Q: So where are things on the feeds front - in terms of what's documented
and what works as advertised (and documented)?
