apex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Priyanka Gugale <priya...@datatorrent.com>
Subject Re: Cassandra query start value
Date Thu, 28 Jan 2016 06:13:01 GMT
The startRow purpose is as you explained, but the value assigned to it is
not serving the right purpose. As per my understanding the value should be
value of primary key of start row. I had to read through code to understand
that. A meaningful name and right javadoc will make it easy to understand.

-Priyanka

On Wed, Jan 27, 2016 at 9:40 PM, Pramod Immaneni <pramod@datatorrent.com>
wrote:

> Priyanka the start in the query should act like a cursor and keep
> getting updated when new data is pulled in and that's possible the
> reason for that name. Is that not happening?
>
> > On Jan 27, 2016, at 3:53 AM, Priyanka Gugale <priyag@apache.org> wrote:
> >
> > Hi,
> >
> > CassandraPOJOInputOpeartor in Malhar accepts query in the form:
> > "select * from %t where token(%p) > %s limit %l"
> > Where %s is start value.
> >
> > *What is "start value" here?*
> > From javadoc in code it looks like %s is the row number in table from
> where
> > you want to start scan. But looking at code it should be
> "primaryKeyValue"
> > of row from where you want to start scan. (I didn't find any way in cql
> to
> > scan starting from some row number, unless you have rowNumber column in
> > your table.)
> >
> > If this understanding is correct we should rename this to something like
> > startRowPrimaryKey and update javadoc.
> >
> > -Priyanka
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message