Mailing-List: contact dev-help@flink.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@flink.apache.org
MIME-Version: 1.0
In-Reply-To: 
 <CANMXwW33mjf4XWb5dqrbR_Ycap-=jRn5kwNt5bBMYfqwtqQwdg@mail.gmail.com>
References: 
 <CAKiyyaGRn4Ggyv2W8cz8PyVVEiv=Lgj4d-u_c-ft0PP=TcqiOg@mail.gmail.com>
 <CANC1h_t=RB+-ZBbOyQV0yScioW8YiCo320jOJzA=7Kt_YpAA9g@mail.gmail.com>
 <CAGco--bJqc6HzbbncACZyhBPJW8PzTrN9uSUioa7RRMvL1=8Ww@mail.gmail.com>
 <CAC27z=PKQv54Y_jfve6F3HiYnK=cfLpB_nR+mYE3UXwYrAUO1g@mail.gmail.com>
 <CANMXwW33mjf4XWb5dqrbR_Ycap-=jRn5kwNt5bBMYfqwtqQwdg@mail.gmail.com>
From: Maximilian Michels <mxm@apache.org>
Date: Mon, 13 Apr 2015 12:17:54 +0200
Message-ID: 
 <CAGco--ZqZ-UqfsAthdz1TsbZ7MUNvjMTwXurrev8TpVpfwfqYg@mail.gmail.com>
Subject: Re: [QUESTION] Scala Table API syntax
To: dev@flink.apache.org
Content-Type: multipart/alternative; boundary=089e0160bc449e94ac0513986e3a

--089e0160bc449e94ac0513986e3a
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hadn't thought about that but it makes perfect sense. Very handy.

On Mon, Apr 13, 2015 at 12:02 PM, Aljoscha Krettek <aljoscha@apache.org>
wrote:

> @Till: Correct
>
> On Mon, Apr 13, 2015 at 10:52 AM, Till Rohrmann <trohrmann@apache.org>
> wrote:
> > IMO the symbol notation makes it slightly easier to write Table API
> > expressions, because the IDE can assist you in what operations are
> > supported by the expression DSL whereas the string notation will only
> give
> > you a syntax error upon compilation, right?
> >
> > On Mon, Apr 13, 2015 at 10:42 AM, Maximilian Michels <mxm@apache.org>
> wrote:
> >
> >> 'Symbol is actually syntactic sugar for Symbol("Symbol") which is part
> of
> >> the Scala standard library. It is a core feature of Scala which IMO is
> >> perfectly fine to use. I'm not sure whether it makes the expression mu=
ch
> >> easier to read but it is a neat feature already in place.
> >>
> >> @Stephan: As far as I know, the String-based parsing is already
> possible in
> >> the Scala Table API.
> >>
> >> On Sun, Apr 12, 2015 at 2:51 PM, Stephan Ewen <sewen@apache.org> wrote=
:
> >>
> >> > I am not a deep Scala progammer either, but I think the symbols are =
a
> >> > pretty wide-spread concept.
> >> > For example, the Scalding tuple API makes heavy use of them as well.
> >> >
> >> > That said, I do like the idea that the Scala Table API supports the
> >> string
> >> > variant as well, for homogeneity.
> >> >
> >> > On Sun, Apr 12, 2015 at 1:12 PM, Ufuk Celebi <uce@apache.org> wrote:
> >> >
> >> > > Hey all,
> >> > >
> >> > > I am not very proficient with Scala and have some questions
> regarding
> >> the
> >> > > Scala Table API:
> >> > >
> >> > > The logical queries in the Java API are all String-based, e.g.
> >> > >
> >> > > table.groupBy("word")
> >> > >
> >> > > In the Scala API, this works as well, but what's further possible =
is
> >> > this:
> >> > >
> >> > > expr.groupBy('word)
> >> > >
> >> > > For comparisions you use something like `a =3D=3D=3D `b. Note that=
 the '
> is a
> >> > > Scala symbol.
> >> > >
> >> > > - How common is this kind of notation for Scala users?
> >> > > - Are both types of expressions equivalent or can you do more with
> the
> >> > > special Scala syntax Table API?
> >> > >
> >> > > I am asking, because I was wondering whether we should stick to th=
e
> >> > > String-based notation in the docs and have the special syntax as a=
n
> >> > > optional thing. There is no reason for this, if this is common in
> the
> >> > Scala
> >> > > world though. :-)
> >> > >
> >> > > =E2=80=93 Ufuk
> >> > >
> >> >
> >>
>

--089e0160bc449e94ac0513986e3a--