From: Fuad Saud <fuadksd@...>
Date: 2013-11-04T18:26:31-02:00
Subject: [ruby-core:58154] Re: [ruby-trunk - Feature #9076] New one-argument block syntax: &.

--527802f7_2ae8944a_7322
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Introducing both syntaxes indeed is a bit confusing. I don=E2=80=99t thin=
k

=5B1, 10, 100=5D.map &.to=5Fs.length

is much better than

=5B1, 10, 100=5D.map =7B =7Ci=7C i.to=5Fs.length =7D

I think the latter is clearer. I'm fond of having some sugar for position=
al parameter access (like % in clojure or & in elixir), though. =20

-- =20
=46uad Saud
Sent with Sparrow (http://www.sparrowmailapp.com/=3Fsig)


On Monday, November 4, 2013 at 1:57 PM, shevegen (markus heiler) wrote:

> =20
> Issue =239076 has been updated by shevegen (markus heiler).
> =20
> =20
> Note that if &: would be removed at the same time and &. added I would =
not mind it that much. You could use & as a reference to a global object,=
 just similar to how =241 or =242 is automagically set for regexes. But t=
o keep two different syntaxes for semi-similar issues is very bad.
> ----------------------------------------
> =46eature =239076: New one-argument block syntax: &.
> https://bugs.ruby-lang.org/issues/9076=23change-42742
> =20
> Author: asterite (Ary Borenszweig)
> Status: Open
> Priority: Normal
> Assignee: =20
> Category: core
> Target version: current: 2.1.0
> =20
> =20
> Hello,
> =20
> I'd like to introduce a new syntax for blocks that have one argument.
> =20
> Currently you can do this:
> =20
> =5B1, 2, 3=5D.map &:to=5Fs
> =20
> With the proposed syntax this will be written as:
> =20
> =5B1, 2, 3=5D.map &.to=5Fs
> =20
> Instead of =22:=22 we use a =22.=22.
> =20
> The idea is that this new syntax is just syntax sugar that is expanded =
by the parser to this:
> =20
> =5B1, 2, 3=5D.map =7B =7Carg=7C arg.to=5Fs =7D
> =20
> This new syntax allows passing arguments:
> =20
> =5B1, 2, 3, 4=5D.map &.to=5Fs(2) =23=3D> =5B=221=22, =2210=22, =2211=22=
, =22100=22=5D
> =20
> It also allows chaining calls:
> =20
> =5B1, 10, 100=5D.map &.to=5Fs.length =23=3D> =5B1, 2, 3=5D
> =20
> You can also use another block:
> =20
> =5B=5B1, -2=5D, =5B-3, -4=5D=5D.map &.map &.abs =23=3D> =5B=5B1, 2=5D, =
=5B3, 4=5D=5D
> =20
> Pros:
> - Doesn't conflict with any existing syntax, because that now gives a s=
yntax error, so it is available.
> - Allows passing arguments and chaining calls
> - It's *fast*: it's just syntax sugar. The =22&:to=5Fs=22 is slower bec=
ause the to=5Fproc method is invoked, you have a cache of procs, etc.
> - It looks ok (in my opinion) and allows very nice functional code (lik=
e the last example).
> =20
> Cons:
> - Only supports one (implicit) argument. But this is the same limitatio=
n of =22&:to=5Fs=22. If you want more than one argument, use the traditio=
nal block syntax.
> - It's a new syntax, so users need to learn it. But to defend this poin=
t, users right now need to understand the &:to=5Fs syntax, which is hard =
to explain (this calls the =22to=5Fproc=22 method of Symbol, which create=
s a block... vs. =22it's just syntax sugar for=22)
> =20
> What do you think=3F
> =20
> We are using this syntax in a new language we are doing, Crystal, which=
 has a syntax very similar to Ruby, and so far we think it's nice, simple=
 and powerful. You can read more about it here: http://crystal-lang.org/2=
013/09/15/to-proc.html
> =20
> =20
> -- =20
> http://bugs.ruby-lang.org/
> =20
> =20



--527802f7_2ae8944a_7322
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline


                <div style=3D=22font-family: Helvetica; font-size: 13px;=22=
>Introducing both syntaxes indeed is a bit confusing. I don=E2=80=99t thi=
nk<div><br></div><div>=5B1, 10, 100=5D.map &amp;.to=5Fs.length</div><div>=
<br></div><div>is much better than</div><div><br></div><div>=5B1, 10, 100=
=5D.map =7B =7Ci=7C i.to=5Fs.length =7D</div><div><br></div><div>I think =
the latter is clearer. I'm fond of having some sugar for positional param=
eter access (like % in clojure or &amp; in elixir), though.</div></div>
                <div><div><br></div><div>--&nbsp;</div><div>=46uad Saud</=
div><div>Sent with <a href=3D=22http://www.sparrowmailapp.com/=3Fsig=22>S=
parrow</a></div><div><br></div></div>
                =20
                <p style=3D=22color: =23A0A0A8;=22>On Monday, November 4,=
 2013 at 1:57 PM, shevegen (markus heiler) wrote:</p>
                <blockquote type=3D=22cite=22 style=3D=22border-left-styl=
e:solid;border-width:1px;margin-left:0px;padding-left:10px;=22>
                    <span><div><div><div><br></div><div>Issue =239076 has=
 been updated by shevegen (markus heiler).</div><div><br></div><div><br><=
/div><div>Note that if &amp;: would be removed at the same time and &amp;=
. added I would not mind it that much. You could use &amp; as a reference=
 to a global object, just similar to how =241 or =242 is automagically se=
t for regexes. But to keep two different syntaxes for semi-similar issues=
 is very bad.</div><div>----------------------------------------</div><di=
v>=46eature =239076: New one-argument block syntax: &amp;.</div><div><a h=
ref=3D=22https://bugs.ruby-lang.org/issues/9076=23change-42742=22>https:/=
/bugs.ruby-lang.org/issues/9076=23change-42742</a></div><div><br></div><d=
iv>Author: asterite (Ary Borenszweig)</div><div>Status: Open</div><div>Pr=
iority: Normal</div><div>Assignee: </div><div>Category: core</div><div>Ta=
rget version: current: 2.1.0</div><div><br></div><div><br></div><div>Hell=
o,</div><div><br></div><div>I'd like to introduce a new syntax for blocks=
 that have one argument.</div><div><br></div><div>Currently you can do th=
is:</div><div><br></div><div>=5B1, 2, 3=5D.map &amp;:to=5Fs</div><div><br=
></div><div>With the proposed syntax this will be written as:</div><div><=
br></div><div>=5B1, 2, 3=5D.map &amp;.to=5Fs</div><div><br></div><div>Ins=
tead of =22:=22 we use a =22.=22.</div><div><br></div><div>The idea is th=
at this new syntax is just syntax sugar that is expanded by the parser to=
 this:</div><div><br></div><div>=5B1, 2, 3=5D.map =7B =7Carg=7C arg.to=5F=
s =7D</div><div><br></div><div>This new syntax allows passing arguments:<=
/div><div><br></div><div>=5B1, 2, 3, 4=5D.map &amp;.to=5Fs(2) =23=3D&gt; =
=5B=221=22, =2210=22, =2211=22, =22100=22=5D</div><div><br></div><div>It =
also allows chaining calls:</div><div><br></div><div>=5B1, 10, 100=5D.map=
 &amp;.to=5Fs.length =23=3D&gt; =5B1, 2, 3=5D</div><div><br></div><div>Yo=
u can also use another block:</div><div><br></div><div>=5B=5B1, -2=5D, =5B=
-3, -4=5D=5D.map &amp;.map &amp;.abs =23=3D&gt; =5B=5B1, 2=5D, =5B3, 4=5D=
=5D</div><div><br></div><div>Pros:</div><div>- Doesn't conflict with any =
existing syntax, because that now gives a syntax error, so it is availabl=
e.</div><div>- Allows passing arguments and chaining calls</div><div>- It=
's *fast*: it's just syntax sugar. The =22&amp;:to=5Fs=22 is slower becau=
se the to=5Fproc method is invoked, you have a cache of procs, etc.</div>=
<div>- It looks ok (in my opinion) and allows very nice functional code (=
like the last example).</div><div><br></div><div>Cons:</div><div>- Only s=
upports one (implicit) argument. But this is the same limitation of =22&a=
mp;:to=5Fs=22. If you want more than one argument, use the traditional bl=
ock syntax.</div><div>- It's a new syntax, so users need to learn it. But=
 to defend this point, users right now need to understand the &amp;:to=5F=
s syntax, which is hard to explain (this calls the =22to=5Fproc=22 method=
 of Symbol, which creates a block... vs. =22it's just syntax sugar for=22=
)</div><div><br></div><div>What do you think=3F</div><div><br></div><div>=
We are using this syntax in a new language we are doing, Crystal, which h=
as a syntax very similar to Ruby, and so far we think it's nice, simple a=
nd powerful. You can read more about it here: <a href=3D=22http://crystal=
-lang.org/2013/09/15/to-proc.html=22>http://crystal-lang.org/2013/09/15/t=
o-proc.html</a></div><div><br></div><div><br></div><div>-- </div><div><a =
href=3D=22http://bugs.ruby-lang.org=22>http://bugs.ruby-lang.org</a>/</di=
v></div></div></span>
                =20
                =20
                =20
                =20
                </blockquote>
                =20
                <div>
                    <br>
                </div>
            
--527802f7_2ae8944a_7322--