Using the absence/presence of a space to influence parsing bugs me a lot...
No idea about using an explicit delimiter ?
My main concern is how can the reader report a decent error message
if there's an input mistake in this context ? I feel that a mistake
can have far reaching side effects because it will change evaluation
of the forms and some odd behaviors could result from such a mistake....
Luc
> Yes, to be exactly, the simple change in the reader is after parsing a
> symbol, if the following char separating the symbol is '(', the symbol
> become the leading item of the following list, otherwise it is a
> regular item in the current list context.
>
> Louis
>
> On Dec 27, 4:16 pm, Softaddicts <lprefonta...@softaddicts.ca> wrote:
> > So if I understand correctly, your proposal makes the absence/presence of
> > space pivotal
> > in making the decision about parsing infix/postfix notation ?
> >
> > Luc
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > > Luc,
> > > I see you and other people put great points on this subject, that is
> > > exactly what I wanted to know about the 'dark corner' in my initial
> > > post.
> > > Just for your info, I have tried on the modified reader, the following
> > > alternatives all works:
> >
> > > Clojure 1.4.0-master-SNAPSHOT
> > > user=> (map first [[1] [2]])
> > > (1 2)
> > > user=> map(first [[1] [2]])
> > > (1 2)
> > > user=> (defmacro mymac [func]
> > > (let [args (rest func)]
> > > `(- ~@args)))
> > > #'user/mymac
> > > user=> (mymac (+ 1 2 3))
> > > -4
> > > user=> mymac((+ 1 2 3))
> > > -4
> > > user=> mymac(+(1 2 3))
> > > -4
> >
> > > Thanks,
> > > Louis
> >
> > > On Dec 27, 1:40 pm, Luc Prefontaine <lprefonta...@softaddicts.ca>
> > > wrote:
> > > > Louis, obviously there's a problem here, the REPL should return
> >
> > > > user=> (map first [[1] [2]])
> > > > (1 2)
> >
> > > > As for my point about macros, it's not about the calls, it's about
> > > > macro processing before
> > > > spitting out the code that will be compiled, what would this return ?
> >
> > > > user=> (defmacro mymac [func]
> > > > (let [args (rest func)]
> > > > `(- ~@args)))
> > > > #'user/mymac
> >
> > > > Presently:
> >
> > > > user=> (mymac (+ 1 2 3))
> > > > -4
> >
> > > > What about (mymac + (1 2 3)) ? Would + (1 2 3) be considered as one
> > > > argument ? Why ? How the compiler achieve this grouping ?
> > > > If the macro argument involves a list ([&body] being a common example)
> > > > how do you decide what is a real
> > > > list versus a call using your new syntax in the variable list of
> > > > arguments ?
> >
> > > > If you cannot deal with this and have to force (mymac (+ 1 2 3)) when a
> > > > macro is involved then what's the use
> > > > of this new syntax ?
> >
> > > > What about embedded calls involving a mix of fns and macros ?
> > > > I cannot see people having to dig to find out if some of these calls
> > > > are macros versus fns and change the shape according to
> > > > what is being called.
> >
> > > > You missed the point that I made about the fact that code is data and
> > > > can be modified accordingly either through macros
> > > > or runtime evaluations.
> >
> > > > Luc
> >
> > > > On Wed, 28 Dec 2011 01:59:12 +0800
> >
> > > > Ambrose Bonnaire-Sergeant <abonnaireserge...@gmail.com> wrote:
> > > > > On Wed, Dec 28, 2011 at 1:46 AM, Louis Yu Lu <louisy...@gmail.com>
> > > > > wrote:
> >
> > > > > > The proposed syntax sugar does not break the existing code
> >
> > > > > clojure.core=> map(first [[1] [2]])
> > > > > #<core$map clojure.core$map@5ec3d2>
> > > > > [1]
> >
> > > > > Ambrose
> >
> > > > --
> > > > Luc P.
> >
> > > > ================
> > > > The rabid Muppet
> >
> > > --
> > > You received this message because you are subscribed to the Google
> > > Groups "Clojure" group.
> > > To post to this group, send email to clojure@googlegroups.com
> > > Note that posts from new members are moderated - please be patient with
> > > your first post.
> > > To unsubscribe from this group, send email to
> > > clojure+unsubscr...@googlegroups.com
> > > For more options, visit this group at
> > >http://groups.google.com/group/clojure?hl=en
> >
> > --
> > Softaddicts<lprefonta...@softaddicts.ca> sent by ibisMail!
>
> --
> You received this message because you are subscribed to the Google
> Groups "Clojure" group.
> To post to this group, send email to clojure@googlegroups.com
> Note that posts from new members are moderated - please be patient with your
> first post.
> To unsubscribe from this group, send email to
> clojure+unsubscr...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/clojure?hl=en
>
--
Softaddicts<lprefonta...@softaddicts.ca> sent by ibisMail!
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
