"Wes" <ffhighwind hotmail.com> writes:

Can you do this in D? If not then what other language?

1. Can you get a list of all classes and variables in the current
scope e.g. __traits()?

2. Can you statically iterate over every field of a class?
It seems like the only returns are string lists from __traits().
I basically am interested in being able to generate a
PrettyPrint/Serializable/Hash template without passing in every
single field.

3. Is there any way of executing code or programs during compile
time?
I've seen an example of CTFE (Compile Time Function Evaluation),
although I'm unsure if this works for stuff like classes.
However, I am considering more advanced execution (not constants)
such as printing to a file during compiling for stuff like how
long compiling a certain function/template takes.

4. Is there anything like Go/Erlang channels? (synchronous queues)
I assume that synchronized blocks with a shared global dynamic
array would be the solution without the special syntax sugar?


This is more of a rhetorical... as I know the language doesn't
need them, nor would I know if they would be binary/unary
prefix/etc or the precedence... although they would be nice to
have. Specifically I'd like $prefix to be stringification.

"bearophile" <bearophileHUGS lycos.com> writes:

Wes:

 Can you do this in D? If not then what other language?

Maybe CommonLisp and Scheme (and maybe Clojure) are able to do 
what you ask for (but normally they don't have the 
stringification with $).


 1. Can you get a list of all classes and variables in the 
 current scope e.g. __traits()?

I think it's a missing feature. But what are the use cases? In 
Python it's easy to do, but it's not a common need.


 2. Can you statically iterate over every field of a class?
 It seems like the only returns are string lists from __traits().
 I basically am interested in being able to generate a
 PrettyPrint/Serializable/Hash template without passing in every
 single field.

With their names and mixin() then maybe you able to do what you 
look for.


 3. Is there any way of executing code or programs during compile
 time?

Only CTFE.


 I've seen an example of CTFE (Compile Time Function Evaluation),
 although I'm unsure if this works for stuff like classes.

CTFE now works with classes too.


 However, I am considering more advanced execution (not 
 constants)
 such as printing to a file during compiling for stuff like how
 long compiling a certain function/template takes.

With the proposed __ctWrtiteln you are able to print things at 
compile time to standard output. But there is no enough 
introspection to know how much time it takes to compile part of a 
program.



 This is more of a rhetorical... as I know the language doesn't
 need them, nor would I know if they would be binary/unary
 prefix/etc or the precedence... although they would be nice to
 have. Specifically I'd like $prefix to be stringification.

The operator overloading syntax is string-based, so adding new 
operators is not too much hard. But I don't know what are the 
advantages and disadvantages. The $ operator is used (in a 
certain context) to denote the length of collections.

Bye,
bearophile

"Wes" <ffhighwind hotmail.com> writes:

 I think it's a missing feature. But what are the use cases? In 
 Python it's easy to do, but it's not a common need.

True, I suppose. It just seems like the programmer could have 
access to some sort of associative array, where you could have 
multiple stored values per key sorted by scope.

 With their names and mixin() then maybe you able to do what you 
 look for.

So basically there's no easy way to make a prettyprint like:
foreach(field;__traits(allFields, myClass)) { str ~= field; }

 With the proposed __ctWrtiteln you are able to print things at 
 compile time to standard output. But there is no enough 
 introspection to know how much time it takes to compile part of 
 a program.

Assuming this is the same as #pragma (msg, ""). I just meant 
actually calling code that runs at compile time that isn't a 
static-ly defined constant.

 The operator overloading syntax is string-based, so adding new 
 operators is not too much hard. But I don't know what are the 
 advantages and disadvantages. The $ operator is used (in a 
 certain context) to denote the length of collections.

I'm not sure what you mean. Do you mean I can go edit the open 
source compiler and add in my own language feature? Or does the 
ability to add a $/  operator already exist?

"bearophile" <bearophileHUGS lycos.com> writes:

Wes:

 So basically there's no easy way to make a prettyprint like:
 foreach(field;__traits(allFields, myClass)) { str ~= field; }

I am saying that probably there is a way, using mixins.


 Assuming this is the same as #pragma (msg, "").

It's not the same. It's usable just like a writeln, but in CTFE 
too. pragma(msg) has a different behavour, and happens in a 
different compilation stage.


 I'm not sure what you mean. Do you mean I can go edit the open 
 source compiler and add in my own language feature? Or does the 
 ability to add a $/  operator already exist?

I mean that D compiler writers don't need to introduce new syntax 
to add that feature. But I don't see lot of people asking for it.

Bye,
bearophile

travert phare.normalesup.org (Christophe Travert) writes:

"bearophile" , dans le message (digitalmars.D:173297), a écrit :
 I'm not sure what you mean. Do you mean I can go edit the open 
 source compiler and add in my own language feature? Or does the 
 ability to add a $/  operator already exist?

 
 I mean that D compiler writers don't need to introduce new syntax 
 to add that feature. But I don't see lot of people asking for it.

The $ prefix operator *is* a new syntax. It impacts the lexer and parser 
and they would have to create new precedence rules. The only think that 
already exists is the syntax to overload such operator, if it existed.

I don't see anything like new operators comming in the langage before 
long.

-- 
Christophe

Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:

On 7/26/12 7:44 AM, bearophile wrote:
 Wes:

 Can you do this in D? If not then what other language?

 Maybe CommonLisp and Scheme (and maybe Clojure) are able to do what you
 ask for (but normally they don't have the stringification with $).


 1. Can you get a list of all classes and variables in the current
 scope e.g. __traits()?

 I think it's a missing feature. But what are the use cases? In Python
 it's easy to do, but it's not a common need.

What happened to __traits(allMembers, module_name)? It doesn't list the 
contents of the current scope, but instead that of a module (including 
the current one).

 2. Can you statically iterate over every field of a class?
 It seems like the only returns are string lists from __traits().
 I basically am interested in being able to generate a
 PrettyPrint/Serializable/Hash template without passing in every
 single field.

 With their names and mixin() then maybe you able to do what you look for.


 3. Is there any way of executing code or programs during compile
 time?

 Only CTFE.

The OP is likely new to the community, so throwing acronyms around may 
not be helpful.

CTFE stands for Compile-Time Function Evaluation. In short you get to 
evaluate a subset of D during compilation if you force a function call 
in a static evaluation context, such as initializing an enum or a static.


Andrei

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

On 7/26/12, Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> wrote:
 The OP is likely new to the community, so throwing acronyms around may
 not be helpful.

http://dlang.org/glossary.html

Now if someone would stop fiddling with the damn javascript which
keeps erroring out.. I'm beginning to join the Nick camp w.r.t. JS. :p

Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:

On Thu, 26 Jul 2012 17:17:39 +0200
Andrej Mitrovic <andrej.mitrovich gmail.com> wrote:

 On 7/26/12, Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> wrote:
 The OP is likely new to the community, so throwing acronyms around
 may not be helpful.

 
 http://dlang.org/glossary.html
 
 Now if someone would stop fiddling with the damn javascript which
 keeps erroring out.. I'm beginning to join the Nick camp w.r.t. JS. :p

JavaScript is a technology (from the makers of clearly the greatest web
browser of all time - Netscape) whose primary usage, and indeed killer
feature, is to enable people to develop new and exciting ways to make
the web slower, more broken, and more obnoxious than could ever be
achieved with mere HTML/CSS alone. This, of course, is a major win for
humanity in general, and is something we can all be proud of.

http://semitwist.com/articles/article/view/my-awesome-web-development-game
 

"H. S. Teoh" <hsteoh quickfur.ath.cx> writes:

On Thu, Jul 26, 2012 at 05:59:08PM -0400, Nick Sabalausky wrote:
 On Thu, 26 Jul 2012 17:17:39 +0200
 Andrej Mitrovic <andrej.mitrovich gmail.com> wrote:

[...]
 Now if someone would stop fiddling with the damn javascript which
 keeps erroring out.. I'm beginning to join the Nick camp w.r.t. JS.
 :p

 
 JavaScript is a technology (from the makers of clearly the greatest
 web browser of all time - Netscape) whose primary usage, and indeed
 killer feature, is to enable people to develop new and exciting ways
 to make the web slower, more broken, and more obnoxious than could
 ever be achieved with mere HTML/CSS alone. This, of course, is a major
 win for humanity in general, and is something we can all be proud of.
 
 http://semitwist.com/articles/article/view/my-awesome-web-development-game

I'm guilty of turning <i> into <em>, actually. You may shoot me now.

(Actually, I resisted many attempts at introducing bold and italics
buttons on my web-based editing page. I even introduced a way of
indicating which table cells are headers, so that styling will work
properly instead of half-assed manual bolding. But no, the users
clamored for bold and italics until they threatened to mob me (ok, that
last part may be an exaggeration), and so the horrible bold/italics
buttons appeared. So far, I've resisted adding a "red" button
(apparently, bold and italics aren't enough, sometimes you want
something in RED too). I managed to evade that one by having a select
box for the status field of the items, one option of which is styled
red, and insisting that's the only place red is allowed. We'll see how
long I hold out on that one.)


T

-- 
Those who don't understand Unix are condemned to reinvent it, poorly.

Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:

On Thu, 26 Jul 2012 16:06:47 -0700
"H. S. Teoh" <hsteoh quickfur.ath.cx> wrote:

 On Thu, Jul 26, 2012 at 05:59:08PM -0400, Nick Sabalausky wrote:
 On Thu, 26 Jul 2012 17:17:39 +0200
 Andrej Mitrovic <andrej.mitrovich gmail.com> wrote:

 [...]
 Now if someone would stop fiddling with the damn javascript which
 keeps erroring out.. I'm beginning to join the Nick camp w.r.t.
 JS. :p

 
 JavaScript is a technology (from the makers of clearly the greatest
 web browser of all time - Netscape) whose primary usage, and indeed
 killer feature, is to enable people to develop new and exciting ways
 to make the web slower, more broken, and more obnoxious than could
 ever be achieved with mere HTML/CSS alone. This, of course, is a
 major win for humanity in general, and is something we can all be
 proud of.
 
 http://semitwist.com/articles/article/view/my-awesome-web-development-game

 
 I'm guilty of turning <i> into <em>, actually. You may shoot me now.
 

Heh, yea, I've actually done it too (in hand-written HTML though, not
mechanically). You know, just following standard recommended practices.
Then it hit me: "WTF am I doing this for? My whole *intent* is for this
to be bold/italic, why am I beating around the bush? Just use <b>/<i>
and be done with it, Nick!" Now my posts are filled with <b> and <i>
tags, and maybe it's just my hunger for rebelling, but I've never
had a problem with it and I couldn't be happier :)

 (Actually, I resisted many attempts at introducing bold and italics
 buttons on my web-based editing page. I even introduced a way of
 indicating which table cells are headers, so that styling will work
 properly instead of half-assed manual bolding. But no, the users
 clamored for bold and italics until they threatened to mob me (ok,
 that last part may be an exaggeration), and so the horrible
 bold/italics buttons appeared. So far, I've resisted adding a "red"
 button (apparently, bold and italics aren't enough, sometimes you want
 something in RED too). I managed to evade that one by having a select
 box for the status field of the items, one option of which is styled
 red, and insisting that's the only place red is allowed. We'll see how
 long I hold out on that one.)
 

Yea, there's certainly benefits to semantic tagging, but I think it's
become too religious and under-pragmatic.

I would draw the line at "I want blinking/scrolling text" though ;)

Brad Anderson <eco gnuk.net> writes:

On Thu, Jul 26, 2012 at 9:17 AM, Andrej Mitrovic <andrej.mitrovich gmail.com
 wrote:

 On 7/26/12, Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> wrote:
 The OP is likely new to the community, so throwing acronyms around may
 not be helpful.

 http://dlang.org/glossary.html

 Now if someone would stop fiddling with the damn javascript which
 keeps erroring out.. I'm beginning to join the Nick camp w.r.t. JS. :p

What are the errors and what browser (with version) are you using? There
were errors in the Chrome Beta release until recently (they stopped after
we updated hyphenator.js).

I just tested in:
- Chrome 21.0.1180.55
- Firefox 8.0.1
- Firefox 12.0
- Firefox 14.0.1
- Opera 10.63
- Opera 12.00
- IE 9.0.8112.16421

and saw nothing wrong.

Jacob Carlborg <doob me.com> writes:

On 2012-07-26 16:53, Andrei Alexandrescu wrote:

 What happened to __traits(allMembers, module_name)? It doesn't list the
 contents of the current scope, but instead that of a module (including
 the current one).

It can be used on classes and structs (I assume) as well,

-- 
/Jacob Carlborg

captaindet <2krnk gmx.net> writes:

 What happened to __traits(allMembers, module_name)? It doesn't list
 the contents of the current scope, but instead that of a module
 (including the current one).


that could have saved me too ... now i tried getting it working, but to no
avail. i mean,  __traits(allMembers, X) works for class X fine, but fails on
modules. using dmd 2.059 for win and some 2.060-ish phobos from some weeks ago.

could anyone provide a working code example pls?

cheers,

det

captaindet <2krnk gmx.net> writes:

On 2012-07-30 00:11, captaindet wrote:
 What happened to __traits(allMembers, module_name)? It doesn't list
 the contents of the current scope, but instead that of a module
 (including the current one).


 that could have saved me too ... now i tried getting it working, but to no
avail. i mean, __traits(allMembers, X) works for class X fine, but fails on
modules. using dmd 2.059 for win and some 2.060-ish phobos from some weeks ago.

 could anyone provide a working code example pls?

 cheers,

 det

my bad, figured it out

Jacob Carlborg <doob me.com> writes:

On 2012-07-26 13:20, Wes wrote:
 Can you do this in D? If not then what other language?

I'm pretty sure you can do all this using Ruby.

 1. Can you get a list of all classes and variables in the current
 scope e.g. __traits()?

In general, no. But you can get a list of all members in a module or 
class scope. Just use __traits(allMembers, ...);

 2. Can you statically iterate over every field of a class?
 It seems like the only returns are string lists from __traits().
 I basically am interested in being able to generate a
 PrettyPrint/Serializable/Hash template without passing in every
 single field.

Yes, if __traits doesn't work it works with .tupleof. For serialization 
Orange is available:

https://github.com/jacob-carlborg/orange

 3. Is there any way of executing code or programs during compile
 time?
 I've seen an example of CTFE (Compile Time Function Evaluation),
 although I'm unsure if this works for stuff like classes.
 However, I am considering more advanced execution (not constants)
 such as printing to a file during compiling for stuff like how
 long compiling a certain function/template takes.

Classes should work. It's not possible to print to a file.

 4. Is there anything like Go/Erlang channels? (synchronous queues)
 I assume that synchronized blocks with a shared global dynamic
 array would be the solution without the special syntax sugar?

I'm not sure. but you could have a look at std.concurrency. 
std.parallelism might be of interest as well.

This seems to good to read as well: 
http://www.informit.com/articles/article.aspx?p=1609144

-- 
/Jacob Carlborg

"Wes" <ffhighwind hotmail.com> writes:

 In general, no. But you can get a list of all members in a 
 module or class scope. Just use __traits(allMembers, ...);

The allMembers example was lacking fields. It only showed 
methods, so I wasn't really sure. The problem with it was that it 
returned strings. I suppose I can still use them in a mixin.

 Yes, if __traits doesn't work it works with .tupleof. For 
 serialization Orange is available:

 https://github.com/jacob-carlborg/orange

Awesome. Definitely sounds like what I'm looking for.

Jacob Carlborg <doob me.com> writes:

On 2012-07-26 14:17, Wes wrote:
 In general, no. But you can get a list of all members in a module or
 class scope. Just use __traits(allMembers, ...);

 The allMembers example was lacking fields. It only showed methods, so I
 wasn't really sure. The problem with it was that it returned strings. I
 suppose I can still use them in a mixin.

If you want to access the fields of a class or struct you can use .tupleof:

class Foo { int a; }
Foo.tupleof;

You can have a look how this is used in my serialization library Orange:

https://github.com/jacob-carlborg/orange/blob/master/orange/util/Reflection.d#L188

Interesting templates/functions would be:

* hasField
* fieldsOf
* TypeOfField
* nameOfFieldAt
* setValueOfField
* getValueOfField

-- 
/Jacob Carlborg

"David Nadlinger" <see klickverbot.at> writes:

On Thursday, 26 July 2012 at 12:17:23 UTC, Wes wrote:
 In general, no. But you can get a list of all members in a 
 module or class scope. Just use __traits(allMembers, ...);

 The allMembers example was lacking fields. It only showed 
 methods, so I wasn't really sure. The problem with it was that 
 it returned strings. I suppose I can still use them in a mixin.

If you don't want to use mixins, you can also use 
__traits(getMember, instance, name). mixin("instance." ~ name) is 
shorter, though.

David

"David Piepgrass" <qwertie256 gmail.com> writes:

 3. Is there any way of executing code or programs during compile
 time?
 I've seen an example of CTFE (Compile Time Function Evaluation),
 although I'm unsure if this works for stuff like classes.
 However, I am considering more advanced execution (not 
 constants)
 such as printing to a file during compiling for stuff like how
 long compiling a certain function/template takes.

You can call any safe and pure D code at compile time (none of 
the code has to be marked pure explicitly, but it cannot access 
any static or global variables, call C code, access files, etc.) 
This is called CTFE=Compile-Time Function Evaluation.

The "pure" limitation isn't a huge restriction, since you can 
still edit member variables (fields) and the compiler can memoize 
the results of CTFE... although I don't know if it memoizes 
automatically, or if you have to use a template to accomplish it. 
For example if I do

enum twoPi = computePi() + computePi();

I don't know if the compiler computes PI once or twice. Does 
someone know? But if I define this template:

 property auto memoize(T, T code)() { return code; }

enum twoPi = memoize!(double,computePi()) + 
memoize!(double,computePi());

Then computePi is surely called only once, and thus you can cache 
the result of any computation for repeated use. (I don't know how 
to get the type 'double' to be inferred automatically, though.) 
You can also, of course, use enums for this purpose:

enum pi = computePi(); // computed only once
enum twoPi = pi + pi;

I don't think you can run "programs" at compile-time, but since 
you can call ordinary functions and use arbitrarily large 
structs, you can accomplish a lot. I believe the current released 
build, 2.059, can't use classes at compile time, but bearophile 
just implied that 2.060 can.


 This is more of a rhetorical... as I know the language doesn't
 need them, nor would I know if they would be binary/unary
 prefix/etc or the precedence... although they would be nice to
 have. Specifically I'd like $prefix to be stringification.

Just to clarify, because other people are making it sound like D 
could do this... no, D does not offer user-defined operators, 
only overloading of predefined operators. User-defined ops would 
certainly be a nice feature that I would like to have, but the D 
developers have too much to do already. Personally I think the D 
syntax and rules feel too ad-hoc and unintuitive right now; it 
should be simplified slightly, formalized more clearly, and 
debugged further before yet more features are piled on.

Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:

On Thu, 26 Jul 2012 19:35:56 +0200
"David Piepgrass" <qwertie256 gmail.com> wrote:
 For example if I do
 
 enum twoPi = computePi() + computePi();
 
 I don't know if the compiler computes PI once or twice. Does 
 someone know? But if I define this template:
 

You could toss a ctfeWriteln (or whatever it's named) into
'computePi' and find out.

It shouldn't NEED to evaluate it twice since CTFE isn't allowed to
have any global mutable state.

"Araq" <rumpf_a web.de> writes:

On Thursday, 26 July 2012 at 11:20:35 UTC, Wes wrote:
 Can you do this in D? If not then what other language?

Nimrod (http://nimrod-code.org/) can do 2, 3, 4, 5 of your list. 
(1 is planned.)
2 can be done very efficiently directly for tuples; for classes 
it can currently only be done with RTTI. And Nimrod uses $ for 
stringification. ;-)

"David Nadlinger" <see klickverbot.at> writes:

On Thursday, 26 July 2012 at 19:02:15 UTC, Araq wrote:
 ;-)

Further context: Araq is the guy behind Nimrod.

Man, we should totally infiltrate their forums as well.

David,
serious as ever

"Araq" <rumpf_a web.de> writes:

On Thursday, 26 July 2012 at 19:38:41 UTC, David Nadlinger wrote:
 On Thursday, 26 July 2012 at 19:02:15 UTC, Araq wrote:
 ;-)

 Further context: Araq is the guy behind Nimrod.

 Man, we should totally infiltrate their forums as well.

 David,
 serious as ever

He explicitely asked about other programming languages...

"Simen Kjaeraas" <simen.kjaras gmail.com> writes:

On Thu, 26 Jul 2012 21:44:24 +0200, Araq <rumpf_a web.de> wrote:

 On Thursday, 26 July 2012 at 19:38:41 UTC, David Nadlinger wrote:
 On Thursday, 26 July 2012 at 19:02:15 UTC, Araq wrote:
 ;-)

 Further context: Araq is the guy behind Nimrod.

 Man, we should totally infiltrate their forums as well.

 David,
 serious as ever

 He explicitely asked about other programming languages...

That's it! He's using logic against us!

We're just pulling your leg, Araq. This reminds me, I should
perhaps do more with Nimrod than just install it...

-- 
Simen

Walter Bright <newshound2 digitalmars.com> writes:

On 7/26/2012 12:44 PM, Araq wrote:
 He explicitely asked about other programming languages...


Araq, you are welcome here.

"David Nadlinger" <see klickverbot.at> writes:

On Thursday, 26 July 2012 at 23:57:32 UTC, Walter Bright wrote:
 On 7/26/2012 12:44 PM, Araq wrote:
 He explicitely asked about other programming languages...

 Araq, you are welcome here.

Yes, of course. If anything, I'd wish there was _more_ 
interaction between PL communities, not less. My post was just an 
attempt at a little humor – I hoped this was clear from the, 
uhm, style.

David

Walter Bright <newshound2 digitalmars.com> writes:

On 7/26/2012 11:59 PM, David Nadlinger wrote:
 On Thursday, 26 July 2012 at 23:57:32 UTC, Walter Bright wrote:
 On 7/26/2012 12:44 PM, Araq wrote:
 He explicitely asked about other programming languages...

 Araq, you are welcome here.

 Yes, of course. If anything, I'd wish there was _more_ interaction between PL
 communities, not less. My post was just an attempt at a little humor – I
hoped
 this was clear from the, uhm, style.

No problem, I just wanted to make sure there was no misunderstanding.

Walter Bright <newshound2 digitalmars.com> writes:

On 7/26/2012 4:20 AM, Wes wrote:

 This is more of a rhetorical... as I know the language doesn't
 need them, nor would I know if they would be binary/unary
 prefix/etc or the precedence... although they would be nice to
 have.

The question shouldn't be framed as "why not X?" but as "does the utility of 
having X justify placing in in the language?"




 Specifically I'd like $prefix to be stringification.

Can you please be more specific about what this might do?

"Wes" <ffhighwind hotmail.com> writes:

On Thursday, 26 July 2012 at 23:55:29 UTC, Walter Bright wrote:
 On 7/26/2012 4:20 AM, Wes wrote:

 This is more of a rhetorical... as I know the language doesn't
 need them, nor would I know if they would be binary/unary
 prefix/etc or the precedence... although they would be nice to
 have.

 The question shouldn't be framed as "why not X?" but as "does 
 the utility of having X justify placing in in the language?"




 Specifically I'd like $prefix to be stringification.

 Can you please be more specific about what this might do?

It's not for more functionality but more for programmer usability 
and terse coding. I just noticed the .stringof being used in some 

typing.

I was starting development of my own language, and looked more at 
what I could do with D and saw that it included pretty much 
everything I wanted, except the stuff I've mentioned. I wanted my 
language to stretch the boundaries of language/compiler or 
metaprogramming.

The following are potential features/lackings of the language (no 
offense to the language developers as I am very impressed 
overall):

1. It has *practically forces use of the GC, since turning it off 
means I can't use many features of the language. This isn't as 
big of a deal to me as it is to many C++ devs (my assumption). A 
GC gives the impression of slow managed, even if the language 
isn't scripted/bytecode.


operators or different variable attributes like myconst).
I don't see this as a major flaw as I don't know of any other 
language other than scheme/lisp/ratchet that allow this.

3. It can't run *all* forms of code at compile time.

4. It doesn't have a simple prettyprint operator. I think 
.stringof is a big step forward from .ToString() for simplicity, 
but obviously $variable would be more terse.

5. It doesn't have a way to iterate over every id in scope.
I can't think of a good reason to do this anyway.

Walter Bright <newshound2 digitalmars.com> writes:

On 7/26/2012 6:18 PM, Wes wrote:
 1. It has *practically forces use of the GC, since turning it off means I can't
 use many features of the language. This isn't as big of a deal to me as it is
to
 many C++ devs (my assumption). A GC gives the impression of slow managed, even
 if the language isn't scripted/bytecode.

C++ doesn't have features that would require a GC - not using the GC in D does 
not mean the language is crippled to be less than C++.



 different variable attributes like myconst).
 I don't see this as a major flaw as I don't know of any other language other
 than scheme/lisp/ratchet that allow this.

We've explicitly decided not to do user defined syntax.


 3. It can't run *all* forms of code at compile time.

 4. It doesn't have a simple prettyprint operator. I think .stringof is a big
 step forward from .ToString() for simplicity, but obviously $variable would be
 more terse.

 5. It doesn't have a way to iterate over every id in scope.
 I can't think of a good reason to do this anyway.

We try not to add features just because we can - we try to do ones only if they 
have compelling use cases.

=?UTF-8?B?QWxleCBSw7hubmUgUGV0ZXJzZW4=?= <alex lycus.org> writes:

On 27-07-2012 06:56, Walter Bright wrote:
 On 7/26/2012 6:18 PM, Wes wrote:
 1. It has *practically forces use of the GC, since turning it off
 means I can't
 use many features of the language. This isn't as big of a deal to me
 as it is to
 many C++ devs (my assumption). A GC gives the impression of slow
 managed, even
 if the language isn't scripted/bytecode.

 C++ doesn't have features that would require a GC - not using the GC in
 D does not mean the language is crippled to be less than C++.

new and assert both result in GC allocation and are not easily replaced 
with functions that use manual memory management. You have to replace 
functions in druntime to get the desired effect. druntime isn't part of 
the language specification, so, strictly speaking, not using the GC in D 
/does/ make the language slightly more crippled than C++.


 operators or
 different variable attributes like myconst).
 I don't see this as a major flaw as I don't know of any other language
 other
 than scheme/lisp/ratchet that allow this.

 We've explicitly decided not to do user defined syntax.


 3. It can't run *all* forms of code at compile time.

 4. It doesn't have a simple prettyprint operator. I think .stringof is
 a big
 step forward from .ToString() for simplicity, but obviously $variable
 would be
 more terse.

 5. It doesn't have a way to iterate over every id in scope.
 I can't think of a good reason to do this anyway.

 We try not to add features just because we can - we try to do ones only
 if they have compelling use cases.

-- 
Alex Rønne Petersen
alex lycus.org
http://lycus.org

Jacob Carlborg <doob me.com> writes:

On 2012-07-27 03:18, Wes wrote:

 4. It doesn't have a simple prettyprint operator. I think .stringof is a
 big step forward from .ToString() for simplicity, but obviously
 $variable would be more terse.

.stringof is not the equivalent to .ToString(). .stringof is more of a 
compile time language feature. It turns an expression into a string:

import std.stdio;

void main ()
{
     string foo = "bar";
     writeln(foo.stringof);
     writeln((3 + 4).stringof);
}

The above code will print:

foo
3 + 4

The equivalent to .ToString() in D would be .toString() or 
to!(string)(variable).

 5. It doesn't have a way to iterate over every id in scope.
 I can't think of a good reason to do this anyway.

There could be a couple of reasons for doing this. For example creating 
a DSL.

-- 
/Jacob Carlborg

Jacob Carlborg <doob me.com> writes:

On 2012-07-27 01:55, Walter Bright wrote:




  > Specifically I'd like $prefix to be stringification.

 Can you please be more specific about what this might do?

If I understand him correctly I think:

$variable

Would be the same as:

to!(string)(variable)

And/or:

"foo $variable bar"

Would be the same as:

"foo " ~ to!(string)(variable) ~ " bar"

The later would be string interpolation.

-- 
/Jacob Carlborg