digitalmars.D.learn - Emulation macros and pattern matching on D
- Dennis Ritchie (11/11) Jun 05 2015 Macros and operation of pattern matching `=>` in Rust I can write
- Steven Schveighoffer (11/22) Jun 05 2015 string foo(string mode, string value)
- Dennis Ritchie (20/29) Jun 05 2015 Thanks. It looks really simple, but I still do not understand the
- Steven Schveighoffer (11/40) Jun 05 2015 Because what foo is constructing is a string that makes sense in the
- Dennis Ritchie (4/21) Jun 05 2015 Thank you. Now everything is clear. Syntax `${e}` in Rust simpler
- anonymous (22/34) Jun 05 2015 This becomes:
- Dennis Ritchie (3/5) Jun 05 2015 Yes, the sample program really made me think that gravis
- Steven Schveighoffer (9/14) Jun 05 2015 It's just so I didn't have to escape the quotes :D
- Dennis Ritchie (8/16) Jun 05 2015 But such an option is possible to do work? : D
- Steven Schveighoffer (4/19) Jun 05 2015 Yes, unfortunately that version is BSD licensed, won't make it into phob...
- Dennis Ritchie (4/12) Jun 05 2015 I think that the developers of D will be able to write better if
Macros and operation of pattern matching `=>` in Rust I can write
something like this:
macro_rules!foo {
(x => $e:expr) => (println!("mode X: {}", $e));
(y => $e:expr) => (println!("mode Y: {}", $e));
}
fn main() {
foo!(y => 3); // mode Y: 3
foo!(x => 2); // mode X: 2
}
How is it possible to simulate in D?
Jun 05 2015
On 6/5/15 9:01 AM, Dennis Ritchie wrote:
Macros and operation of pattern matching `=>` in Rust I can write
something like this:
macro_rules!foo {
(x => $e:expr) => (println!("mode X: {}", $e));
(y => $e:expr) => (println!("mode Y: {}", $e));
}
fn main() {
foo!(y => 3); // mode Y: 3
foo!(x => 2); // mode X: 2
}
How is it possible to simulate in D?
string foo(string mode, string value)
{
return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`;
}
void main()
{
mixin(foo("Y", "3"));
mixin(foo("X", "2"));
}
-Steve
Jun 05 2015
On Friday, 5 June 2015 at 13:13:15 UTC, Steven Schveighoffer
wrote:
string foo(string mode, string value)
{
return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`;
}
void main()
{
mixin(foo("Y", "3"));
mixin(foo("X", "2"));
}
Thanks. It looks really simple, but I still do not understand the
concept of using mixins in full.
I do not understand why in this line:
return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`;
use the following syntax:
~ mode ~ , ~ value ~
For example, why here I can simply write:
void main() {
int b = 5;
mixin(`int a = b;`);
assert(a == 5);
}
Why should not I write like this:
void main() {
int b = 5;
mixin(`"int a = " ` ~ b ~ ` ";"`);
assert(a == 5);
}
Jun 05 2015
On 6/5/15 10:15 AM, Dennis Ritchie wrote:On Friday, 5 June 2015 at 13:13:15 UTC, Steven Schveighoffer wrote:Because what foo is constructing is a string that makes sense in the *caller*, not inside foo. What those statements do is concat the *value* of mode (i.e. "Y" or "X") and the *value* of value (i.e. "3" or "2") to the string. It's equivalent to rust using the ${e} to do variable substitution.string foo(string mode, string value) { return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`; } void main() { mixin(foo("Y", "3")); mixin(foo("X", "2")); }Thanks. It looks really simple, but I still do not understand the concept of using mixins in full. I do not understand why in this line: return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`; use the following syntax: ~ mode ~ , ~ value ~For example, why here I can simply write: void main() { int b = 5; mixin(`int a = b;`); assert(a == 5); }Because b makes sense in the context of main.Why should not I write like this: void main() { int b = 5; mixin(`"int a = " ` ~ b ~ ` ";"`); assert(a == 5); }Because it won't compile :) Mixin strings must be constructable at compile time, the value of b depends on runtime. Not to mention that you can't concat strings with ints. -Steve
Jun 05 2015
"Dennis Ritchie" <dennis.ritchie mail.ru> On Friday, 5 June 2015 at 14:23:15 UTC, Steven Schveighoffer wrote:On 6/5/15 10:15 AM, Dennis Ritchie wrote:Thank you. Now everything is clear. Syntax `${e}` in Rust simpler than D ` ~ substitute the value ~ ` :)Thanks. It looks really simple, but I still do not understand the concept of using mixins in full. I do not understand why in this line: return `writefln("mode ` ~ mode ~ `: %s", ` ~ value ~ `);`; use the following syntax: ~ mode ~ , ~ value ~Because what foo is constructing is a string that makes sense in the *caller*, not inside foo. What those statements do is concat the *value* of mode (i.e. "Y" or "X") and the *value* of value (i.e. "3" or "2") to the string. It's equivalent to rust using the ${e} to do variable substitution.
Jun 05 2015
On Friday, 5 June 2015 at 14:15:09 UTC, Dennis Ritchie wrote:For example, why here I can simply write: void main() { int b = 5; mixin(`int a = b;`); assert(a == 5); }This becomes: int b = 5; int a = b; assert(a == 5);Why should not I write like this: void main() { int b = 5; mixin(`"int a = " ` ~ b ~ ` ";"`); assert(a == 5); }This is a train wreck. If you want to use `b` in a mixin, it needs to be a compile time constant, e.g. an enum. You can't concatenate an int with strings. You messed up the quotes. You might be mistaking backticks for something special, but they're basically just quotes. Fixing it to compile: import std.conv: to; enum int b = 5; mixin("int a = " ~ b.to!string ~ ";"); /* alternatively: mixin(`int a = ` ~ b.to!string ~ `;`); */ assert(a == 5); After the mixin this becomes: enum int b = 5; int a = 5; assert(a == 5); Not that here it's `a = 5;`, whereas above it's `a = b;`.
Jun 05 2015
On Friday, 5 June 2015 at 14:31:19 UTC, anonymous wrote:You messed up the quotes. You might be mistaking backticks for something special, but they're basically just quotes.Yes, the sample program really made me think that gravis (backticks) - is something special...
Jun 05 2015
On 6/5/15 11:52 AM, Dennis Ritchie wrote:On Friday, 5 June 2015 at 14:31:19 UTC, anonymous wrote:It's just so I didn't have to escape the quotes :D Otherwise it looks like this: return "writefln(\"mode " ~ mode ~ ": %s\", " ~ value ~ ");" I think we can probably do a compile-time substitution processor like rust (and other languages, including php and ruby) which just uses the variable name inside a string with some escape around it. `~var~` is not very succinct, I like $var better. -SteveYou messed up the quotes. You might be mistaking backticks for something special, but they're basically just quotes.Yes, the sample program really made me think that gravis (backticks) - is something special...
Jun 05 2015
On Friday, 5 June 2015 at 19:08:19 UTC, Steven Schveighoffer wrote:It's just so I didn't have to escape the quotes :D Otherwise it looks like this: return "writefln(\"mode " ~ mode ~ ": %s\", " ~ value ~ ");"But such an option is possible to do work? : D return `writefln(q"["mode]" ` ~ mode ~ `q"[: %s"]", ` ~ value ~ `);`;I think we can probably do a compile-time substitution processor like rust (and other languages, including php and ruby) which just uses the variable name inside a string with some escape around it. `~var~` is not very succinct, I like $var better.It would be very good, because many have long been trying to implement this functionality handmade :) http://www.prowiki.org/wiki4d/wiki.cgi?DanielKeep/shfmt
Jun 05 2015
On 6/5/15 4:03 PM, Dennis Ritchie wrote:On Friday, 5 June 2015 at 19:08:19 UTC, Steven Schveighoffer wrote:This looks horrid :)It's just so I didn't have to escape the quotes :D Otherwise it looks like this: return "writefln(\"mode " ~ mode ~ ": %s\", " ~ value ~ ");"But such an option is possible to do work? : D return `writefln(q"["mode]" ` ~ mode ~ `q"[: %s"]", ` ~ value ~ `);`;Yes, unfortunately that version is BSD licensed, won't make it into phobos. -SteveI think we can probably do a compile-time substitution processor like rust (and other languages, including php and ruby) which just uses the variable name inside a string with some escape around it. `~var~` is not very succinct, I like $var better.It would be very good, because many have long been trying to implement this functionality handmade :) http://www.prowiki.org/wiki4d/wiki.cgi?DanielKeep/shfmt
Jun 05 2015
On Friday, 5 June 2015 at 20:14:51 UTC, Steven Schveighoffer wrote:On 6/5/15 4:03 PM, Dennis Ritchie wrote:I think that the developers of D will be able to write better if they take it! :DIt would be very good, because many have long been trying to implement this functionality handmade :) http://www.prowiki.org/wiki4d/wiki.cgi?DanielKeep/shfmtYes, unfortunately that version is BSD licensed, won't make it into phobos.
Jun 05 2015