• Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> Jun 17 2009
• BCS <ao pathlink.com> Jun 17 2009
• Ary Borenszweig <ary esperanto.org.ar> Jun 17 2009
• Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> Jun 18 2009
• Ary Borenszweig <ary esperanto.org.ar> Jun 19 2009
• Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> Jun 19 2009

Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> writes:

Is there a way to see the source for a D program after the mixins and
templates, etc., are expanded?

I get occasional error messages saying I've got incompatible types, for
example, but the error message only makes sense when I mentally instantiate a
template. Is there a way to see this instantiation in print?

At what phase of compilation does this occur? Has the code been tokenized?
(It's okay with me if it is--I just want to know.)

Paul

BCS <ao pathlink.com> writes:

Reply to Paul,

 Is there a way to see the source for a D program after the mixins and
 templates, etc., are expanded?
 


Descent has a "compile time view" that is supposed to do some of this.

If you are only looking at string mixins you can use pragma(msg, string) 
to view the string

 I get occasional error messages saying I've got incompatible types,
 for example, but the error message only makes sense when I mentally
 instantiate a template. Is there a way to see this instantiation in
 print?
 
 At what phase of compilation does this occur? Has the code been
 tokenized? (It's okay with me if it is--I just want to know.)
 


This is mostly informed guess work but: it's all mixed up; the file is
tokenized 
and parsed, then some semantic stuff happens including some template evaluation 
and CTFE to construct the string that is then tokenized and your back where 
you started. In concept, the tokenization, parsing and semantics are different 
phases. The first two might happen at the same time but I think the third 
is separate.

 Paul
 

Ary Borenszweig <ary esperanto.org.ar> writes:

BCS escribió:
 Reply to Paul,
 
 Is there a way to see the source for a D program after the mixins and
 templates, etc., are expanded?


 Descent has a "compile time view" that is supposed to do some of this.
 
 If you are only looking at string mixins you can use pragma(msg, string) 
 to view the string
 
 I get occasional error messages saying I've got incompatible types,
 for example, but the error message only makes sense when I mentally
 instantiate a template.




You can also debug the template instantiation with Descent.

But only for D1, aw...

Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> writes:

Paul D. Anderson Wrote:

 Is there a way to see the source for a D program after the mixins and
templates, etc., are expanded?
 
 I get occasional error messages saying I've got incompatible types, for
example, but the error message only makes sense when I mentally instantiate a
template. Is there a way to see this instantiation in print?
 
 At what phase of compilation does this occur? Has the code been tokenized?
(It's okay with me if it is--I just want to know.)
 
 Paul
 


Thanks for the help. So the short answer is no.

A related question: which D language constructs do (virtual) text substitution?

My list:
* alias
* mixin
* special tokens
* special token sequences
* templates
* template mixins

Are there others?

Paul

Ary Borenszweig <ary esperanto.org.ar> writes:

Paul D. Anderson wrote:
 Paul D. Anderson Wrote:
 
 Is there a way to see the source for a D program after the mixins and
templates, etc., are expanded?

 I get occasional error messages saying I've got incompatible types, for
example, but the error message only makes sense when I mentally instantiate a
template. Is there a way to see this instantiation in print?

 At what phase of compilation does this occur? Has the code been tokenized?
(It's okay with me if it is--I just want to know.)

 Paul


 Thanks for the help. So the short answer is no.


I thought the short answer was yes. Did you try Descent? You can hover a 
template instance to see it's expansion. Same goes with mixins. Doesn't 
this help you?

Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> writes:

Ary Borenszweig Wrote:

 Paul D. Anderson wrote:
 Paul D. Anderson Wrote:
 
 Is there a way to see the source for a D program after the mixins and
templates, etc., are expanded?

 I get occasional error messages saying I've got incompatible types, for
example, but the error message only makes sense when I mentally instantiate a
template. Is there a way to see this instantiation in print?

 At what phase of compilation does this occur? Has the code been tokenized?
(It's okay with me if it is--I just want to know.)

 Paul


 Thanks for the help. So the short answer is no.


 I thought the short answer was yes. Did you try Descent? You can hover a 
 template instance to see it's expansion. Same goes with mixins. Doesn't 
 this help you?


Sorry, didn't mean to sound ungrateful. I appreciate your help, but...

I don't use Descent -- the last time I tried to use it (and I know it's been
improved a lot) I couldn't get it installed properly.

But it sounds like Descent does what I want. I was looking for a non-IDE
specific variety. And, oh yeah, I need it for D2. Maybe I'll peek at the
Descent implementation and see if I can't come up with something that I can
pull from there.

Again, thanks.

Paul