Re: [Patch, fortran] PR51976 - [F2003] Support deferred-length character components of derived types (allocatable string length)

[Tobias Burnus <burnus at net-b dot de>]

Paul Richard Thomas wrote:
> The attached patch represents progress to date.  It fixes the original
> problem in this PR and allows John Reid's version of
> iso_varying_string/vocabulary_word_count.f90 to compile and run
> correctly.  It even bootstraps and regtests!

Great! It seems as if 4.9 will get a much-wished feature soon after 4.8 
is branched.

> +   /* Is a parameter associated with a deferred type component.  */
> +   unsigned deferred_parameter:1;

I quickly glanced at the patch and I am still not sure what 
"deferred_parameter" is exactly doing nor what "parameter" means in this 
context. I know a named constant ("PARAMETER") and (kind/length) type 
parameters of intrinsic types and (via "kind" and "len") derived types. 
Can you make it a bit clearer what this attribute is good for? – Do you 
need to handle it in module.c or in dump-parse-tree.c?

(A while later)

Looking at the patch a bit more carefully, "deferred_parameter" seems to 
be set for the hidden length component to a deferred-length character 
component. I think it is not really a (type) parameter; maybe 
"deferred_length_comp"?


>          /* If it was not found, try the default initializer if there's any;
> ! 	 otherwise, it's an error.  */
>          if (!comp_iter)
>    	{
>    	  if (comp->initializer)
> --- 2349,2355 ----
>          /* If it was not found, try the default initializer if there's any;
> ! 	 otherwise, it's an error unless this is a deferred parameter.  */

"a hidden deferred-length component"?

> +       /* Add the hidden deferred length field.  */

Or "field" ;-)


Otherwise, it looks okay to me – but I have not yet fully studied the 
patch nor applied and tried to break it ;-)
For instance, trying "select type(comp%str); case (character(len=*); 
print *, len(comp%str)" where "str" is "class(*)".

Tobias