Status of allocatable components

[Erik Edelmann <erik dot edelmann at iki dot fi>]

All,

My and Paul T's effort to implement allocatable components of derived
types began with a short period of hard work, followed by a longer
period of inactivity (on my part, at least).  Since the work has begun
to move forwards again, I thought would send a status report:

* Explicit allocation and deallocation works.

* Automatic deallocation when a variable goes out of scope works.

* Code to handle automatic deallocation for INTENT(OUT) arguments and
  function results is there, but I will not be very suprised if it
  doesn't work for all cases. (All parts of the code may have bugs,
  but I suspect this part of the code more than others. Also see my note
  to Paul T. below).

* Allocatable components will be nullified when created (i.e. they will
  start their life with an allocation status defined to "not
  allocated").

* The "automatic deallocation&allocation if necessary" on assignments
  is the main thing left to be done.

* I have probably forgotten something.

I attach the current (partly a bit ugly) version of the patch (someday,
I will clean things up and write some kind of explanation for it), and a
testcase.



Paul,

Changes I've made since last time: I've messed (= probably broken
something) a bit with the deallocation of scalar INTENT(OUT) arguments
in trans-expr.c (gfc_conv_function_call), and tried to make sure
function results aren't deallocated until after they have been used.



        Erik

PS. Next time I read email will be on sunday evening.

Attachment: ac_17.diff
Description: Text document

Attachment: allocatable_components_1.f90
Description: Text document