[patch, fortran] PR20373 - intrinsic symbols can be given the wrong type (take 3)

[Daniel Franke]

Following a private discussion with Janne and Paul's comments on the list, I 
changed my last proposal slightly:

The warning for _any_ type specified together with an INTRINSIC keyword now 
reads:

	INTEGER, INTRINSIC :: SIN
	                        1
	Warning: Type specified for intrinsic function 'sin' at (1) is ignored

Reasoning: 
  * comparing with the internally available type may give spurious
    warnings (e.g. "complex, intrinsic :: sum"; SUM has default type REAL)
  * enumerating allowed types for each intrinsic is not an option
  * the specified type IS ignored
  * the behaviour is consistent with that of the SUN compiler


The testcase was adjusted accordingly. Further, the patch is now correctly 
tabified.

If there are no objections, I will commit after 48 hours.


gcc/fortran:
2007-06-20  Daniel Franke  <franke.daniel@gmail.com>

       PR fortran/20373
       * intrinsic.c (add_functions): Additional function types.
       * resolve.c (resolve_symbol): Added type checks to explicitly defined
       intrinsics.

gcc/testsuite:
2007-06-20  Daniel Franke  <franke.daniel@gmail.com>

       PR fortran/20373
       * gfortran.dg/intrinsic.f90: New test.


Regards
	Daniel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pr20373.patch
Type: text/x-diff
Size: 6766 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20070620/47058da9/attachment.bin>