[Bug fortran/52335] New: [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[Bug fortran/52335] New: [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

             Bug #: 52335
           Summary: [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects
                    valid DELIM= in OPEN
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Keywords: rejects-valid
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: burnus@gcc.gnu.org


http://groups.google.com/group/comp.lang.fortran/browse_thread/thread/9942915c9c41b151


open(unit=lunnml,file=nmlfile,delim='apostrophe',status='old') 
                                                               1
Error: Fortran 2003: DELIM= at (1) not allowed in Fortran 95


However, Fortran 95 has:

  R904 open-stmt     is   OPEN ( connect-spec-list )
  R905 connect-spec  is   ...
                     or   DELIM = scalar-default-char-expr

and

"9.3.4.9 DELIM= specifier in the OPEN statement
 The scalar-default-char-expr shall evaluate to APOSTROPHE, QUOTE, or NONE.
[...]"


Note: gfortran correctly allows:
  inquire(nunit, delim=str)



I think the bogus check has been introduced when Fortran 2003's
  "9.5.1.7 DELIM= specifier in a data transfer statement"
was implemented.

The reason seems to be that OPEN and data-transfer (READ/WRITE) share the
"delim=" checking.

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2012-02-22
                 CC|                            |burnus at gcc dot gnu.org
         AssignedTo|unassigned at gcc dot       |burnus at gcc dot gnu.org
                   |gnu.org                     |
   Target Milestone|---                         |4.4.7
     Ever Confirmed|0                           |1

--- Comment #1 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-22 08:04:02 UTC ---
Actually, the check is in (fortran/io.c's) gfc_match_open, thus, one can simply
remove it without taking care of data transfer statements.

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4

--- Comment #2 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-22 10:03:46 UTC ---
Patch:
  http://gcc.gnu.org/ml/fortran/2012-02/msg00090.html

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

--- Comment #3 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-22 17:25:21 UTC ---
Author: burnus
Date: Wed Feb 22 17:25:11 2012
New Revision: 184485

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184485
Log:
2012-02-22  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * io.c (gfc_match_open): Remove bogus F2003 DELIM= check.

2012-02-22  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * gfortran.dg/io_constraints_10.f90: New.


Added:
    trunk/gcc/testsuite/gfortran.dg/io_constraints_10.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/io.c
    trunk/gcc/testsuite/ChangeLog

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-23 22:53:58 UTC ---
Author: burnus
Date: Thu Feb 23 22:53:54 2012
New Revision: 184534

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184534
Log:
2012-02-23  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * io.c (gfc_match_open): Remove bogus F2003 DELIM= check.

2012-02-23  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * gfortran.dg/io_constraints_10.f90: New.


Added:
    branches/gcc-4_6-branch/gcc/testsuite/gfortran.dg/io_constraints_10.f90
Modified:
    branches/gcc-4_6-branch/gcc/fortran/ChangeLog
    branches/gcc-4_6-branch/gcc/fortran/io.c
    branches/gcc-4_6-branch/gcc/testsuite/ChangeLog

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

--- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-23 22:54:31 UTC ---
Author: burnus
Date: Thu Feb 23 22:54:26 2012
New Revision: 184535

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184535
Log:
2012-02-23  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * io.c (gfc_match_open): Remove bogus F2003 DELIM= check.

2012-02-23  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52335
        * gfortran.dg/io_constraints_10.f90: New.


Added:
    branches/gcc-4_5-branch/gcc/testsuite/gfortran.dg/io_constraints_10.f90
Modified:
    branches/gcc-4_5-branch/gcc/fortran/ChangeLog
    branches/gcc-4_5-branch/gcc/fortran/io.c
    branches/gcc-4_5-branch/gcc/testsuite/ChangeLog

[Bug fortran/52335] [4.4/4.5/4.6/4.7 Regression] I/O: -std=f95 rejects valid DELIM= in OPEN

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52335

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|4.4.7                       |4.5.4

--- Comment #6 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-23 22:58:14 UTC ---
FIXED on the trunk and the 4.5/4.6 branches. (Possibly in time for GCC 4.6.3,
but later then 4.6.3 RC 1, which uses Rev. 184520.)