Re: [PATCH v2 1/2] fortran: enable ptype/whatis for user defined types.

[Doug Evans <xdje42 at gmail dot com>]

On Wed, Dec 11, 2013 at 4:35 AM, Keven Boell <keven.boell@intel.com> wrote:
>         (gdb) ptype type
>         old> No symbol "type" in current context.
>         new> type = Type type
>              integer(kind=4) :: t_i
>              End Type type
>
> 2013-11-19  Sanimir Agovic  <sanimir.agovic@intel.com>
>             Keven Boell  <keven.boell@intel.com>
>
>         * f-exp.y (yylex): Add domain array to enable lookup
>         in multiple domains. Loop over lookup domains and try
>         to find requested symbol. Add STRUCT_DOMAIN to lookup
>         domains to be able to query for user defined types.
>
> testsuite/
>         * gdb.fortran/type.f90: New file.
>         * gdb.fortran/whatis_type.f90: New file.
>
> Signed-off-by: Keven Boell <keven.boell@intel.com>
> ---
>  gdb/f-exp.y                               |   33 ++++++++++++--------
>  gdb/testsuite/gdb.fortran/type.f90        |   28 +++++++++++++++++
>  gdb/testsuite/gdb.fortran/whatis_type.exp |   48 +++++++++++++++++++++++++++++
>  3 files changed, 97 insertions(+), 12 deletions(-)
>  create mode 100644 gdb/testsuite/gdb.fortran/type.f90
>  create mode 100644 gdb/testsuite/gdb.fortran/whatis_type.exp
>
> diff --git a/gdb/f-exp.y b/gdb/f-exp.y
> index 567cd00..a7e59df 100644
> --- a/gdb/f-exp.y
> +++ b/gdb/f-exp.y
> @@ -1175,21 +1175,30 @@ yylex (void)
>      char *tmp = copy_name (yylval.sval);
>      struct symbol *sym;
>      struct field_of_this_result is_a_field_of_this;
> +    enum domain_enum_tag lookup_domains[] = {STRUCT_DOMAIN, VAR_DOMAIN};
> +    int i;
>      int hextype;
> -
> -    /* Initialize this in case we *don't* use it in this call; that
> -       way we can refer to it unconditionally below.  */
> -    memset (&is_a_field_of_this, 0, sizeof (is_a_field_of_this));
> -
> -    sym = lookup_symbol (tmp, expression_context_block,
> -                        VAR_DOMAIN,
> -                        parse_language->la_language == language_cplus
> -                        ? &is_a_field_of_this : NULL);
> -    if (sym && SYMBOL_CLASS (sym) == LOC_TYPEDEF)
> +
> +    for (i = 0; i < ARRAY_SIZE (lookup_domains); ++i)
>        {
> -       yylval.tsym.type = SYMBOL_TYPE (sym);
> -       return TYPENAME;
> +       /* Initialize this in case we *don't* use it in this call; that
> +          way we can refer to it unconditionally below.  */
> +       memset (&is_a_field_of_this, 0, sizeof (is_a_field_of_this));
> +
> +       sym = lookup_symbol (tmp, expression_context_block,
> +                            lookup_domains[i],
> +                            parse_language->la_language == language_cplus
> +                            ? &is_a_field_of_this : NULL);
> +       if (sym && SYMBOL_CLASS (sym) == LOC_TYPEDEF)
> +         {
> +           yylval.tsym.type = SYMBOL_TYPE (sym);
> +           return TYPENAME;
> +         }
> +
> +       if (sym)
> +         break;
>        }
> +
>      yylval.tsym.type
>        = language_lookup_primitive_type_by_name (parse_language,
>                                                 parse_gdbarch, tmp);

I don't have any comments on this part of the patch,
Looks ok but I if there's a fortran-specific issue here I wouldn't know it. :-)

> diff --git a/gdb/testsuite/gdb.fortran/type.f90 b/gdb/testsuite/gdb.fortran/type.f90
> new file mode 100644
> index 0000000..e0c699e
> --- /dev/null
> +++ b/gdb/testsuite/gdb.fortran/type.f90
> @@ -0,0 +1,28 @@
> +! Copyright 2013 Free Software Foundation, Inc.
> +!
> +! This program is free software; you can redistribute it and/or modify
> +! it under the terms of the GNU General Public License as published by
> +! the Free Software Foundation; either version 3 of the License, or
> +! (at your option) any later version.
> +!
> +! This program is distributed in the hope that it will be useful,
> +! but WITHOUT ANY WARRANTY; without even the implied warranty of
> +! MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +! GNU General Public License for more details.
> +!
> +! You should have received a copy of the GNU General Public License
> +! along with this program.  If not, see <http://www.gnu.org/licenses/>.
> +
> +program type
> +  implicit none
> +
> +  type :: t1
> +    integer :: t1_i
> +    real    :: t1_r
> +  end type t1
> +
> +  type (t1) :: t1v
> +
> +  t1v%t1_i = 42
> +  t1v%t1_r = 42.24    ! bp1
> +end program type
> diff --git a/gdb/testsuite/gdb.fortran/whatis_type.exp b/gdb/testsuite/gdb.fortran/whatis_type.exp
> new file mode 100644
> index 0000000..5a90b40
> --- /dev/null
> +++ b/gdb/testsuite/gdb.fortran/whatis_type.exp
> @@ -0,0 +1,48 @@
> +# Copyright 2013 Free Software Foundation, Inc.
> +
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 3 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program.  If not, see <http://www.gnu.org/licenses/>.
> +
> +if { [skip_fortran_tests] } { continue }
> +
> +standard_testfile type.f90
> +
> +if { [prepare_for_testing ${testfile}.exp ${testfile} \
> +                          ${srcfile} {debug f90}] } {
> +    return -1
> +}
> +
> +if ![runto MAIN__] {
> +    perror "Couldn't run to MAIN__"
> +    continue
> +}

I never liked calling perror for this particular failure.
grepping, I see it's still used a lot though I'll bet that's all
cut-n-paste (and presumably you're just going with the flow here as
well).
For new patches I'd like to just use fail here (that's what runto will
do internally if asked).

> +
> +gdb_breakpoint [gdb_get_line_number "bp1"]
> +gdb_continue_to_breakpoint "bp1"
> +
> +set t1_i "integer\\\(kind=4\\\) :: t1_i"
> +set t1_r "real\\\(kind=4\\\) :: t1_r"
> +
> +gdb_test "whatis t1" \
> +  "type = Type t1\r\n${t1_i}\r\n${t1_r}\r\nEnd Type t1" \
> +  "whatis t1"
> +gdb_test "whatis t1v" \
> +  "type = Type t1\r\n${t1_i}\r\n${t1_r}\r\nEnd Type t1" \
> +  "whatis t1v"
> +
> +gdb_test "ptype t1" \
> +  "type = Type t1\r\n    ${t1_i}\r\n    ${t1_r}\r\nEnd Type t1" \
> +  "ptype t1"
> +gdb_test "ptype t1v" \
> +  "type = Type t1\r\n    ${t1_i}\r\n    ${t1_r}\r\nEnd Type t1" \
> +  "ptype t1"

Typo: "ptype t1v"