From maintainers-request at octave dot org Wed Dec  7 19:32:15 2005
Subject: Re: message ids for warnings and errors
From: Paul Kienzle <pkienzle at users dot sourceforge dot net>
To: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
Cc: octave maintainers mailing list <maintainers at octave dot org>
Date: Wed, 7 Dec 2005 20:30:57 -0500


On Dec 7, 2005, at 7:24 PM, John W. Eaton wrote:

> On  7-Dec-2005, Paul Kienzle wrote:
>
> | It would be nice if the warning state were automatically
> | restored when the function completes, otherwise we are going
> | to end up with unwind protect cruft sprinkled everywhere again.
>
> One of the reasons for adding the message-id thing was compatibility.
> Having the state scoped would break compatibility.
>
> In CVS Octave, we have 21 knobs for warning control:
>
>   warn_assign_as_truth_value
>   warn_function_name_clash
>   warn_neg_dim_as_zero
>   warn_separator_insert
>   warn_variable_switch_label
>   warn_associativity_change
>   warn_future_time_stamp
>   warn_num_to_str
>   warn_single_quote_string
>   warn_divide_by_zero
>   warn_imag_to_real
>   warn_precedence_change
>   warn_str_to_num
>   warn_empty_list_elements
>   warn_matlab_incompatible
>   warn_reload_forces_clear
>   warn_string_concat
>   warn_fortran_indexing
>   warn_missing_semicolon
>   warn_resize_on_range_error
>   warn_undefined_return_values
>
> and 12 of them are off by default.  So I guess we have a situation
> similar to Matlab, where we will start with
>
>   warning on all
>
> plus the 12 exceptions.
>
> Is it really a problem if the state is global?  Are there very many
> (any) places now where we need unwind_protect to play with the warning
> state?

~/cvs/octave/scripts$ grep -h "^ *warn_" */*.m */*/*.m | sed -e"s/^ 
*//;s/ *= */ = /" | sort | uniq

   warn_empty_list_elements = 0;
   warn_empty_list_elements = save_warn_empty_list_elements;
   warn_fortran_indexing = 0;
   warn_fortran_indexing = wfi;
   warn_str_to_num = 0;
   warn_str_to_num = tmp;

~/cvs/octave/scripts$ grep -l "^ *warn_" */*.m */*/*.m | wc
   16      16     357


Similarly, octave-forge has 27 instances, including

   warn_divide_by_zero = 0;
   warn_divide_by_zero = wdz;

Maybe warn could have another state 'expected' which
suppresses the warning in the local scope?  So rather
than:

    wfi = warning('query','octave:empty_list_elements');
    unwind_protect
      warning('off','octave:empty_list_elements');
      ...
    unwind_protect_cleanup
      warning(wfi,'octave:empty_list_elements');
    end_unwind_protect

you would use:

    warning('expected','octave:empty_list_elements');
    ...

- Paul