[Octave-bug-tracker] [bug #49536] many functions that expect a "string" silently truncate char matrix arguments (octave_value::string_value)

[Mike Miller]

Follow-up Comment #8, bug #49536 (project octave):

The warning comes from the following backtrace


#0  octave_char_matrix_str::string_value[abi:cxx11](bool) const
(this=<optimized out>) at ../libinterp/octave-value/ov-str-mat.cc:238
#1  0x00007fe5b75c66bf in octave_char_matrix_str::short_disp (this=<optimized
out>, os=...) at ../libinterp/octave-value/ov-str-mat.cc:274
#2  0x00007fe5b7b4f25b in octave_value::short_disp (this=0x7fe57cfed9b0,
os=...) at ../libinterp/octave-value/ov.h:1228
#3  symbol_table::do_workspace_info[abi:cxx11]() const (this=0x7fe56019de80)
at ../libinterp/corefcn/symtab.cc:1507
#4  0x00007fe5b7abfbd9 in symbol_table::workspace_info[abi:cxx11](int)
(scope=1) at ../libinterp/corefcn/symtab.h:2095
#5  octave_link::set_workspace () at ../libinterp/corefcn/octave-link.cc:71


I am now using this change


diff --git a/libinterp/octave-value/ov-str-mat.cc
b/libinterp/octave-value/ov-str-mat.cc
--- a/libinterp/octave-value/ov-str-mat.cc
+++ b/libinterp/octave-value/ov-str-mat.cc
@@ -234,6 +234,9 @@ octave_char_matrix_str::string_value (bo
 
   charMatrix chm (matrix);
 
+  if (chm.rows () > 1)
+    warning ("evaluating a character matrix as a string, only the first row
is used");
+
   // FIXME: Is this correct?
   return chm.row_as_string (0);
 }
@@ -268,7 +271,8 @@ octave_char_matrix_str::short_disp (std:
 {
   if (matrix.ndims () == 2 && numel () > 0)
     {
-      std::string tmp = string_value ();
+      charMatrix chm (matrix);
+      std::string tmp = chm.row_as_string (0);
 
       // FIXME: should this be configurable?
       size_t max_len = 100;


and can now work with the GUI and run the full test suite in the GUI with no
warnings.

Needs some polish and a warning ID, but I think this is a good first start to
draw a line.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?49536>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/