On 13-Mar-2009, Ben Abbott wrote:
| On Mar 13, 2009, at 3:38 PM, John W. Eaton wrote:
|
| > On 13-Mar-2009, Ben Abbott wrote:
| >
| > | On Mar 13, 2009, at 9:40 AM, Rob Mahurin wrote:
| > |
| > | > On Mar 12, 2009, at 9:08 PM, Ben Abbott wrote:
| > | >> Can someone explain what this warning is intended to imply?
| > | >>
| > | >> warning: suggest parenthesis around assignment used as truth
| > value
| > | >
| > | > This is a gcc warning? The idiom
| > | > if (x=1) { ... }
| > | > is legal and useful but often a typo for
| > | > if (x==1) { ... }
| > | > Extra parentheses
| > | > if ((x=1)) { ... }
| > | > apparently suppress the warning.
| >
| > The intent of the Octave warning is the same as the GCC
warning. In
| > what case is it emitted incorrectly?
| >
| > | Your examples do give me some additional context for what is
imlied.
| > | Unfortunately, I've searched my sources and don't see such an
| > instance
| > | in my if/elseif statments. Further when I convert warnings to
errors
| > | ("warning error"), I get what looks like a parser error.
| >
| > Using "warning error" will likely generate a lot of unwanted
errors
| > from code that is distributed with Octave. We should probably
limit
| > these warnings to code that is not part of Octave itself.
| >
| > jwe
|
| I'm still getting the original warning. Can anyone recommend what I
| might do to determine the respective file and line number which is
| responsible?
I should have a change checked in shortly that will give you a line
number and a filename for this parser warning. I've also checked in
another change that should make
warning on backtrace
work again. But that is no help here, since the warning you are
seeing is emitted when Octave is parsing an expression, not when it is
evaluating it.
jwe