Re: [lmi] Safer is the Yoda style?

[Greg Chicares]

On 2019-06-10 12:07, Vadim Zeitlin wrote:
> On Sun, 9 Jun 2019 21:17:24 +0000 Greg Chicares <address@hidden> wrote:
> 
> GC> When writing an assertion, I experimentally deviated from the "Yoda"
> GC> style, to see whether gcc would flag the error. It didn't.

[...because it was an assertion macro implemented as 'if(!(condition))'...]

>  And, for completeness, I also checked if wxASSERT() also suffers from the
> same problem and, unfortunately, it does. This is even stranger, however,
> as it doesn't add extra parentheses and uses "!(expr)" in its expansion
> instead -- and it turns out that this is enough to disable -Wparentheses
> from both gcc and clag too (while MSVC still gives it in this case too).
> So at the very least, I'm going to apply this change:
> 
> https://github.com/wxWidgets/wxWidgets/pull/1345/commits/04435144ac9f27c7958398177741b796987f0daa

I just committed a similar change in lmi.

>  The conclusion I'd like to draw from this is that we shouldn't be using
> assert() and modify LMI_ASSERT() in the same way I've modified wxASSERT()
> above, i.e. use direct condition instead of its negation.
> 
>  Would you agree with that?

I did modify LMI_ASSERT, and I don't use the C assert macro in real code.

I'll continue using the "Yoda" style, though. In the code I was writing
yesterday, it saved me from an awful mistake. If a native speaker of APL
you were, dislike it you would not.