Re: Assertion failure

[mike]

On 28 août 2012, at 15:17, David Kastrup <address@hidden> wrote:

> "address@hidden" <address@hidden> writes:
> 
>> Hey all,
>> 
>> With --disable-optimisation passed to configure, Janek's score at:
>> 
>> www.mikesolomon.org/tota-pulchra.zip
>> 
>> fails with the assertion failure on line 252 of the current grob-property.cc
>> 
>> assert (value == SCM_EOL || value == marker);
>> 
>> I started a git bisect but have no clue when the first good commit is
>> - it hits this failure at least as far back as June 20th.
>> 
>> I'm a bit short on time but could someone do some snooping to try and
>> figure out what the problematic commit is?
> 
> Not sure it's not a case of historical nonsense.  Cf.
> 
> <URL:http://codereview.appspot.com/6449126/diff/1003/lily/grob-property.cc#newcode254>
> 
> 
> -- 
> David Kastrup
> 
> 
> _______________________________________________
> lilypond-devel mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/lilypond-devel

Neil's right (hey Neil!) that the property shouldn't be set in the user-defined 
function the way that it is currently set. I should have checked the file first.

Slightly changing subject, I'm of the opinion that no user input, however 
unconventional, should lead to an assertion failure.  When someone puts an 
assertion failure in the code, to me, it is saying "if this thing is ever 
triggered, every developer should stop what she is doing and figure out what is 
wrong with LilyPond and pull all nighters until it's fixed."  Otherwise a 
programming error seems more appropriate.

Cheers,
MS