Re: Bug in Saxon?

The software is working as designed, and it's in conformance with the spec.

Warnings tend to attract strong opinions, some people value them and others
don't. Some people have a self-imposed rule that the code must execute without
warnings, and by doing that they are making life a bit more difficult for
themselves. There are plenty of facilities for suppressing warnings, either
generally or very specifically.

If you have a line of code that will always fail if executed, such as
xs:integer("XYZ"), I think that's grounds for a warning, even if the logic
ensures that it isn't executed. In fact having code that can never be executed
is grounds for a warning as well.

My IDE (IntelliJ) gives me lots of similar warnings for my Java code
(unreachable code, conditions that will always be false, casts that can never
succeed, expressions that might give an NPE, etc.) On the whole I find that
these are helpful: they enable me to improve the quality of my code.

Of course there are cases where you want to ignore the warning, for example
code that is deliberately unreachable because you've put it in an "if (false)"
block. It's an area where you can't please everyone all of the time.

Michael Kay
Saxonica

> On 7 Aug 2024, at 14:09, Martin Honnen martin.honnen@xxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
>
>
> On 07/08/2024 15:00, Roger L Costello costello@xxxxxxxxx
<mailto:costello@xxxxxxxxx> wrote:
>>
>> Okay, I upgraded to saxon-ee-12.5.jar and I get the same warning message.
When I run my actual program, I get dozens of these warning messages until
finally I get a message bNo more warnings will be displayedb Turning off
warning messages is not acceptable.
>
> The bogus warning seems a bug in 12(.5) EE, I don't see it with HE. That
suggests you might be able to get rid of it by suppressing some optimizations
EE does but I haven't tried to figure out which one(s) that might be.
>
>
>
>
>
> XSL-List info and archive <http://www.mulberrytech.com/xsl/xsl-list>
> EasyUnsubscribe <http://lists.mulberrytech.com/unsub/xsl-list/3500899> (by
email <>)