Recent checkin breaks 32-bit builds

[David Bertoni]

Hi all,

The recent checkin to use XMLFileLoc consistently to represent
line/column numbers has broken most of the 32-bit platforms.

I've almost finished a patch, just in case anyone else has decided to
work on it.

Dave

---------------------------------------------------------------------
To unsubscribe, e-mail: c-dev-unsubscribe@...
For additional commands, e-mail: c-dev-help@...

[Boris Kolpackov-2]

Hi David,

David Bertoni <dbertoni@...> writes:

> The recent checkin to use XMLFileLoc consistently to represent
> line/column numbers has broken most of the 32-bit platforms.

That's my fault, sorry about that.

> I've almost finished a patch, just in case anyone else has decided to
> work on it.

Thanks for working on it. Unfortunately, I don't think we can use
long long type directly since some of the platforms that we try to
support (notably VC6) lack it. I will try to come up with a more
portable fix.

Thanks,
Boris

--
Boris Kolpackov, Code Synthesis Tools   http://codesynthesis.com/~boris/blog
Open source XML data binding for C++:   http://codesynthesis.com/products/xsd
Mobile/embedded validating XML parsing: http://codesynthesis.com/products/xsde

---------------------------------------------------------------------
To unsubscribe, e-mail: c-dev-unsubscribe@...
For additional commands, e-mail: c-dev-help@...

[David Bertoni]

Boris Kolpackov wrote:

> Hi David,
>
> David Bertoni <dbertoni@...> writes:
>
>> The recent checkin to use XMLFileLoc consistently to represent
>> line/column numbers has broken most of the 32-bit platforms.
>
> That's my fault, sorry about that.
>
>> I've almost finished a patch, just in case anyone else has decided to
>> work on it.
>
> Thanks for working on it. Unfortunately, I don't think we can use
> long long type directly since some of the platforms that we try to
> support (notably VC6) lack it. I will try to come up with a more
> portable fix.

It's about time we got rid of support for VC6, since it's long out of
service.

We already have a similar workaround in Xalan-C, which I first pursued
for this issue, but it involves some ifdefs.  I can revive that if you
want me to.

Dave

---------------------------------------------------------------------
To unsubscribe, e-mail: c-dev-unsubscribe@...
For additional commands, e-mail: c-dev-help@...

[Boris Kolpackov-2]

Hi David,

David Bertoni <dbertoni@...> writes:

> It's about time we got rid of support for VC6, since it's long out of
> service.

I am personally in favor of dropping VC6 (and even more so BCB) but
other developers might not share this view. I think we will have to
vote for it so feels free to initiate it (you've got my +1; I would
first ask the community if anyone is planning to use 3.0.0 with VC6
or BCB.


> We already have a similar workaround in Xalan-C, which I first pursued
> for this issue, but it involves some ifdefs.  I can revive that if you
> want me to.

I've already committed a fix that should take care of this.

BTW, I haven't had a chance to review your memory manager patch yet
but I am planning to do so in the next couple of days.

Thanks,
Boris

---------------------------------------------------------------------
To unsubscribe, e-mail: c-dev-unsubscribe@...
For additional commands, e-mail: c-dev-help@...