On 05/17/11 09:33 AM, Nico Williams wrote:
> On Tue, May 17, 2011 at 11:27 AM, Alan Coopersmith
> <alan.coopersm...@oracle.com> wrote:
>> On 05/17/11 09:05 AM, Nico Williams wrote:
>>> I wonder if xsltproc could provide a pseudo-random version of
>>> generate_id() (and a way to seed it).
>>
>> You could probably LD_PRELOAD a shared object with an init routine to
>> malloc a psuedo-random amount of memory on the heap before the file is
>> loaded & parsed, if you wanted to test with different memory addresses.
>>
>> Since the sparc & x86 memory layouts are definitely different, diff'ing
>> those is a simpler way to test that, if you can build on both architectures.
>> If you only have one, you might try 32-bit vs. 64-bit builds of xsltproc.
>
> What I meant is that it'd be nice if xsltproc provided a deterministic
> ID generation facility for this instead of using a
> not-so-deterministic one. File an RFE with the xsltproc community?
Ah, I liked the idea from the blog I linked to of having it be the offset
from the start of the doc, which may be as simple as having a base address
subtracted from the tag's address in the id generation, but I don't know
enough xslt to claim to understand the methods it used to do it.
I don't see any relevant reports for it in the libxslt product under
bugzilla.gnome.org yet.
--
-Alan Coopersmith- alan.coopersm...@oracle.com
Oracle Solaris Platform Engineering: X Window System
_______________________________________________
ksh93-integration-discuss mailing list
ksh93-integration-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/ksh93-integration-discuss
