[File] Big endianess problem on s390x?

[Christos Zoulas]

> On May 19, 2026, at 3:54 PM, Guy Harris <gharris at sonic.net> wrote:
> 
> On May 18, 2026, at 8:15 AM, Christos Zoulas <christos at zoulas.com> wrote:
> 
> 
>> Is this after the most recent change that includes byte swapping?
>> 
>> 
> 
> 
> Looking at the tip of the main branch in https://github.com/file/file.git, it appears that there are data item types "beguid", "leguid", and "guid", with "guid" being equivalent to "leguid". The only difference between them appears to be how they're printed, i.e. how they're converted from binary format to text format; that would suggest that it should affect how they're *parsed* in magic files, i.e. how they're converted from text format to binary format, as well.
> 
I think that's exactly right and then way the code is done now is probably wrong and should be fixed.
The question is do we default "guid" to the endianness of the system (because that would require fixing all the
magic that is currently guid to leguid) or do we alias it with leguid?

christos

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.astron.com/pipermail/file/attachments/20260519/91cb13e8/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 235 bytes
Desc: Message signed with OpenPGP
URL: <http://mailman.astron.com/pipermail/file/attachments/20260519/91cb13e8/attachment.asc>