[ome-devel] 'Modulo' tag Start/End types

[Berg, Stuart]

Hi Ian,

Thanks for the explanation.  Now I've submitted a patch to the tifffile.py
author to handle this case.

Best regards,
Stuart


On 7/29/16, 7:45 AM, "ome-devel on behalf of Munro, Ian"
<ome-devel-bounces at lists.openmicroscopy.org.uk on behalf of
i.munro at imperial.ac.uk> wrote:

>Hi Stuart
>
>These fields allow a scale to be added to the extra Dimension.
>There is therefore no fundamental reason why they should be integers.
>My feeling is that this would be very restrictive.
>
>To give an example from FLIM: - the modality that I am most familiar with.
>The data stored in the extra dimension has units of time - though not
>real-time.
>Typically data is recorded over 12.5ns so “End” would be 12.5 if  ns
>rather than ps units were to be used.
>
>
>Regards
>
>
>Ian.
>
>
>
>
>Thiere’s no fundamental reason why these fields
>> On 29 Jul 2016, at 00:25, Berg, Stuart <bergs at janelia.hhmi.org> wrote:
>> 
>> Hi,
>> 
>> I use Christoph Gohlke's 'tifffile.py' library for interpreting OME-TIFF
>> metadata.  One of my users recently reported that tifffile.py chokes on
>>a
>> file created by the 'bfconvert' tool[1].  The problem turns out to be
>>this
>> tag:
>> 
>> <ModuloAlongC End="1.0" Start="0.0" Step="1.0" Type="other"
>> TypeDescription="TCSPC" />
>> 
>> The tifffile.py code expects Start/End/Step to be straightforwardly
>> convertible to int, not float.  If I understand correctly, these fields
>> will always be integers, so they could have been written as Start="1",
>> Step="1", etc.  Therefore, it seems wrong for bfconvert to write "1.0",
>> right?
>> 
>> But if you think this is expected output from bfconvert, then I'm happy
>>to
>> submit a tiny patch to the tifffile.py repo.
>> 
>> Best regards,
>> Stuart
>> 
>> [1]: https://github.com/ilastik/ilastik/issues/1289
>> 
>> _______________________________________________
>> ome-devel mailing list
>> ome-devel at lists.openmicroscopy.org.uk
>> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
>
>_______________________________________________
>ome-devel mailing list
>ome-devel at lists.openmicroscopy.org.uk
>http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel