[ome-users] Bioformats works only outside OME (more data)
Mario Valle
mvalle at cscs.ch
Thu Jan 31 13:13:05 GMT 2008
Ciao Curtis!
The new loci_tools.jar seems better, but there are apparent good Leica files (.lif) that
are not imported.
As per your suggestion, I have changed the order of readers in the database to put
Bioformats in the first position.
You can find the two files under ftp.cscs.ch/out/mvalle/
2007-11-12_L15_AF001.lif (21MB)
this file has been correctly imported as a set of distinct images
2007-11-13_DMEM_GFPcells.lif (11MB)
this instead has been rejected as 'unknown format'
Could you check those images? BTW, the both open correctly in bfview, even if the show
only a single contained image instead of 14. Any idea?
Also I had to set permissions on the images to rwxrwxrwx; with r--r--r-- no import takes
places in OME 2.6.1 on Mac.
Thanks for your help!
mario
Curtis Rueden wrote:
> Hi again,
>
> It looks like I spoke too soon about the new loci_tools.jar not being
> able to stitch files. Upon further investigation, it seems to be
> working fine. I think my problem before was that OME's Perl Bio-Rad
> PIC reader was overriding Bio-Formats on my system at the time, so it
> read my test dataset as 12 separate Images, rather than a single
> Image.
>
> Once again, you can find the updated loci_tools.jar here:
> http://skyking.microscopy.wisc.edu/curtis/jar/loci_tools.jar
>
> Please let me know if it doesn't work for you.
>
> -Curtis
>
> On Jan 15, 2008 1:56 PM, Curtis Rueden <ctrueden at wisc.edu> wrote:
>> Hi Mario, Linda, and everyone,
>>
>> I isolated the most serious problem to a change we made on December
>> 11th (r3465). I have fixed several problems including that one, and
>> placed a working loci_tools.jar you can try here:
>>
>> http://skyking.microscopy.wisc.edu/curtis/jar/loci_tools.jar
>>
>> Just overwrite the one in /OME/java with that one, and hopefully
>> things will work better.
>>
>> We will try to make another official release of the library soon, to
>> minimize the number of new OME installations that are affected by this
>> problem.
>>
>> Please note what Ilya and I said earlier about rearranging the order
>> of importers, though -- if you don't, Bio-Formats will not be used for
>> certain formats including Deltavision, BMP, Molecular Dynamics GEL,
>> Metamorph STK, DICOM, TIFF & OME-TIFF, Bio-Rad PIC, and Zeiss LSM.
>>
>> Also please note that there are undoubtably still some (hopefully more
>> minor) bugs -- in particular, the file stitching is not working
>> properly. This means you cannot yet import a single Image distributed
>> across multiple numbered files (e.g., data1.tiff, data2.tiff,
>> data3.tiff, etc.). This functionality used to work, so is probably
>> just a matter of me fixing more bugs in the Bio-Formats file stitching
>> code somewhere.
>>
>> Linda, to revisit your earlier question about metadata: this updated
>> version of Bio-Formats should properly include all the original
>> key/value metadata pairs in the OME import. To see these values, after
>> importing the Image, click the Image thumbnail to bring up the Image
>> details screen. Among other things, you'll see a "Module executions"
>> box on the top right. The first link in that box should say "Image
>> import" -- click it, and you'll receive a detailed list of
>> information, including the key/value pairs in their original format
>> beneath a section labeled "OriginalMetadata." Please let me know if it
>> doesn't work for you.
>>
>> Lastly, in the process of debugging, I regenerated several old
>> versions of loci_tools.jar. I placed them online, in case anyone else
>> wants them to do their own testing:
>>
>> http://skyking.microscopy.wisc.edu/curtis/archive/
>>
>> Please let me know if you continue to have trouble, or have any
>> additional questions.
>>
>> -Curtis
>>
>>
>> On Jan 11, 2008 12:22 PM, Curtis Rueden <ctrueden at wisc.edu> wrote:
>>> Hi Mario & Ilya,
>>>
>>> As I said yesterday, there is a critical problem with the latest
>>> Bio-Formats and OME server. I was afraid of this, as I have honestly
>>> been too neglectful of testing with OME server recently. I am looking
>>> into the problem today, and hope to have a solution as soon as
>>> possible. In the meantime, Ilya, if you have a working version of
>>> loci_tools.jar that you could post online somewhere temporarily for
>>> people to use, that would be really helpful. Otherwise I can try
>>> building a few of the old versions until I find one that works well.
>>>
>>> My sincerest apologies to Mario, Linda and any others who have been
>>> struggling with the Bio-Formats/OME connectivity.
>>>
>>> -Curtis
>>>
>>>
>>> On Jan 11, 2008 9:08 AM, Ilya Goldberg <igg at nih.gov> wrote:
>>>> It seems more and more people are struggling with OME server's over-
>>>> riding the new and better bioformats with its old and deprecated
>>>> importers.
>>>> See Curits's earlier post about re-arranging the importer priority to
>>>> either eliminate OME's importers altogether, or move bioformats up in
>>>> the priority list.
>>>> We will make these changes permanent in future releases so that BF
>>>> takes priority over everything else.
>>>>
>>>> There may be other problems with the connection, but in the default
>>>> configuration, OME's importers will over-ride TIFF, LSM, and STK.
>>>> They should still import them, but apparently there are potentially
>>>> other issues with these files, and BF does a better job. LIFF and
>>>> LIF should have been delegated to BF, because there are no "native"
>>>> importers for these. Which makes me think there may be an unrelated
>>>> problem with the BF setup.
>>>>
>>>> -Ilya
>>>>
>>>>
>>>> On Jan 11, 2008, at 6:31 AM, Mario Valle wrote:
>>>>
>>>>> Continuing investigation I have found:
>>>>>
>>>>> - From the OME web interface I can import only TIFF files. No
>>>>> error, nothing, only
>>>>> "unrecognized format"
>>>>>
>>>>> - From the command line: ome import -d "dataset" image
>>>>> I can import TIFF, STK, ie. everything except formats supported
>>>>> by the bioformats
>>>>> library.
>>>>>
>>>>> So seems there are two problems:
>>>>> 1) the bioformats library not found
>>>>> 2) the web server accessing only a minimal subset of the import
>>>>> functionality.
>>>>>
>>>>> Help!
>>>>> mario
>>>>>
>>>>>
>>>>> Mario Valle wrote:
>>>>>> I'm still stuck at the image importing step...
>>>>>>
>>>>>> 1) Moved to OME 2.6.1 on Mac (here I'm administrator, so I can
>>>>>> experiment)
>>>>>> 2) installed latest loci_tools.jar under /OME/java
>>>>>> 3) downloaded bftools, unpacked and added the same loci_tools.jar
>>>>>> ar OME in the resulting
>>>>>> directory
>>>>>> 3) tried bfview with TIFF, LIF, LSM, STK, LIFF files. Everything
>>>>>> works. Only TIFF with
>>>>>> character encoding Mac cannot be visualized, but this is another
>>>>>> problem.
>>>>>> 3) tried to import the same files in OME, only the tiff files are
>>>>>> imported.
>>>>>> Seems like OME is using some old import method, but no other copy
>>>>>> of loci_tools.jar is
>>>>>> anywhere on the machine.
>>>>>> Help!
>>>>>> mario
>>>>>>
>>>>>>
>>>>> --
>>>>> Ing. Mario Valle
>>>>> Visualization Group | http://
>>>>> www.cscs.ch/~mvalle
>>>>> Swiss National Supercomputing Centre (CSCS) | Tel: +41 (91)
>>>>> 610.82.60
>>>>> v. Cantonale Galleria 2, 6928 Manno, Switzerland | Fax: +41 (91)
>>>>> 610.82.82
>>>>> _______________________________________________
>>>>> ome-users mailing list
>>>>> ome-users at lists.openmicroscopy.org.uk
>>>>> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users
>>>>>
>>>> _______________________________________________
>>>> ome-users mailing list
>>>> ome-users at lists.openmicroscopy.org.uk
>>>> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users
>>>>
>
--
Ing. Mario Valle
Visualization Group | http://www.cscs.ch/~mvalle
Swiss National Supercomputing Centre (CSCS) | Tel: +41 (91) 610.82.60
v. Cantonale Galleria 2, 6928 Manno, Switzerland | Fax: +41 (91) 610.82.82
More information about the ome-users
mailing list