[ome-users] [OME] #12420: Bug: MIPAV importer plugin broken due to model changes

Sat Jun 28 18:47:48 BST 2014

Hi Pascal,

> Would it not be better to mention this in the current documentation that
the
> present Bioformats Plugin is not any more supported for MIPAV?

I agree that updating the documentation is a good idea. However:

1) I do not control the documentation, so suggesting it to me in a private
mail will not accomplish it. I have re-CCed the ome-users list.

2) Of course, I think it would be "better" to fix the plugin so that it
actually works in MIPAV again.

Regards,
Curtis

On Sat, Jun 28, 2014 at 10:03 AM, Dr. Pascal Detampel <pdetampe at ucalgary.ca>
wrote:

> Hi Curtis,
>
> Thanks for your reply.
> Would it not be better to mention this in the current documentation that
> the present Bioformats Plugin is not any more supported for MIPAV?
> Right now, one gets the impression that the installation is just straight
> forward.
>
> I will be away till Wednesday.
>
> Best,
> Pascal
>
>
>
> Am 28.06.2014 um 06:51 schrieb Curtis Rueden <ctrueden at wisc.edu>:
>
> > Hi Pascal,
> >
> > > But with respect to Zeiss czi stacks, the importer throws an error
> > > message that “An I/O error occurred reading the file: Import data
> > > error: bounds incorrect”. So I guess, as already pointed out by Roger,
> > > the code has some additional problems.
> >
> > The plan we discussed a couple of years ago with the MIPAV team was for
> them to adopt the plugin upstream as part of MIPAV. We did some work [1] to
> make it possible.
> >
> > But I guess it never panned out. If you are interested in seeing that
> happen you could write a mail to the MIPAV mailing list [2].
> >
> > Regards,
> > Curtis
> >
> > [1] https://github.com/openmicroscopy/bioformats/pull/64
> > [2] http://mipav.cit.nih.gov/pubwiki/index.php/MIPAV_mailing_list
> >
> >
> > On Fri, Jun 27, 2014 at 10:14 PM, Dr. Pascal Detampel <
> pdetampe at ucalgary.ca> wrote:
> > Just in case somebody runs into the same issue.
> > With the help of Roger, and some digging in the corresponding classes, I
> got it to work again.
> >
> > Next to replacing getDimensions to getPixels, dropping the index, and
> changing the object type to PositiveFloat (thanks Roger), I had to call the
> method *.getValue() instead of *.floatValue() and change the type from
> float to double in the lines below. Only the object "dimTimeInc” should be
> Double and the elements of the array “res” (a few lines below) have to be
> casted (float). In addition, the object type PositiveFloat has to be
> imported at the top of the file (import
> ome.xml.model.primitives.PositiveFloat;).
> >
> > Except for the import statement at the top, my changes looked like the
> following:
> >
> > >          // harvest physical resolution
> > >          PositiveFloat dimPhysSizeX = store.getPixelsPhysicalSizeX(0);
> > >          PositiveFloat dimPhysSizeY = store.getPixelsPhysicalSizeY(0);
> > >          PositiveFloat dimPhysSizeZ = store.getPixelsPhysicalSizeZ(0);
> > >          Double dimTimeInc = store.getPixelsTimeIncrement(0);
> > >          double physSizeX = dimPhysSizeX == null ?
> > >            1.0f : dimPhysSizeX.getValue();
> > >          double physSizeY = dimPhysSizeY == null ?
> > >            1.0f : dimPhysSizeY.getValue();
> > >          double physSizeZ = dimPhysSizeZ == null ?
> > >            1.0f : dimPhysSizeZ.getValue();
> > >          float timeInc = dimTimeInc == null ? 1.0f :
> dimTimeInc.floatValue();
> > >
> > >          // compute dimensional extents
> > >          int[] dimExtents = {sizeX, sizeY, sizeZ, sizeT};
> > >          float[] res = {(float) physSizeX, (float) physSizeY, (float)
> physSizeZ, timeInc};
> > >          int[] units = {
> > >            FileInfoBase.MICROMETERS, FileInfoBase.MICROMETERS,
> > >            FileInfoBase.MICROMETERS, FileInfoBase.SECONDS
> > >          };
> >
> > Maybe there is an more elegant way, but at least it compiles again and
> Mipav shows the BioFormatsImporter under Plugins.
> > But with respect to Zeiss czi stacks, the importer throws an error
> message that “An I/O error occurred reading the file: Import data error:
> bounds incorrect”.
> > So I guess, as already pointed out by Roger, the code has some
> additional problems.
> >
> > Thanks,
> > Pascal
> >
> >
> > On Jun 27, 2014, at 7:42 AM, OME <trac at cvs.openmicroscopy.org.uk> wrote:
> >
> > > #12420: Bug: MIPAV importer plugin broken due to model changes
> > >
> ------------------------------+---------------------------------------------
> > >      Reporter:  rleigh       |        Owner:  mlinkert
> > >          Type:  task         |       Status:  new
> > >      Priority:  minor        |    Milestone:  Unscheduled
> > >     Component:  Bio-Formats  |      Version:  5.0.2
> > >    Resolution:               |     Keywords:
> > > Remaining Time:               |       Sprint:
> > >     Resources:               |   References:
> > > Referenced By:               |
> > >
> ------------------------------+---------------------------------------------
> > > PlugInBioFormatsImporter.java will not compile:
> > >
> > > {{{
> > > labtop:plugins pascal$ javac -cp
> > > /Applications/mipav:/Applications/mipav/bioformats_package.jar
> > > PlugInBioFormatsImporter.java
> > >
> > > PlugInBioFormatsImporter.java:201: cannot find symbol
> > > symbol  : method getDimensionsPhysicalSizeX(int,int)
> > > location: interface loci.formats.meta.IMetadata
> > >           Float dimPhysSizeX = store.getDimensionsPhysicalSizeX(0, 0);
> > >
> > > PlugInBioFormatsImporter.java:202: cannot find symbol
> > > symbol  : method getDimensionsPhysicalSizeY(int,int)
> > > location: interface loci.formats.meta.IMetadata
> > >           Float dimPhysSizeY = store.getDimensionsPhysicalSizeY(0, 0);
> > >
> > > PlugInBioFormatsImporter.java:203: cannot find symbol
> > > symbol  : method getDimensionsPhysicalSizeZ(int,int)
> > > location: interface loci.formats.meta.IMetadata
> > >           Float dimPhysSizeZ = store.getDimensionsPhysicalSizeZ(0, 0);
> > >
> > > PlugInBioFormatsImporter.java:204: cannot find symbol
> > > symbol  : method getDimensionsTimeIncrement(int,int)
> > > location: interface loci.formats.meta.IMetadata
> > >           Float dimTimeInc = store.getDimensionsTimeIncrement(0, 0);
> > >
> > > 4 errors
> > > }}}
> > >
> > > The method calls in question were removed from the OME data model in
> 2010,
> > > and so it has been non-functional for some time.
> > >
> > > Fixing the above error is simple (change getDimensions to getPixels and
> > > drop the pixel index).  However, there may be additional problems with
> the
> > > code.
> > >
> > > Ideally, there should be a maven pom to allow this to be added to our
> > > regular CI infrastructure to pick up any future breakage.
> > >
> > > --
> > > Ticket URL: <http://trac.openmicroscopy.org.uk/ome/ticket/12420>
> > > OME <http://www.openmicroscopy.org/>
> > > OME Project
> >
> >
> > _______________________________________________
> > ome-users mailing list
> > ome-users at lists.openmicroscopy.org.uk
> > http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users
> >
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20140628/b9fd3308/attachment.html>