Re: Multiview Reconstruction plugin fails because of missing class in imglib2

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Multiview Reconstruction plugin fails because of missing class in imglib2

Nico Stuurman
Hi Stephan,

Moving this thread to the imagej-devel list since Wayne indicated that is the place for imglib2 related issues.  Please see below, and please continue this discussion on the imagej-devel list rather than the image list (sorry for starting it there!):

I am a bit confused why that happens. In ImgLib2, the Pair class was moved from net.imglib2 to net.imglib2.util for the release version. To the best of my knowledge, Fiji currently still comes with imglib2-2.0.0-beta-26.jar - so that’s why it doesn’t find the class. But the multi-view reconstruction plugin (http://fiji.sc/Multiview-Reconstruction <http://fiji.sc/Multiview-Reconstruction>) that comes with the bigdataviewer update site is still build for the imglib2-2.0.0-beta-26.jar. If you build it yourself, please use the branch “compatibility” as “master” works with the release version of imglib2.

This is with the binary distribution of the Multiview Reconstruction plugin, i.e. the version of the plugin that is installed through the Fiji updater.   Is it possible that the code on the BigDatatViewer update site is build from the master branch rather than the “compatibility” branch?

Best,

Nico


Cheers,
Stephan

On Dec 12, 2014, at 20:52 , Nico Stuurman <[hidden email]> wrote:

Using an up-to-date Fiji installation, the Multiview Registration plugin fails because it can not find the class net.imglib2.util.Pair.  Indeed, the same error can be triggered with the following Beanshell script:

import net.imglib2.util.*;
t = new Pair();

It is a bit unclear to me whether the function has just been removed or was just added, but clearly, the plugin code and Fiji itself use different version of the imglib2. Is there any way to get this to work (I would love to try out the plugin with my data)?  

I did try the lifeline version, but not sure if that autoupdates to the latest when I add the BigDataViewer update site, in any case, that version gives me the same problem.  

B.t.w., I am not sure how the developers would like to be contacted, but I did file a bug report (#975). Also happy to try fix this myself, but this seems to be a higher level packaging problem that is not easily penetrable.

Best,

Nico

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html


--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

Nico Stuurman

Vale Lab
HHMI / Dept. of Cellular and Molecular Pharmacology
University of California, San Francisco
Genentech Hall, Room N316
600 16th street
San Francisco, CA 94158-2517

phone: (415) 514-3927
fax: (415) 476-5233







_______________________________________________
ImageJ-devel mailing list
[hidden email]
http://imagej.net/mailman/listinfo/imagej-devel
Reply | Threaded
Open this post in threaded view
|

Re: Multiview Reconstruction plugin fails because of missing class in imglib2

Curtis Rueden
Hi Nico,

> I am not sure how the developers would like to be contacted, but I did
> file a bug report (#975). Also happy to try fix this myself, but this
> seems to be a higher level packaging problem that is not easily
> penetrable.

The bug report is great -- let's use that for any further discussion.


Regards,
Curtis

On Fri, Dec 12, 2014 at 1:52 PM, Nico Stuurman <[hidden email]> wrote:
Using an up-to-date Fiji installation, the Multiview Registration plugin fails because it can not find the class net.imglib2.util.Pair.  Indeed, the same error can be triggered with the following Beanshell script:

import net.imglib2.util.*;
t = new Pair();

It is a bit unclear to me whether the function has just been removed or was just added, but clearly, the plugin code and Fiji itself use different version of the imglib2. Is there any way to get this to work (I would love to try out the plugin with my data)?

I did try the lifeline version, but not sure if that autoupdates to the latest when I add the BigDataViewer update site, in any case, that version gives me the same problem.

B.t.w., I am not sure how the developers would like to be contacted, but I did file a bug report (#975). Also happy to try fix this myself, but this seems to be a higher level packaging problem that is not easily penetrable.

Best,

Nico

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html


_______________________________________________
ImageJ-devel mailing list
[hidden email]
http://imagej.net/mailman/listinfo/imagej-devel
Reply | Threaded
Open this post in threaded view
|

Re: Multiview Reconstruction plugin fails because of missing class in imglib2

Stephan Preibisch-2
I uploaded a new version from the compatibility branch (which builds against imglib-2.0.0-beta26), should work now. The master branch still builds against the first stable release of imglib-2.0.0, as far as this is in Fiji, these problems should be gone.

Thanks everyone!

Cheers, Steffi

On Dec 15, 2014, at 17:42 , Curtis Rueden <[hidden email]> wrote:

Hi Nico,

> I am not sure how the developers would like to be contacted, but I did
> file a bug report (#975). Also happy to try fix this myself, but this
> seems to be a higher level packaging problem that is not easily
> penetrable.

The bug report is great -- let's use that for any further discussion.


Regards,
Curtis

On Fri, Dec 12, 2014 at 1:52 PM, Nico Stuurman <[hidden email]> wrote:
Using an up-to-date Fiji installation, the Multiview Registration plugin fails because it can not find the class net.imglib2.util.Pair.  Indeed, the same error can be triggered with the following Beanshell script:

import net.imglib2.util.*;
t = new Pair();

It is a bit unclear to me whether the function has just been removed or was just added, but clearly, the plugin code and Fiji itself use different version of the imglib2. Is there any way to get this to work (I would love to try out the plugin with my data)?

I did try the lifeline version, but not sure if that autoupdates to the latest when I add the BigDataViewer update site, in any case, that version gives me the same problem.

B.t.w., I am not sure how the developers would like to be contacted, but I did file a bug report (#975). Also happy to try fix this myself, but this seems to be a higher level packaging problem that is not easily penetrable.

Best,

Nico

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

_______________________________________________
ImageJ-devel mailing list
[hidden email]
http://imagej.net/mailman/listinfo/imagej-devel


_______________________________________________
ImageJ-devel mailing list
[hidden email]
http://imagej.net/mailman/listinfo/imagej-devel