Upload/Download OME-Tiff

[cxallan]

When I use the CommandLineImporter, it reports:

Depth: 4 Metadata Level: MINIMUM

and then proceeds to only upload minimal metadata. Is there a way to change the metadata level? I'm using 4.2.2.

You're kind of abusing the command line importer interface slightly by doing things in the manner that you are. I've created an example which uses the lower level import library (which is used by the command line importer) that should cover the use cases and issues that you raised including metadata levels and dataset linkage. It's available here:

https://gist.github.com/1710116

It's using the 4.3.3+ API but it should be close enough to give you an idea of what you need to do.

And for some reason I'm able to set X,Y,C and T using the .setPixelSizeX,Y,C, and T methods on a omexmlservice object, but not Z. So when I upload my image, it always has the same number of Z channels as the source image. Basically, I'm taking a source Tiff, extracting its metadata, attaching it to a new tiff, and uploading it. I'm getting nowhere with setting the Z size so I was thinking of just writing a simple XML parser to adjust it *manually*.

Any ideas?

Jake

Physical pixel sizes or size of the given dimension? Can you post the entire example somewhere so we can see it?

[sgaric]

Hi Chris,

I've been slightly absent from the forums lately due to a break so I have missed some of this discussion. Wanted to let you know that uploading images, like you suggested, using ImportLibrary.importImage() function,
worked as intended. I pretty much got everything working as I want it. One thing is missing for me though, so I was wondering if this is possible with OMERO APIs or not.

During an upload of the image is it possible to get and ID (or handle of some sort) for that image within the OMERO system so that I could refer to it if needed down the track in my code. What I would like to do for example is do some import the image to the server, continue on to do some processing on that image and then attach those results to that image in the OMERO server.

I believe this can be done in two steps already, first upload all the images I am interested in. Then retrieve the images from the server (which will give me the image ID within the OMERO system), do processing and attach the results. I would just like to avoid retrieving the image from the system if that is possible.

Also do you have an example of java/python code where some document is being attached to images/datasets? Are the examples on this page what I should be using for that (about 1/3 down the page) http://trac.openmicroscopy.org.uk/ome/wiki/OmeroJava ?

Regards,
Slavisa

[jmoore]

Hi Slavisa,

One thing is missing for me though, so I was wondering if this is possible with OMERO APIs or not.

During an upload of the image is it possible to get and ID (or handle of some sort) for that image within the OMERO system so that I could refer to it if needed down the track in my code. What I would like to do for example is do some import the image to the server, continue on to do some processing on that image and then attach those results to that image in the OMERO server.

The easiest solution is probably to add an import event which will be given the pixel id. For example, you could create an anonymous inner class of the LoggingImportMonitor which handles the event type: IMPORT_DONE:

Code: Select all

    public static class IMPORT_DONE extends PROGRESS_EVENT {
        public final List<Pixels> pixels;
        public IMPORT_DONE(int index, String filename, IObject target,
                Long pixId, int series, ImportSize size, List<Pixels> pixels) {
            super(index, filename, target, pixId, series, size, null, null);
            this.pixels = pixels;
        }
    }


Also do you have an example of java/python code where some document is being attached to images/datasets? Are the examples on this page what I should be using for that (about 1/3 down the page) http://trac.openmicroscopy.org.uk/ome/wiki/OmeroJava ?

Yes. See the section "Create a file annotation and link to an image."

Cheers,
~Josh