fiji-hpc / hpc-datastore Goto Github PK

in "Share BDS" option in Fiji, the newly added channel is missing an icon

.spim.data.SpimDataIOException: java.io.IOException: Server returned HTTP response code: 500 for URL: http://localhost:9080/bdv/a534ba9f-578d-49e9-8eb5-6f1e28ffe66e/1/

when accessing newly added channel

as a result, the new channel cannot be opened in BDV

....simply when, e.g., a dataset consists of four channels each being displayed translated/rotated/scaled w.r.t. to some common reference frame

One can download a test dataset, unzip, a point BDV to the 807191f5-.../0/export.xml file to understand what the dataset looks like. It should look like this:

when zoomed-out a little, and displayed using the Fuse mode on.

➡️
The issue comes when accessing this dataset using any option from the menu Fiji -> Plugins -> HPC DataStore -> BigDataViewer, which means to access the data remotely via the DS server itself, in this regime the opening fails. The error is the following.

Worth noticing is that the JSON listing from alfeios, where this dataset is also installed, shows nulls for transformations and viewRegistration fields.

It will allow to add directory of dataset that was copied from another server.

to get rid errors like this one:
java.io.FileNotFoundException: /tmp/datastore/0b1db55d-8bcb-430e-8f66-dc56c9b8ca94

[ERROR] Command errored: Create new dataset from XML/HDF5
java.lang.NoSuchMethodError: net.imglib2.img.basictypeaccess.ArrayDataAccessFactory.get(Lnet/imglib2/type/NativeType;)Ljava/lang/Object;
at bdv.export.ExportScalePyramid$BlockCreator.forType(ExportScalePyramid.java:431)
at bdv.export.ExportScalePyramid.writeScalePyramid(ExportScalePyramid.java:237)
at cz.it4i.fiji.datastore.rest_client.WriteSequenceToN5.writeScalePyramid(WriteSequenceToN5.java:249)
at cz.it4i.fiji.datastore.rest_client.WriteSequenceToN5.writeN5File(WriteSequenceToN5.java:208)
at cz.it4i.fiji.datastore.ij.ExportSPIMAsN5PlugIn.run(ExportSPIMAsN5PlugIn.java:220)
at org.scijava.command.CommandModule.run(CommandModule.java:196)
at org.scijava.module.ModuleRunner.run(ModuleRunner.java:165)
at org.scijava.module.ModuleRunner.call(ModuleRunner.java:125)
at org.scijava.module.ModuleRunner.call(ModuleRunner.java:64)
at org.scijava.thread.DefaultThreadService.lambda$wrap$2(DefaultThreadService.java:247)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

...or figure out how non-gray16 is solved when opening N5 data

(related to #24)

Při typech uint8, float32 (uint16 funguje, jiné jsem nezkoušel) vrací server dimenze částečného bloku (prvních 12B) nekorektně. resp. vrátí (-1,-1,-1), nicméně zbytek dat má validní velikost, která odpovídá očekávání.

Postup reprodukce:
zvolit typ datasetu float32 nebo uint8
vytvořit dataset o velikosti, která není dělitelná velikostí bloku
poslat request na získání částečného bloku (tedy toho na okraji obrázku).

Chování:
Server vrátí prvních 12B reprezentujících (-1,-1,-1) a potom data odpovídající korektnímu bloku

Očekávané chování:
Server vrátí prvních 12B odpovídajích dimenzi získaného block (x,y,z) a data o odpovídajíc velikosti...
celý obsah odpovědi by tedy měl být velký: 12 (dimenze) + x * y * z * sizeof(voxelType)