Closed sevaa closed 11 months ago
@sevaa - We only provide the ome xml file in cases that it contains useful information. The ome XML does not currently contain useful information in this workflow it's therefore not provided in this workflow.
So the current response is ..."fileMetadata": {}...
Please let us know if you'd like us to not provide fileMetadata in some other manner.
That's what my suspicion was. Can you please pass null
instead, or don't add "fileMetadata" to the enclosing JSON object at all?
EDIT: null
for fileMetadata
, not null
for omeXml
.
Hi @sevaa, can you verify these change in qa
branch?
Not sure how.
@sevaa - the fix is applied to the QA server, you should be able to retest if you like.
Which pipeline was it? I'm looking at the results of the ROI pipeline run. Hard to debug in QA, but from what I see the metadata is still an empty object. Pipeline run for ID S9468. Same in S9469.
large 2d conversion
see: {"files": [{"primaryFilePath": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S1.png", "thumbnailIndex": 0, "title": "IA_M2_S1", "fileMetadata": {"omeXml": null}, "imageSet": [{"imageName": "IA_M2_S1", "imageMetadata": null, "assets": [{"type": "thumbnail", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S1/IA_M2_S1_sm.jpeg"}, {"type": "keyImage", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S1/IA_M2_S1_lg.jpeg"}, {"type": "neuroglancerZarr", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S1/IA_M2_S1.zarr/0", "metadata": {"shader": "RGB", "dimensions": "XY", "shaderParameters": {}}}]}]}, {"primaryFilePath": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S2.png", "thumbnailIndex": 0, "title": "IA_M2_S2", "fileMetadata": {"omeXml": null}, "imageSet": [{"imageName": "IA_M2_S2", "imageMetadata": null, "assets": [{"type": "thumbnail", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S2/IA_M2_S2_sm.jpeg"}, {"type": "keyImage", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S2/IA_M2_S2_lg.jpeg"}, {"type": "neuroglancerZarr", "path": "LID/jecohen/maximovao-2023-1102-MTB/RNA-2023-1102-Morphology_Tag-IA_M1_S1-4/Pre_ROI_Images/IA_M2_S2/IA_M2_S2.zarr/0", "metadata": {"shader": "RGB", "dimensions": "XY", "shaderParameters": {}}}]}]}]}
Note: "fileMetadata": {"omeXml": null},
etc
Noted. Although this is not an empty object, that's still the opposite of what I'm asking for - a metadata object with no payload. So not fixed, I presume...
UPDATE: the Multichannel pipeline produced a proper metadata piece, with a nonblank path in it. Can't tell if there is a code path in that pipeline that omits the path.
{}
with {"omeXml":null}
is not what I was after. Please reopen this issue, or I'll create another.The new response should be as follows,
omeXml
, the response should be fileMetadata: null
.omeXml
, the response should be fileMetadata: {"omeXml": "/path/ab/c"}
.We assumed that is what was expected. Please correct me if I am wrong.
That's what I was after, yes. That's not the current behavior, as seen in QA - I'm getting "fileMetadata": {"omeXml": null}
instead.
Retested, this time fileMetadata is null.
Debugging https://jira.niaid.nih.gov/browse/HEDWG2-1063 . The file level metadata for the referenced image goes like this was either a blank object (
{}
), or an object with null omeXml ({"omeXml":null}
).The pipeline run to check was in QA, and the ID on the pipeline run was S9468.
Please pass a null/missing fileMetadata instead of that. Storing objects with no payload is a waste of storage.