Update Tests for HyBIG #76
Conversation
This removes an aster test without any parameters because that causes massive tiling. It also updates the test for tiling due to the new cell count based tiling trigger.
test/hybig/HyBIG_Regression.ipynb
Outdated
| "id": "e8deff1f", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "### Test: Scale Sizes are Positive\n" |
There was a problem hiding this comment.
I got a failure here (I tried it a couple of times):
HTTPError: 400 Client Error: Bad Request for url: http://localhost:3000/C1256584478-EEDTEST/ogc-api-coverages/1.0.0/collections/all/coverage/rangeset?forceAsync=true&outputcrs=EPSG%3A4326&scaleSize=0.002%2C-0.002&granuleId=G1256584570-EEDTEST&format=image%2Fjpeg
There was a problem hiding this comment.
Interesting, what happens if you just visit that url in the failure?
http://localhost:3000/C1256584478-EEDTEST/ogc-api-coverages/1.0.0/collections/all/coverage/rangeset?forceAsync=true&outputcrs=EPSG%3A4326&scaleSize=0.002%2C-0.002&granuleId=G1256584570-EEDTEST&format=image%2FjpegBecause this particular test is actually failing in harmony, not HyBIG, but I left the test in anyway. This is related to the question, should Harmony have a validation check for scaleExtent? See comments on the ticket.
I have the latest and greatest harmony-in-a-box and this is what I see when I visit the URL:
There was a problem hiding this comment.
I did see the comments in the ticket, yup. If this is testing Harmony-core functionality, it probably lives in one of the Harmony-core regression test notebooks.
I also have the latest Harmony-in-a-box. The URL fails when placed directly in the browser, with the error that you have in this comment. But I continue to get this other error when using harmony-py, which likely means that harmony-py is doing something additional here in validating the URL before it gets sent to Harmony, maybe. Do you see different behaviour when running the request via harmony-py?
There was a problem hiding this comment.
Using harmony-py, like the test? It passes for me. Raises the expected error.
There was a problem hiding this comment.
As discussed in Slack - differences here were due to my local environment having an older version of harmony-py. I get this test to pass now.
https://pypi.org/project/jupyter-black/ 👀 |
Update words in comments.
There was a problem hiding this comment.
Thanks for fixing up the whitespace, and sorry for having a stale version of the harmony-py leading to some confusion.
I think it's a good question about the location of the scale extent ordering validation. My gut instinct is that is really something that lives in a Harmony-core test suite, but that is probably a broader effort, because there aren't really those sort of tests in either of the harmony or harmony-regression notebooks.
I am going to leave them here for now, but would like to change that if harmony ends up validating scale_extents. |
Description
Updates the existing regression tests to work with the recent changes to HyBIG.
These include:
Jira Issue ID
DAS-1969
Local Test Steps
Ensure latest HyBIG image is running in your Harmony in a box.
Pull this branch locally and build regression images, specifically hoss and hybig.
Edit the hoss and hybig notebooks to run against localstack and run each notebook against localstack.
PR Acceptance Checklist