Skip tasks or error out when resources are missing #351
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
☂️ Python Coverage
Overall Coverage
New FilesNo new covered files... Modified Files
|
mikix
force-pushed
the
mikix/skip-missing-resources
branch
from
d6d8f64
to
e66b65c
Compare
mikix
force-pushed
the
mikix/skip-missing-resources
branch
from
e66b65c
to
6262610
Compare
dogversioning
approved these changes
Oct 23, 2024
Comment on lines
43
to
44
| if self.root.protocol in {"http", "https"}: | ||
| return None # we haven't done the export yet! |
There was a problem hiding this comment.
nit - with this and the i2b2 tcp thing below, if i was coming into this codebase fresh, a little more context as to what this means might be helpful (i.e. it's we've got the URL of the place to get the data, not the data itself)
There was a problem hiding this comment.
OK, I added some more context:
# We haven't done the export yet, so there are no files to inspect yet.
# Returning None means "dunno" (i.e. "just accept whatever you eventually get").
return None
- If tasks are selected with --task or --task-filter and a needed resource is missing, we now error out. - If tasks are not specifically selected, we now only run tasks for which resources are available (and error out if no resources exist at all) - If --allow-missing-resources is passed, we skip both above checks and do historical behavior of running any given task with zero input rows (which pushes up a schema, cleans up the delta lake, and stamps the resource as complete in the completion tracking tables).
mikix
force-pushed
the
mikix/skip-missing-resources
branch
from
6262610
to
96442bc
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This addresses a lingering issue from #296, where we didn't adequately protect ourselves from incorrectly handling "empty set" inputs and wrongly marking them as complete. I've decided to go the simplest route and refuse to handle empty-sets unless specifically told to. I think this is appropriate given how unlikely empty sets are in the real world. (We see them with Cerner for some resources like ServiceRequest, but it's the exception not the rule.)
(An alternate solution that I did consider was looking at the bulk export log if it was present, and seeing what resources were exported - then continuing if the export log says it exported Patients, even if no patient files where present. But it seemed more complicated code-wise and more brittle if users are moving files around. The approach I went with requires a human in the loop to agree to allow empty sets. Or... someone just adds the --allow flag to their scripts and we lose this protection anyway - but that's at least not on me. 😄)
Checklist
docs/) needs to be updated