DM-18739: Generate defects from standardized format #66
Conversation
Since pipe_tasks depends on this repository, we can't depend on the ingestDefect.py script to ingest. We need to ingest and persist the ingested version in this repository
|
Note that defects in standard form now reside in obs_test_data. |
data/utils/defectsFromBias.py
Outdated
| assert len(bboxList) == len(test2BBoxList) | ||
| for boxA, boxB in zip(bboxList, test2BBoxList): | ||
| assert boxA == boxB | ||
| test2defectList = Defects.readText(DefectsPath+".ecsv") |
There was a problem hiding this comment.
Would it work to include the ".ecsv" suffix on DefectsPath itself -- would the correct file get written? If so, I think it would make the code a bit simpler and easier to understand. Also the description string would give the correct path for the output file (right now it is missing the ".ecsv" suffix).
There was a problem hiding this comment.
I think you want to get the return value from writeText and use that directly in readText (writeText forces the file extension based on the type of text file it is writing). The debug print message should also use the actual name used as returned by writeText.
There was a problem hiding this comment.
That sounds reasonable, though it means that the full name of the file cannot be included in the help string for the command line parser is produced. I suppose one could always print a message when saving the data.
There was a problem hiding this comment.
I believe I have addressed this.
SimonKrughoff
force-pushed
the
tickets/DM-18739
branch
from
896aa4a
to
a238aa8
Compare
SimonKrughoff
force-pushed
the
tickets/DM-18739
branch
from
a238aa8
to
e0c72cc
Compare
This means the tickets/DM-18739 ticket will need to be merged with the tickets/DM-19857 branch of ap_verify
| @@ -0,0 +1,14 @@ | |||
| # Config override for lsst.pipe.tasks.IngestCalibsTask | |||
There was a problem hiding this comment.
Did you mean IngestDefectsTask?
No description provided.