Fix native_msg reader for ROI input and multi-part order file patterns #282
Conversation
…ns. Add a UMARF check when reading the header. Clean-up code.
…g in the calibration function and a code cleanup
There was a problem hiding this comment.
Why weren't any of these changes for ROI inputs included before? Did no one ever test ROI before?
Additionally, we should probably migrate the areas file to be in YAML (since pyresample supports it now), but that doesn't have to happen in this PR.
Otherwise, I just have an issue with the file patterns.
| file_patterns: ['{satid:4s}-{instr:4s}-MSG{product_level:2d}-0100-NA-{processing_time1:%Y%m%d%H%M%S.%f}000Z-{processing_time2:%Y%m%d%H%M%S}-{order_id:d}.nat', | ||
| '{satid:4s}-{instr:4s}-MSG{product_level:2d}-0100-NA-{processing_time1:%Y%m%d%H%M%S.%f}000Z-NA.nat', | ||
| '{satid:4s}-{instr:4s}-MSG{product_level:2d}-0100-NA-{processing_time1:%Y%m%d%H%M%S.%f}000Z-{order_id:d}.nat'] | ||
| file_patterns: ['{satid:4s}-{instr:4s}-MSG{product_level:2d}-0100-NA-{processing_time1:%Y%m%d%H%M%S.%f}000Z-{order_id:s}.nat'] |
There was a problem hiding this comment.
This doesn't include the first original pattern which had a processing_time2. If this file pattern doesn't actually exist (check with person who originally added it) then this is fine.
There was a problem hiding this comment.
processing_time2 is not used anywhere, the rest of the filename is now caught under order_id.
There was a problem hiding this comment.
yes, exactly. those file parts are not needed.
There was a problem hiding this comment.
@djhoese what do mean with "Why weren't any of these changes for ROI inputs included before?" The area_extent?
|
And by "not included before" I mean in old satpy or mpop. |
djhoese
changed the title
Codecov Report
@@ Coverage Diff @@
## develop #282 +/- ##
===========================================
- Coverage 64.47% 64.39% -0.08%
===========================================
Files 114 114
Lines 13997 14008 +11
===========================================
- Hits 9024 9020 -4
- Misses 4973 4988 +15
Continue to review full report at Codecov.
|
|
@sjoro My question about ROI changes: This reader was part of mpop right? Or at least it existed in satpy before (version <0.9), right? Were people not using this reader with ROI datasets before this? Basically, why were you the first one to find this bug? I figured it was some mistake you made in converting the reader to xarray/dask, but this PR looks like it is all new logic in the reader for figuring out of it is an ROI dataset. |
|
@djhoese ah, ok. yes, the reader was in mpop, but support for ROI-files was never implemented, only full disk files. we just got the ROI-area support fixed at the last workshop in Oslo. so, basically ROI-features were not tested before. |
|
Ah ok, sounds good then. |
Fixes issues #277 and #278. Fixes a bug in native_msg.py calibration, a wrong index was selected for the calibration coefficients. Fixes the area_extent-definition for native_msg. Adds two new areas to area.def. Performs code clean-up.