add option to fits reader to open L1 products also in UTC >L1 UTC per default (via keyword in header)#447
add option to fits reader to open L1 products also in UTC >L1 UTC per default (via keyword in header)#447nicHoch wants to merge 4 commits intoi4Ds:masterfrom
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #447 +/- ##
==========================================
- Coverage 72.23% 67.04% -5.19%
==========================================
Files 78 76 -2
Lines 8085 8136 +51
==========================================
- Hits 5840 5455 -385
- Misses 2245 2681 +436 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
samaloney
left a comment
There was a problem hiding this comment.
I'm not sure allowing operations in different time systems is a good idea. A product should either have time in OBT or UTC and the conversion should only be done once in the from_level0 or similar methods?
All products (>L0) will always have obt_timerange from products will have an utc_timerange but only if the time sys in header is UTC and not sure supporting on the fly conversion is a good idea because the spice kernel dependence.
This aren't blocker just thoughts
|
i agree that it is not ideal to provide some of the on the fly conversions. The auto converting most of the time (and if proper used never) will not happen as our process steps will do this explicit. In general this L1 product could have state depended time systems and i think it would be bad experience if the properties will raise an exception if used |
9b9c9dc to
748ae6a
Compare
748ae6a to
5a7c670
Compare
add keyword: get_timeformat_from_TIMESYS to Product(path) method.
If set time and timedel column in data table are converted to timesystem provided in the TIMESYS header keyword.
some products properties like utc_timerange, min_exposure ... might do autoconversions based on the internal timesystem but it should be avoided to open L1 files in UTC for processing them further.
Use get_timeformat_from_TIMESYS=True only for reading excess to L1 data, read/write/combing might lead to timing accuracy issues