Opened 19 years ago
Last modified 16 years ago
#1008 assigned defect
Recipes may be read multiple times — at Initial Version
| Reported by: | Paul Price | Owned by: | Paul Price |
|---|---|---|---|
| Priority: | low | Milestone: | |
| Component: | psModules | Version: | unspecified |
| Severity: | minor | Keywords: | |
| Cc: |
Description
I've suspected some problems in the code that reads the recipes during configuration (psModules/config/pmConfig.c) for some time, but hadn't really noticed anything substantive until just now.
Performing ppArith on a newly created image, an error is generated:
price@mithrandir:/home/mithrandir/price/data2/simtest>ppArith -file1 path://SIMTEST/detrend/SIMTEST.DARKMASK.3/SIMTEST.DARKMASK.norm.3.0.fits -op '|' -file2 path://SIMTEST/detrend/SIMTEST.DARKMASK.4/SIMTEST.DARKMASK.norm.4.0.fits path://SIMTEST/detrend/SIMTEST.20.0 -stats path://SIMTEST/detrend/SIMTEST.20.0.stats -recipe PPSTATS CHIPSTATS -mask
2007-11-13 04:12:41Z|mithrandir |I|psModules.config
Camera SIMTEST, format SIMTEST matches header.
2007-11-13 04:12:41Z|mithrandir |W|formatFromHeader (pmConfig.c:985)
Camera SIMPLE, format SIMPLE also matches header --- ignored.
2007-11-13 04:12:41Z|mithrandir |I|psModules.config
Camera SIMTEST, format SIMTEST matches header.
psMetadataAddItem (psMetadata.c:684) : Duplicate metadata item name: OVER_VAL is not allowed. Use a psMetadataFlags option to allow such action.
This appears to be because the PPSTATS recipe is processed multiple times, and since HEADER is a MULTI, we end up with multiple versions of each of the requested headers.
For now, the problems aren't too bad, but eventually, some careful thought needs to go in to the details of reading the recipes.
