Remove warning for empty (NRTI) granules
Add metadata filter
Explicitly add the actual filename to the extracted metadata (issue #27361)
Capture empty input granules before they cause damage
Silence some known warnings, or reduce them to info messages
Add job-order file name and configuration file name(s) to netCDF4 output as attributes
Add check on fraction of "good" pixels before reading a granule
Add robustness against missing groups
Report time difference at info level
Add difference configuration maker
Improve handling of empty granules in NRT processing
Update zonal average methods, is now a separate routine.Add hemispheric and land/sea statistics to Histogram.Update report creation.
Avoid taking the minimum of a zero-length array
Catch error for empty data
Report range after applying filter
Add data selection based on valid_min/valid_max
Handle incorrect tracking_id attribute in NPP-Cloud productChange time handling in time dependent plottingHandle missing support variables in NPP-CloudHandle row count mismatch for cloud comparison
Match scanlines by time
Change handling of 'empty' variables.
We now check for the solar zenith angles in the file.If these are all > 88 degrees, we do not expect to findvalid data, and the granule is skipped. Before the variablewas removed.
Support map_scale key, improve error handling when reading scaling
Allow for selective exclusion of variables from L3 creation
Updated parameters for CH4 and CO
Skip files where the scanline dimension has length zero
Add input granule counter variable
Change input file count storage
Add the number of input granules as a global attribute.
Exclude variables from scatterplot if range is 0
Do not cast irradiance data to a masked array.
Improved handling of unconfigured variables; clarifying error/warning messages; improved product detection; better full_synchronize detection.
Remove variable when an empty granule is encountered, when "full_synchronize" is not in effect
Add full_synchronize key, cleanu up reporting
Update syncing rules to avoid masking all data by an empty variable
Ensure that transformer object is only used when not None
Update version number to 0.3.2
Write and read version number of PyCAMA to/from output to avoid version mismatches
Transformation I/F change, real product name mapping support, count events per granule.
Transformer filters now store keyword parameters per filter, not per variable.
Properly deal with empty variables.
Clarify error message in case of length mismatch (see above).Allow of internal and external product types (for comparing different versions of the same product).Ensure that fill values are correctly dealt with.
Correct custom mapping of products.
Handle special products; get variable in read_irradiance_data() to avoid stale references; close file references asap.
Add include_scatter option to allow exclusion from scatter density plots
Fixes for NP_BDx and OMI L2 files combined
Cleanup of the temporary QA4ECV hack
Add processing mode dependent variables
Periodically check the warning counter; make sure # ground_pixel is lowest encountered (filter will deal with rest)
Changes for reading OMI L2 data as well, update to version 0.1.3
Capture that other error as well
Extract processing performance figures
Correctly capture transformation errors
Clarify error message for easier debugging
Show progress to 0.1 % (NRTI contains 144 files, 1% is insufficient).
Print correct orbit number (not orbit data). Capture more errors.
Improve robustness agains missing or empty attributes.
Remove variables that are not available in the input, no longer stop on missign variables; Clean up logging, add progress messages
switch to h5py
Add input pointer extraction; detect variables without scanline dimension; deal with non-synchonization (incomplete)
Add option to configure (build) variables from configuration input.
Add collecting of variables that do not depend on the 'scanline' dimension.
Add event counting
PyCAMA updates, removal of vlen data (as much as possible), others
Check that first dimension is a time dimension of length 1 before cutting it off
Filter out NaN values from synthetic variables
Worldplot & histogram now working
More analysis types & plot variants.
Add combined data for global analysis
Include loop over time steps
Turn PyCAMA into a real package, including load magic for plugins.