|Mar 1, 2016|
|5 years 4 months ||More warning fixes, removed/fixed a few "HACK" comments
|5 years 4 months ||Fixing some warnings on osX with CLang
|Feb 29, 2016|
|5 years 4 months ||Lots of small improvements and bug fixes for reading ARPS files. Now
the 2D grids should work quite well but the 3D grids are still some
kind of a hack...
|Feb 26, 2016|
|5 years 5 months ||More flexible handling of metadata: if some fields are missing, it
now keeps going
|5 years 5 months ||Suppressed a warning that was getting too verbose at times. Made
ARPSIO more thread-safe.
|Feb 24, 2016|
|5 years 5 months ||Removing member variable "ofstream fout"
|Feb 23, 2016|
|5 years 5 months ||More plugins becoming thread-safe
|5 years 5 months ||Making the plugins thread safe by removing the global fin/fout.
Please try to avoid setting these as members for new plugins!
|Feb 20, 2016|
|5 years 5 months ||Now GSN uses the virtual sensor name as station_id, making it much
more logical when using exclude/keep/merge statements. The NetCDF
plugin does not require anymore to be configured for the inputs in
order to produce output grids. Otherwise, code and
|Feb 10, 2016|
|5 years 5 months ||Code cleanup: more consistency in the handling of
keep/exclude/merge. Better, more explicit error messages.
|5 years 5 months ||Small fix: invalid user/password now throws a proper exception
(instead of the default exception that there is no data)
|Feb 9, 2016|
|5 years 5 months ||Fixed a bug: since the variance could be nodata, it should be
checked before taking the sqrt
|5 years 5 months ||Fixed a warning found on osX
|5 years 5 months ||Making the order of the min and max rates of change unimportant in
the Rate filter. Fixed a small logic bug in the snow/noSnow filter.
|Feb 8, 2016|
|5 years 5 months ||Some small optimizations, code cleanup, documentation clean up and
better handling of the nodata cases. The filter should now be ready
|Feb 5, 2016|
|5 years 5 months ||Several small bugs have been found in FilterSnowNosnow. Now it
behaves in a more logical manner. However, the logic could still be
improved (it should keep the same filtering strategy for the whole
day instead of re-evaluating at each time step). A p
|5 years 5 months ||Now the full method has decsribed in the CCES report has been
implemented. Please note that there is still a bug hidden somewhere
that has to be found...
|Feb 3, 2016|
|5 years 5 months ||More detailed error messages when the timestamps are not properly
sorted. Better behavior of the GSN plugin when the data is missing
but the metadata could be retrieved (so at least we know that the
station is valid).
|Jan 29, 2016|
|5 years 5 months ||Code cleanup, removing some unecessary stuff for CTest
|Jan 28, 2016|
|5 years 5 months ||Now the SHADE filter will use the last valid splitting coefficient
if it can not compute one, within 24 hours. When the measured
radiation is too low, no correction is applied because it can only
be diffuse radiation (at dawn/dusk).