|May 31, 2012|
|7 years 5 months ||New methods "getNx(), getNy(), getNz()" have been
implemented for easy looping over arrays/grids.
|7 years 5 months ||An optimization from Yesterday was actually breaking the whole
resampling... This has been fixed (thanks to Nander and David for
finding this problem!)
|7 years 5 months ||reverting wrongly committed files
|7 years 5 months ||Fixed a type mismatch
|7 years 5 months ||A cleaner way of implementing the Date optimization has been
implemented (relying on a define to switch back to the previous
implementation if necessary). The compilation of the GSN plugin is
now turned on by default. The version numbers have been pu
|May 30, 2012|
|7 years 5 months ||A few "consts" have been added. The dates are now
considered to be positive which allows comparison operators to be
optimized: converting 1 year of hourly data with data_converter got
almost 30% faster. Moreover, when looking for resampling window, t
|7 years 5 months ||oups, fixed a typo
|7 years 5 months ||Two new members have been added to Grid3DObject, to store the
heights of the 3D levels (ie: level 1 = 10m above ground, etc). This
is similar as what is done by ARPS, GRIB, etc A boolean specifies if
these are absolute or relative heights (ie: above
|May 25, 2012|
|7 years 5 months ||The smoothing filters now behave differently depending on the
centering that has been specified. The IOManager backreference for
Meteo2DInterpolator (pointer this given in the initialization list)
has been replaced by a setter.
An invalid flag for V
|May 24, 2012|
|7 years 5 months ||Fixed more warnings, found by Code::Blocks. A bug when compiling
gsoap files with migw has been fixed by a workaround (adding a macro
definition in the CMakeLists.txt).
|7 years 5 months ||Fixed several warnings found with Visual C++
|7 years 5 months ||Better error messages in IMISIO, better safety (regarding
"undef") in Date and some code cleanup/const/reserve
business (all benchmarked) for the others. We also now avoid
stacking two "/" at the end of the PLUGINPATH if one is
|May 23, 2012|
|7 years 5 months ||oups, fixed an include that had to be removed... By the way, the
last changes lead to good compiler optimizations: in optimized mode,
on 1 year of data, the speed up is above 30%
|7 years 5 months ||The last filters that had not been ported to the new filtering
infrastructure have been ported and their documentation updated. All
filters now use window specifications (instead of vector of
pointers). The misleading filter argument "index" (for par
|7 years 5 months ||All the filters using windows have been modified so they use the
window specifications (start, end) instead of a vector of pointers.
The FilterWindAvg was not properly handling nodata, this has been
fixed. By the way, when using a reasonnably large b
|May 22, 2012|
|7 years 5 months ||This is the new implementation of the filters' windowing scheme.
Now, specifying 4 points and 6 hours will guarantee that at least 4
points are in the window AND 6 hours, otherwise it returns
unsuccessfully. The run time penalty is ~4% compared to th
|7 years 5 months ||First correct implementation of the filtering windows. This is not
active yet as it only provides window specifications. But it
properly fulfills both the number of points criteria and the time
|May 21, 2012|
|7 years 5 months ||Undoing the last commit: the /2 was right and I was wrong...
|May 16, 2012|
|7 years 5 months ||Fixed a division by zero in libsmet and the "none"
resampling was broken (it was deleting all data instead of
propagating available data). Fixed!
|7 years 5 months ||The "passive_T" filter had a bug in its implementation,
this has been fixed (using the wrong statistical fit). It should
still be tested before operational use, though...