libsmet: a new convert_to_double implementation, based on strtod from the cstdlib, cuts conversion time to about a fourth of the stringstream implementation.

GSNIO: Missing NULL values and missing fields and units keys in the header have been fixed on the GSN server side. Plugin updated accordingly.

GSNIO: Now we can deal with offsets and multipliers, which are sometimes indicated in a UTF8 string after the fields definition in the header. The units are given as "°C" or "%" for instance. Thus we can perform a calculation to MKSA. 

GSNIO: Cosmetic adjustments. If the user requested station does not exist, there will be an exception thrown. Otherwise the most pressing issue remains the NULL values in GSN.

GSNIO: reading of meteo data works now. However there are the usual GSN issues, which are plentiful and painful:
1) Big problem: NULL values seem to actually come as 0 values
2) parameter naming conventions: totally inconsistent, also one station might have more than one sensor for a parameter, which one to use?
3) Slow, impossible to retrieve meta data for one station only, meta data formatted in obscure way -> necessary to guess where the timestamp is (current guess position 1 or 2 in the csv formatted data row).

GSNIO: reading of StationData now works perfectly,although terribly slow, due to the fact that it is not possible to retrieve meta data for a selected number of stations only.

The LIDW_LAPSE method is now back. It still works the same, ie it can produce very unatural results depending on the stations set that is used!

Deleting old GSNIO plugin based on SOAP messages and adding new framework based on REST API and libcurl (gnutls variant).

The version number has been changed to 2.4.1 (this will most probably be an internal release shipped with Snowpack).

Bringing back the LIDW spatial interpolation method that perfoms IDW on a subset of the stations. At least it compiles...

The Forland1996 undercatch correction as well as the coefficients for WFJ as calculated by Nander Wever have been implemented

Two new undercatch corrections have been implemented: for the Chinese Standard Precipitation Gauge and for the shielded Geonor gauge.

A documentation bug has been fixed in GeneratorAlgorithms. Some warnings and white spaces have been fixed in GeneratorAlgorithms.

A line was commented out that should not have been... Therefore the geolocalization information was not properly read!

This addresses issue 335 by providing data quality output (it has to be enabled at compile time): it reports all points that have been filtered or corrected as well as all points that had to be resampled to replace a nodata value by a proper value (ie resampling between measured points is not reported).

A small bug has been found in the Unsworth generator (when iswr>0 && rswr==0, this was leading to a division by 0). A suppression filter has been implemented (it simply deletes all the data for a given parameter).

More work on the Unshade processing element. Unfortunatelly, it is slow, it works very poorly and it is complicated... 

This is the very preliminary version of a filter to remove shade on a short wave radiation sensor: when both ISWR and RSWR are measured, we compute the measured albedo, remove all points that are out of range, filter with a MAD filter, linearly re-interpolate the albedo for the missing points (either no measured albedo or filtered out) and then reconstruct the chosen signal (either ISWR ort RSWR) using the albedo and the other signal as reference. This would most probably be most useful on ISWR, but who knows...

The date rounding was not always working properly since it was rounding the GMT date. This has been fixed (so rounding a local time down to the nearest day now properly returns the local day at 00:00). Identation correction in libinterpol1D.

Documentation update: hopefuly clearer for first time users (more links added) and removing an obsolete comment in Accumulate

Finally! Handling precipitation redistribution before re-accumulation (ie the 24 hours sums provided in an otherwise hourly data file) now works with non-uniform timesteps. This introduced some runtime performance hit (~10% slower) but better accomodate the needs of our users!