Improved an error message and added a note in the documentation that libproj is currently not thread-safe (as well as a little code cleanup)

This is finally a proper version of the Solar resampling: it works for multiple stations and is well optimized.

The Solar resampling now works with multiple stations, but this canceled out some of the optimization gains...

The new Solar resampling algorithm has been re-written for performance. It is now 16% faster when compiled in debug mode. It will still need to be tweaked to handle multiple stations (otherwise, the cached parameters would be reused between stations).

This is the first version of a solar radiation resampling algorithm. It seems to work quite well but it is still quite slow (ie it will need to be optimized by caching quite a few things, for each station calling it).

First commit to adress issue 543: now on Posix systems, the readDirectory function can perform a recursive search

The Karsten method for converting cloud cover (in okta) to clearness index has been implemented and is also available as a data generator (looking for a "CLD" parameter to use as cloud cover data)

code cleanup and added the possibility to provide a date range to the smet_range script (there is no default "." directory anymore, the directory MUST be provided)

A OSHD_DEBUG key has been added and the units are now properly supported when using commit https://github.com/tbeu/matio/commit/12b7d4051369716ef21a53540d1f3354fe0d4e97 of libmatio (the UTF-16 is now returned as ISO-8859-1). 

Small optimization in the SNIO plugin.

The OshdIO plugin, based on libmatio for reading Matlab binary files, is coming into shape... almost there!

A bug has been fixed in CosmoXMLIO: it was reading all the data with a time-shift (ie always reading the file *after* the timestep that it should have read).

A few documentation glitches have been fixed in IOInterface and Date and an invalid default value for the time zones fixed in SMETIO (instead of taking 0 as intended, it was always nodata). The missing files for the new OSHD plugin are now committed, although the plugin is NOT READY to be used yet.

Introduced the basic structure that will be required to build a new plugin to read Matlab files prepared by the OSHD project/service. This is not usable yet but should come quite soon.

When opening the xml files with a forced encoding, the files were not being properly closed, leading to a possible "Too many open files" error message

The SUPPR filter can now take as argument a file containing a list of station IDs and timesteps where the parameter should be deleted

A resampling method has been added to the Grid2DObject so it is possible to change the cellsize. The support for the '*' wildcard has been added to the KEEP and EXCLUDE keys.

The MOD operator has been added to Date (it takes a number of seconds in argument). This is used by the new VSTATIONS_REFRESH_RATE key to avoid calling the spatial interpolations too often in the virtual stations. More checks have been implemented in the virtual stations and dowscaling for added safety. 

A major bug has been found when reading 3D grids in ARPS: the indexes when not in the right order.

A bug has been found in NetCDFIO: it should look for the GRID2D key, not the METEO key! In Coords, the lat/lon are checked for validity in setLatLon() in order to catch grossly out of range coordinates (as was found on the Snowpack operational system for a few stations).

Fixed a bug in libinterpol2D, improved the documentation of the ALS_SCALING interpolation method, fixed a few bugs in BormaIO and added a constant in Meteoconst.h

Fixed a bug when reading 3D fileds, improved the documentation and renamed a method to a more proper name

The layers handling is not so clear... it seems that the first layer is most often a purely numeric layer and therefore must be skiped. Some fields may not have such a layer, this will have to be clarified.

In order to add 3D grid reading/writing methods to the plugins, the whole architecture of IOInterface has been changed: this is not a purely virtual class anymore, it now always implements the "nothing implemented here" exception. This makes the plugins more compact (there is no need to implement empty methods for what the plugin can not support) and much clearer. The documentation and template have been changed acoordingly.