Added two parameters to the StationData class: slope angle and slope azimuth (with setter and getters)

Bugfix for issue 147: When copying MeteoData object it could not be guarenteed that the index for a certain paramater would remain constant, thus accessing let's say md.param(13) and md2.param(13) could result in accessing different values. Since MeteoIO uses the index extensively it is of vital importance for the index to stay constant.

This problem only affected additional parameters, not the standard parameters.

Speedup for the MeteoData constructor and copy constructor, by looping through the individual parameters instead of seeking them out one by one.

Removed potentially deadly line in BufferedIOHandler: setDfltBufferProperties was called in the destructor (without being able to hold the no throw promise. Added some brackets to make if statements easier to comprehend.

Completed the feature to push meteo data into the IOManager. Now both levels IOManager::raw and IOManager::filtered are supported permitting the user to push data either in a raw state or in a filtered state into the IOManager. This feature is essential in order to bypass internal reading of meteo data.

Added a possibility to push meteo data into the IOManager in case the user does not wish to read meteo data with MeteoIO but wishes to use other functionality provided by the IOManager (e.g. resampling, interpolation, filtering):

iomanager.push_meteo_data(IOManager::filtered, date_start, date_end, vec_meteo);

Bugfix: The function bufferData is not equivalent to iomanager.readMeteoData! bufferData resets the buffer indices, thus a wrong date for buffer_start was present when requesting a big chunk of data (e.g. 1 month).

Finally, the Windows packaging works cleanly. The start menu links work, the add/remove software texts and icon work, the components installation works.

The popc autotools build was not working (I forgot to put the WindAvg filter). This has been fixed alongside some cpack fixes for the Windows packaging.

An icon has been designed and the CMakeLists.txt improved for building the Windows installer. This is not yet perfect, but we are moving forward!

Fixed a few compilation warnings found on Illusion. The buffer centering has been introduced in BufferedIOHandler: the user can specify centering as a fraction of the buffer size or as a duration before the requested date (when rebuffering). The documentation has been updated accordingly.

Debugging output cleanup in CosmoXML.

Write meteo data: OK

Read station data: implemented.
In this routine, the user must choose (= comment/uncomment) he we write every station or if we use the given date to filter which station are written

Notice: have to check how dateStart and dateEnd are chosen if we provide a single date (eg in meteo_reading.cc)... Now it takes "date - 6 hours" --> "date + 15 days - 6 hours"

The Date::setFromSys() method should now be portable (it was only BSD/GNU before). A constness issue has been fixed in DynamicLibrary for Windows platforms.

No commit message

The WIND_AVG filter has been ported. The problem is that it is not possible to get at the same time VW and DW. This will have to be fixed later... 

More documentation has been written. 

Documentation update. A new section "how to build your io.ini configuration file" has been added to the end user section of the documentation.

The autotools' Makefile.in that is necessary for the popc version has been updated. The io.ini and data_converter.cc have been improved (some errors/obsolete keys were in io.ini)

Writes XML-files without the need of an external text-file

TO DO
Interpolation? Depends on the FieldExtra output format
Add variable MaxWindSpeed

Fixed a bug when doing dates additions and substractions (the += and -= operators were fine). Fixed the copy constructor so that it can copy an undefined date. Added the support for Matlab dates.

setFromSys() now properly sets the system time zone. A bug in the default constructor has been fixed.

The popc version has been updated. It is far from perfect (lots of popc specific hacks), it lacks an IOManager parclass, but this is the minimum that we need for Alpine3D...

The ptr_to_function to ptr_to_object cast warning has been turned off (at least for gcc). This does not fix the core issue (that has to be dealt with by the future c++ standard), but at least we don't see it anymore...

Fix for issue 131: If the setProcessingLevel(IOManager::filtered) procedure is called, no more resampling is performed when calling getMeteoData(date, vecMeteo).

More meteo laws have been implemented, alongside new constants. An extra check has been added for the USER spatial interpolation algorithm.

The custom grids spatial interpolations method is now working! The user can force loading his own grids for specific parameters and timesteps.

The Standard Deviation filter has been moved to the new structure. Some optimizations for windowed_filters will have to be implemented at some point... (see issue 132 for some ideas)

Another commit related to timezones... The Unix date always being in UTC, this has been fixed. The resampling was not using dates in a proper way, this has been fixed. The GSN plugin has also been fixed, it now rounds time to the closest minute (in order to be consistent with our advertised "1 minute time resolution"). A rounding method has been implemented for dates.

The timezone parameter in io.ini is now named TIME_ZONE instead of TZ for clarity. 

Undefined dates should now be fully handled (ie: also in arithmetic).

- Date.*: allow to handle undefined Dates (== 0. for now), add new members undef, setUndef(), isUndef(), and rndJulianDate()
- IOUtils.*: replaced TZ with time_zone and TIME_ZONE
- ImisIO.*: correct typo seperateDrift to separateDrift