An optimization from Yesterday was actually breaking the whole resampling......

An optimization from Yesterday was actually breaking the whole resampling... This has been fixed (thanks to Nander and David for finding this problem!)

meteoio/ResamplingAlgorithms.cc

@@ -403,7 +403,7 @@ void ResamplingAlgorithms::getNearestValidPts(const size_t& pos, const size_t& p
 	indexP2=IOUtils::npos;
 
 	const Date dateStart = resampling_date - window_size;
-	if(vecM[0].date<=dateStart) { //we try to find a valid point only if it's even worth it...
+
 	for (size_t ii=pos; (ii--) > 0; ) {
 		if (vecM[ii].date < dateStart) break;
 		if (vecM[ii](paramindex) != IOUtils::nodata){
@@ -411,12 +411,10 @@ void ResamplingAlgorithms::getNearestValidPts(const size_t& pos, const size_t& p
 			break;
 		}
 	}
-	}
 
 	//make sure the search window remains window_size
 	const Date dateEnd = (indexP1 != IOUtils::npos)? vecM[indexP1].date+window_size : resampling_date+window_size;
 
-	if(vecM[vecM.size()].date>=dateEnd) { //we try to find a valid point only if it's even worth it...
 	for (size_t ii=pos; ii<vecM.size(); ii++) {
 		if (vecM[ii].date > dateEnd) break;
 		if (vecM[ii](paramindex) != IOUtils::nodata) {
@@ -425,7 +423,6 @@ void ResamplingAlgorithms::getNearestValidPts(const size_t& pos, const size_t& p
 		}
 	}
 }
-}
 
 /**
  * @brief This function solves the equation y = ax + b for two given points and returns y for a given x