PSI sics-cvs-psi_pre-ansto

integrate.w

@@ -0,0 +1,44 @@
+\subsection{Integrate}
+For four circle diffractometers it is usefull to be able to integrate the
+intensity of the peak from a collected counting profile. This module does
+just this. It uses the method as described by Grant \& Gabe in J. Appl. Cryst
+(1987), 11, 114-120. This method uses a moving window technique for finding
+the limits of the peak. This method was choosen for reason of its
+simplicity. The authors state, that the results are comparable to the
+Lehmann Larsen method used elsewhere. This is a utility module. Just a
+single function is implemented:
+
+@o integrate.h @{
+/*--------------------------------------------------------------------------
+
+                             I N T E G R A T E
+
+  Peak integration follwing the method described by Grant & Gabe in
+  J. Appl. Cryst (1978), 11, 114-120.
+
+  Mark Koennecke, March 1998
+----------------------------------------------------------------------------*/
+#ifndef GABEINTEGRATE
+#define GABEINTEGRATE
+
+#define INTEGLEFT      -1
+#define INTEGRIGHT     -2
+#define INTEGNOPEAK    -3
+#define INTEGFUNNYBACK -4
+
+    int GabePeakIntegrate(int m, int iCounts, long lCounts[], 
+                          float *fIntens, float *fVariance);
+#endif
+@}
+
+Input to this routine is the number of counts and the counts themselves.
+Then m, the number of lookahead points to use. As a result fIntens will
+contain the summed Intensity and fVariance the sigma of the summed
+intensity. The function retusn 1 for success or a negative error code in
+case of trouble. These error codes are:
+\begin{description}
+\item[INTEGLEFT] no left side background found.
+\item[INTEGRIGHT] nor right side background found.
+\item[INTEGNOPEAK] The data does not contain a peak.
+\item[INTEGFUNNYBACK] the data has a strange background.
+\end{description}