Google
 

Trailing-Edge - PDP-10 Archives - decuslib10-02 - 43,50145/rtab.doc
There are 2 other files named rtab.doc in the archive. Click here to see a list.
SUBROUTINE RTAB

PURPOSE
   TABULATE ROWS OF A MATRIX TO FORM A SUMMARY MATRIX

USAGE
   CALL RTAB(A,B,R,S,N,M,MS,L)

DESCRIPTION OF PARAMETERS
   A - NAME OF INPUT MATRIX
   B - NAME OF INPUT VECTOR OF LENGTH N CONTAINING KEY
   R - NAME OF OUTPUT MATRIX CONTAINING SUMMARY OF ROW DATA.
       IT IS INITIALLY SET TO ZERO BY THIS SUBROUTINE.
   S - NAME OF OUTPUT VECTOR OF LENGTH L+1 CONTAINING COUNTS
   N - NUMBER OF ROWS IN A
   M - NUMBER OF COLUMNS IN A AND R
   L - NUMBER OF ROWS IN R
   MS  - ONE DIGIT NUMBER FOR STORAGE MODE OF MATRIX A
	  0 - GENERAL
	  1 - SYMMETRIC
	  2 - DIAGONAL

REMARKS
   MATRIX R IS ALWAYS A GENERAL MATRIX

SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
   LOC
   RADD

METHOD
   ROWS OF DATA IN MATRIX A ARE TABULATED BASED ON THE KEY
   CONTAINED IN VECTOR B. THE FLOATING POINT NUMBER IN B(I) IS
   TRUNCATED TO FORM J. THE ITH ROW OF A IS ADDED TO THE JTH
   ROW OF R ELEMENT BY ELEMENT AND ONE IS ADDED TO S(J). IF J
   IS NOT BETWEEN ONE AND L, ONE IS ADDED TO S(L+1). THIS
   PROCEDURE IS REPEATED FOR EVERY ELEMENT IN VECTOR B.
   UPON COMPLETION, THE OUTPUT MATRIX R CONTAINS A SUMMARY OF
   ROW DATA AS SPECIFIED BY VECTOR B. EACH ELEMENT IN VECTOR S
   CONTAINS A COUNT OF THE NUMBER OF ROWS OF A USED TO FORM THE
   CORRESPONDING ROW OF R. ELEMENT S(L+1) CONTAINS A COUNT OF
   THE NUMBER OF ROWS OF A NOT INCLUDED IN R AS A RESULT OF J
   BEING LESS THAN ONE OR GREATER THAN L.