arrayof_ttr_mapper_append arrayof_ttr_mapper_append_unique arrayof_ttr_mapper_delete arrayof_ttr_mapper_insert arrayof_ttr_mapper_insert_array arrayof_ttr_mapper_insert_sorted arrayof_ttr_mapper_insert_unique arrayof_ttr_mapper_packarray arrayof_ttr_mapper_remove count_distinct_sorted_ttr_mapper count_distinct_ttr_mapper index_sorted_ttr_mapper index_ttr_mapper inssor_ttr_mapper map_distinct_ttr_mapper map_inv_distinct_ttr_mapper pack_distinct_sorted_ttr_mapper pack_distinct_ttr_mapper sort_ttr_mapper

pack_distinct_sorted_ttr_mapper() type(ttr_mapper) function, dimension(dim) pack_distinct_sorted_ttr_mapper ( type(ttr_mapper), dimension(:), intent(in) vect, integer, intent(in) dim, logical, dimension(:), intent(in), optional mask ) compatta gli elementi distinti di vect in un sorted array Definizione alla linea 938 del file stat_proc_engine.F90. IF (time == datetime_miss .OR. .NOT.c_e(timerange%p1) .OR. .NOT.c_e(timerange%p2) .OR. & ! (timerange%p1 > 0 .AND. timerange%p1 < timerange%p2) .OR. & timerange%p1 < 0 .OR. timerange%p2 < 0) THEN ! is this too pedantic and slow? decoded%pstart = datetime_miss decoded%pend = datetime_miss decoded%reftime = datetime_miss RETURN ENDIF IF (time_definition == 0) THEN ! time == reference time decoded%reftime = time decoded%pend = time + p1 decoded%pstart = decoded%pend - step ELSE IF (time_definition == 1 .OR. time_definition == 2) THEN ! time == verification time decoded%pend = time decoded%pstart = time - step decoded%reftime = time - p1 ELSE decoded%pstart = datetime_miss decoded%pend = datetime_miss decoded%reftime = datetime_miss ENDIF END SUBROUTINE ttr_decoder_decode_pop ! set time, timerange%p1, timerange%p2 according to pstart, pend,