onestep takes one step

Author: kingaa

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: pomp
 Type: Package
 Title: Statistical Inference for Partially Observed Markov Processes
-Version: 5.9.0.0
-Date: 2024-06-02
+Version: 5.9.1.0
+Date: 2024-06-07
 Authors@R: c(person(given=c("Aaron","A."),family="King",role=c("aut","cre"),email="kingaa@umich.edu",comment=c(ORCID="0000-0001-6159-3207")),
 	  person(given=c("Edward","L."),family="Ionides",role="aut",comment=c(ORCID="0000-0002-4190-0174"))	,
 	  person(given="Carles",family="Bretó",role="aut",comment=c(ORCID="0000-0003-4695-4902")),

R/rprocess_spec.R

@@ -52,7 +52,7 @@
 ##' They differ as to how this is done.
 ##' Specifically,
 ##' \enumerate{
-##' \item \code{onestep} takes a single step to go from any given time \code{t1} to any later time \code{t2} (\code{t1 < t2}).
+##' \item \code{onestep} takes a single step to go from any given time \code{t1} to any later time \code{t2} (\code{t1 <= t2}).
 ##' Thus, this plug-in is designed for use in situations where a closed-form solution to the process exists.
 ##' \item To go from \code{t1} to \code{t2}, \code{euler} takes \code{n} steps of equal size, where
 ##' \preformatted{

inst/NEWS

@@ -1,5 +1,11 @@
 _N_e_w_s _f_o_r _p_a_c_k_a_g_e '_p_o_m_p'
 
+_C_h_a_n_g_e_s _i_n '_p_o_m_p' _v_e_r_s_i_o_n _5._9._1:
+
+        • The rprocess plugin ‘onestep’ now always takes exactly to get
+          from one observation time to the next, even when the interval
+          between them is zero.
+
 _C_h_a_n_g_e_s _i_n '_p_o_m_p' _v_e_r_s_i_o_n _5._8._4:
 
         • To add additional elements for use by the basic model

inst/NEWS.Rd

@@ -1,5 +1,10 @@
 \name{NEWS}
 \title{News for package `pomp'}
+\section{Changes in \pkg{pomp} version 5.9.1}{
+  \itemize{
+    \item The rprocess plugin \code{onestep} now always takes exactly to get from one observation time to the next, even when the interval between them is zero.
+  }
+}
 \section{Changes in \pkg{pomp} version 5.8.4}{
   \itemize{
     \item To add additional elements for use by the basic model components (i.e., \dQuote{userdata}), the \code{userdata} argument should now be used.

man/rprocess_spec.Rd

@@ -208,7 +208,7 @@ SEXP euler_simulator
     switch (method) {
     case onestep: default:      // one step
       dt = time[step]-t;
-      nstep = (dt > 0) ? 1 : 0;
+      nstep = 1;
       break;
     case discrete:              // fixed step
       dt = deltat;