[Vegan-commits] r2809 - in pkg/vegan: R inst

[noreply at r-forge.r-project.org]

Author: jarioksa
Date: 2013-12-06 19:37:41 +0100 (Fri, 06 Dec 2013)
New Revision: 2809

Modified:
   pkg/vegan/R/envfit.default.R
   pkg/vegan/inst/ChangeLog
Log:
envift does not choke on character variables but ignores them (with a warning)

Modified: pkg/vegan/R/envfit.default.R
===================================================================
--- pkg/vegan/R/envfit.default.R	2013-12-05 12:34:46 UTC (rev 2808)
+++ pkg/vegan/R/envfit.default.R	2013-12-06 18:37:41 UTC (rev 2809)
@@ -1,4 +1,4 @@
-"envfit.default" <-
+`envfit.default` <-
     function (ord, env, permutations = 999, strata, choices = c(1, 2), 
              display = "sites", w = weights(ord), na.rm = FALSE, ...) 
 {
@@ -32,9 +32,13 @@
     }
     if (is.data.frame(env)) {
         facts <- sapply(env, is.factor)
+        vects <- sapply(env, is.numeric)
+        if (!all(facts | vects))
+            warning("the following variables are ignored because they are neither numeric nor factors:\n",
+                    paste(colnames(env)[!(facts | vects)], collapse=", "))
         if (sum(facts)) {  # have factors
             Pfac <- env[, facts, drop = FALSE]
-            P <- env[, !facts, drop = FALSE]
+            P <- env[, vects, drop = FALSE]
             if (length(P)) { # also have vectors
                 vectors <- vectorfit(X, P, permutations, strata, 
                                      choices, w = w, ...)

Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog	2013-12-05 12:34:46 UTC (rev 2808)
+++ pkg/vegan/inst/ChangeLog	2013-12-06 18:37:41 UTC (rev 2809)
@@ -20,6 +20,15 @@
 	adapted to new anova.cca and this brought along similar changes in
 	API. 
 
+	* envfit: function assumed that environmental variables are either
+	factors or numeric, and choked if they were neither but, say,
+	character strings. Now the function ignores variables which are
+	neither factors nor numeric (with a warning). This is one possible
+	reason for problems reported in
+	https://stat.ethz.ch/pipermail/r-sig-ecology/2013-December/004217.html,
+	although the reported case is irreproducible and we cannot be
+	sure.
+
 	* tests for cca/rda/capscale: commented out test that failed with
 	the anova.cca. Known issue was that 'by = "term"' and 'by =
 	"axis"' stop with error with missing data. This was designed and