Final revdep checks

Author: hadley

cran-comments.md

@@ -1,16 +1,27 @@
 ## Test environments
-* local OS X install, R 3.1.2
-* ubuntu 12.04 (on travis-ci), R 3.1.2
+* local OS X install, R 3.2.0
+* ubuntu 12.04 (on travis-ci), R 3.2.0
 * win-builder (devel and release)
 
 ## R CMD check results
 There were no ERRORs, WARNINGs or NOTEs. 
 
 ## Downstream dependencies
 This release changes the backend of stringr from R's built in regular 
-expression engine to use instead use ICU's, as provided by the stringi
-package. The potential for backward incompatible changes was high so I have
-R CMD check on all 155 downstream dependencies 
-(https://github.com/hadley/stringr/blob/master/revdep/summary.md). As well
-as carefully reviewing the results, I notified all downstream maintainers
-of the pending change in early Dec and again on April 14.
+expression engine to ICU's, as provided by the stringi package. The potential 
+for backward incompatible changes was high so I have R CMD check on all 158 
+downstream dependencies (https://github.com/hadley/stringr/blob/master/revdep/summary.md). 
+
+All downstream maintainers (both en masse, and individually in the case of 
+problems) were notified of the change in early Dec and again on April 14
+(two weeks ago). 
+
+Despite these warnings, I have identifed six packages that now fail R CMD 
+check with either an error or a warning:
+
+* Errors: CSS, mpoly, NMF, pathological, rlme
+* Warnings: MazamaSpatialUtils
+
+Two other packages failed with non-installed-related errors (RSiteCatalyst, 
+REDCapR). Judging from the error messages, I don't believe these failures
+at related to stringr.

revdep/check.R

@@ -1,5 +1,7 @@
 library("devtools")
 
+dr_devtools()
+
 install.packages(c("RcppEigen", "RcppArmadillo"), lib = "~/R-revdep")
 install.packages("rgeos", lib = "~/R-revdep", type = "source")
 res <- revdep_check()