CRAN release (#3060)

* remove old revdep reslts

* ignore

* revdep results, CC @hadley

* checks with #3055

* rerun, no pRoloc problems anymore

* add necessary file

* CRAN notes

* ignore local issues dir

* bump version

* fix typos

* make test much more unlikely to fail

* add protection to hybrid function

to (hopefully) work around ASAN errors that show on CRAN on Fedora (but not locally on Debian+clang+ASAN)

* fix typo

* fix typo

* CRAN comments

* NEWS

* bump

* bump

* try to fix CRAN failures

* up comments

* try to fix CRAN failures, again

.Rbuildignore

@@ -27,3 +27,4 @@ demo/pandas
 ^vignettes/rsconnect$
 ^docs$
 ^_pkgdown\.yml$
+^issues$

.gitignore

@@ -17,3 +17,4 @@ vignettes/*.R
 /GTAGS
 /TAGS
 /.dir-locals.el
+/issues

DESCRIPTION

@@ -1,6 +1,6 @@
 Type: Package
 Package: dplyr
-Version: 0.7.2.9000
+Version: 0.7.3
 Title: A Grammar of Data Manipulation
 Description: A fast, consistent tool for working with data frame like objects,
     both in memory and out of memory.

NEWS.md

@@ -1,21 +1,20 @@
-# dplyr 0.7.2.9000
+# dplyr 0.7.3
 
-* Grouping by character vectors is now faster (#2204).
-
-* Fix protection error that occurred when creating a character column
-  using grouped `mutate()` (#2971).
+* Fixed protection error that occurred when creating a character column using grouped `mutate()` (#2971).
 
-* Semi- and anti-joins now preserve order of left-hand-side data frame (#3089).
+* Fixed a rare problem with accessing variable values in `summarise()` when all groups have size one (#3050).
 
-* `select()`, `rename()` and `summarise()` no longer changes the grouped vars of the original data (#3038).
+* `select()`, `rename()` and `summarise()` no longer change the grouped vars of the original data (#3038).
 
 * `nth(default = var)`, `first(default = var)` and `last(default = var)` fall back to standard evaluation in a grouped operation instead of triggering an error (#3045).
 
 * `case_when()` now works if all LHS are atomic (#2909), or when LHS or RHS values are zero-length vectors (#3048).
 
 * `case_when()` accepts `NA` on the LHS (#2927).
 
-* Fixed a rare problem with accessing variable values in `summarise()` when all groups have size one (#3050).
+* Semi- and anti-joins now preserve the order of left-hand-side data frame (#3089).
+
+* Grouping by character vectors is now faster (#2204).
 
 # dplyr 0.7.2
 

R/compat-dbplyr.R

@@ -1,7 +1,7 @@
-#' dbplyr compatiblity functions
+#' dbplyr compatibility functions
 #'
 #' @description
-#' In dplyr 0.6.0, a number of databse and SQL functions moved from dplyr to
+#' In dplyr 0.6.0, a number of database and SQL functions moved from dplyr to
 #' dbplyr. The generic functions stayed in dplyr (since there is no easy way
 #' to conditionally import a generic from different packages), but many other
 #' SQL and database helper functions moved. If you have written a backend,

R/copy-to.r

@@ -27,7 +27,7 @@ copy_to <- function(dest, df, name = deparse(substitute(df)),
 
 #' Copy tables to same source, if necessary
 #'
-#' @param x,y `y` will be copied to `x`, if neccessary.
+#' @param x,y `y` will be copied to `x`, if necessary.
 #' @param copy If `x` and `y` are not from the same data source,
 #'   and `copy` is `TRUE`, then `y` will be copied into the
 #'   same src as `x`.  This allows you to join tables across srcs, but

R/explain.r

@@ -1,7 +1,7 @@
 #' Explain details of a tbl
 #'
 #' This is a generic function which gives more details about an object than
-#' [print()], and is more focussed on human readable output than
+#' [print()], and is more focused on human readable output than
 #' [str()].
 #'
 #' @section Databases:

R/id.r

@@ -4,11 +4,11 @@
 #' \itemize{
 #'   \item `order(id)` is equivalent to `do.call(order, df)`
 #'   \item rows containing the same data have the same value
-#'   \item if `drop = FALSE` then room for all possibilites
+#'   \item if `drop = FALSE` then room for all possibilities
 #' }
 #'
 #' @param .variables list of variables
-#' @param drop drop unusued factor levels?
+#' @param drop drop unused factor levels?
 #' @return a numeric vector with attribute n, giving total number of
 #'   possibilities
 #' @keywords internal

R/join.r

@@ -58,7 +58,7 @@
 #'   same src as `x`.  This allows you to join tables across srcs, but
 #'   it is a potentially expensive operation so you must opt into it.
 #' @param suffix If there are non-joined duplicate variables in `x` and
-#'   `y`, these suffixes will be added to the output to diambiguate them.
+#'   `y`, these suffixes will be added to the output to disambiguate them.
 #'   Should be a character vector of length 2.
 #' @param ... other parameters passed onto methods
 #' @name join

R/lead-lag.R

@@ -4,9 +4,9 @@
 #' ahead of or behind the current values.
 #'
 #' @param x a vector of values
-#' @param n a postive integer of length 1, giving the number of positions to
+#' @param n a positive integer of length 1, giving the number of positions to
 #'   lead or lag by
-#' @param default value used for non-existant rows. Defaults to `NA`.
+#' @param default value used for non-existent rows. Defaults to `NA`.
 #' @param order_by override the default ordering to use another vector
 #' @param ... Needed for compatibility with lag generic.
 #' @importFrom stats lag

R/manip.r

@@ -28,7 +28,7 @@
 #'   for [tbl_df()], [dtplyr::tbl_dt()] and [dbplyr::tbl_dbi()].
 #' @param ... Logical predicates defined in terms of the variables in `.data`.
 #'   Multiple conditions are combined with `&`. Only rows where the
-#'   conditon evalutes to `TRUE` are kept.
+#'   condition evaluates to `TRUE` are kept.
 #'
 #'   These arguments are automatically [quoted][rlang::quo] and
 #'   [evaluated][rlang::eval_tidy] in the context of the data
@@ -211,7 +211,7 @@ summarize_ <- summarise_
 #'
 #' * [if_else()], [recode()], [case_when()]
 #'
-#' @section Scoped mutation and transmuation:
+#' @section Scoped mutation and transmutation:
 #'
 #' The three [scoped] variants of `mutate()` ([mutate_all()],
 #' [mutate_if()] and [mutate_at()]) and the three variants of

R/src_dbi.R

@@ -21,7 +21,7 @@
 #' that you are grouping by. Use [explain()] to check that the database is using
 #' the indexes that you expect.
 #'
-#' There is one excpetion: [do()] is not lazy since it must pull the data
+#' There is one exception: [do()] is not lazy since it must pull the data
 #' into R.
 #'
 #' @param dbname Database name
@@ -136,7 +136,7 @@ src_postgres <- function(dbname = NULL, host = NULL, port = NULL,
 #' @param path Path to SQLite database. You can use the special path
 #'   ":memory:" to create a temporary in memory database.
 #' @param create if `FALSE`, `path` must already exist. If
-#'   `TRUE`, will create a new SQlite3 database at `path` if
+#'   `TRUE`, will create a new SQLite3 database at `path` if
 #'   `path` does not exist and connect to the existing database if
 #'   `path` does exist.
 src_sqlite <- function(path, create = FALSE) {

R/tbl-cube.r

@@ -39,7 +39,7 @@
 #'
 #' @export
 #' @param dimensions A named list of vectors. A dimension is a variable
-#'   whose values are known before the experiement is conducted; they are
+#'   whose values are known before the experiment is conducted; they are
 #'   fixed by design (in \pkg{reshape2} they are known as id variables).
 #'   `tbl_cubes` are dense which means that almost every combination of
 #'   the dimensions should have associated measurements: missing values require
@@ -203,7 +203,7 @@ as.tbl_cube <- function(x, ...) UseMethod("as.tbl_cube")
 
 #' @export
 #' @rdname as.tbl_cube
-#' @param dim_names names of the dimesions. Defaults to the names of
+#' @param dim_names names of the dimensions. Defaults to the names of
 #' @param met_name a string to use as the name for the measure
 #'   the [dimnames()].
 as.tbl_cube.array <- function(x, dim_names = names(dimnames(x)), met_name = deparse(substitute(x)),

cran-comments.md

@@ -1,11 +1,11 @@
 ## Release summary
 
-This submission moves run-time vs build-time checks out of .onLoad and into a new function that can be run on demand (`dr_dplyr()`)
+This is a minor bugfix release in response to CRAN's request. This is the second release attempt (with the same version), following up test failures with reverse dependencies.
 
 ## Test environments
 
-* local OS X install, R 3.4.0
-* ubuntu 12.04 (on travis-ci), R 3.4.0, R 3.2.5, R-devel.
+* local OS X install, R 3.4.1
+* ubuntu 12.04 (on travis-ci), R 3.4.1, R-oldrel, R-devel.
 * win-builder (devel)
 
 ## R CMD check results
@@ -16,6 +16,20 @@ This submission moves run-time vs build-time checks out of .onLoad and into a ne
 
   This is all compiled code in the libs/ directory.
 
+## ASAN errors on CRAN
+
+We could not replicate the ASAN errors on Fedora with clang.  We changed code that we believe may be responsible for the problem, and we'll watch the check results once the package is released.
+
 ## Reverse dependencies
 
-As no changes affected behaviour I did not re-run the revdep checks.
+We checked 779 reverse dependencies (702 from CRAN + 77 from BioConductor), comparing R CMD check results across CRAN and dev versions of this package.
+
+ * We saw 0 new problems
+ * We failed to check 2 packages
+
+Issues with CRAN packages are summarised below.
+
+### Failed to check
+
+* loon        (failed to install)
+* MonetDBLite (check timed out)

inst/include/dplyr/Result/Rank.h

@@ -226,7 +226,7 @@ class Rank_Impl : public Result, public Increment {
   }
 
 
-  SEXP data;
+  Vector<RTYPE> data;
   Map map;
 };
 
@@ -259,7 +259,8 @@ class RowNumber : public Result {
       Slice slice(data, index);
       // order( gdf.group(i) )
       Visitor visitor(slice);
-      std::sort(tmp.begin(), tmp.begin() + m, Comparer(visitor));
+      Comparer comparer = Comparer(visitor);
+      std::sort(tmp.begin(), tmp.begin() + m, comparer);
       int j = m - 1;
       for (; j >= 0; j--) {
         if (Rcpp::traits::is_na<RTYPE>(slice[ tmp[j] ])) {
@@ -340,7 +341,8 @@ class Ntile : public Result {
 
       // order( gdf.group(i) )
       Visitor visitor(slice);
-      std::sort(tmp.begin(), tmp.begin() + m, Comparer(visitor));
+      Comparer comparer = Comparer(visitor);
+      std::sort(tmp.begin(), tmp.begin() + m, comparer);
       int j = m - 1;
       for (; j >= 0; j--) {
         if (Rcpp::traits::is_na<RTYPE>(slice[tmp[j]])) {

revdep/.gitignore

@@ -0,0 +1,3 @@
+checks
+library
+data.sqlite