The R Project SVN R

% File src/library/base/man/merge.Rd
% Part of the R package, https://www.R-project.org
% Copyright 1995-2018 R Core Team
% Copyright 2002-2016 The R Foundation
% Distributed under GPL 2 or later

\name{merge}
\alias{merge}
\alias{merge.default}
\alias{merge.data.frame}
\concept{join}
\title{Merge Two Data Frames}
\description{
  Merge two data frames by common columns or row names, or do other
  versions of database \emph{join} operations.
}
\usage{
merge(x, y, \dots)

\method{merge}{default}(x, y, \dots)

\method{merge}{data.frame}(x, y, by = intersect(names(x), names(y)),
      by.x = by, by.y = by, all = FALSE, all.x = all, all.y = all,
      sort = TRUE, suffixes = c(".x",".y"), no.dups = TRUE,
      incomparables = NULL, \dots)
}
\arguments{
  \item{x, y}{data frames, or objects to be coerced to one.}
  \item{by, by.x, by.y}{specifications of the columns used for merging.
    See \sQuote{Details}.}
  \item{all}{logical; \code{all = L} is shorthand for \code{all.x = L} and
    \code{all.y = L}, where \code{L} is either \code{\link{TRUE}} or
    \code{FALSE}.}
  \item{all.x}{logical; if \code{TRUE}, then extra rows will be added to
    the output, one for each row in \code{x} that has no matching row in
    \code{y}.  These rows will have \code{NA}s in those columns that are
    usually filled with values from \code{y}.  The default is
    \code{FALSE}, so that only rows with data from both \code{x} and
    \code{y} are included in the output.}
  \item{all.y}{logical; analogous to \code{all.x}.}
  \item{sort}{logical.  Should the result be sorted on the \code{by}
    columns?}
  \item{suffixes}{a character vector of length 2 specifying the suffixes
    to be used for making unique the names of columns in the result
    which are not used for merging (appearing in \code{by} etc).}
  \item{no.dups}{logical indicating that \code{suffixes} are appended in
    more cases to avoid duplicated column names in the result.  This
    was implicitly false before \R version 3.5.0.}
  \item{incomparables}{values which cannot be matched.  See
    \code{\link{match}}.  This is intended to be used for merging on one
    column, so these are incomparable values of that column.}
  \item{\dots}{arguments to be passed to or from methods.}
}
\details{
  \code{merge} is a generic function whose principal method is for data
  frames: the default method coerces its arguments to data frames and
  calls the \code{"data.frame"} method.

  By default the data frames are merged on the columns with names they
  both have, but separate specifications of the columns can be given by
  \code{by.x} and \code{by.y}.  The rows in the two data frames that
  match on the specified columns are extracted, and joined together.  If
  there is more than one match, all possible matches contribute one row
  each.  For the precise meaning of \sQuote{match}, see
  \code{\link{match}}.

  Columns to merge on can be specified by name, number or by a logical
  vector: the name \code{"row.names"} or the number \code{0} specifies
  the row names.  If specified by name it must correspond uniquely to a
  named column in the input.

  If \code{by} or both \code{by.x} and \code{by.y} are of length 0 (a
  length zero vector or \code{NULL}), the result, \code{r}, is the
  \emph{Cartesian product} of \code{x} and \code{y}, i.e.,
  \code{dim(r) = c(nrow(x)*nrow(y), ncol(x) + ncol(y))}.

  If \code{all.x} is true, all the non matching cases of \code{x} are
  appended to the result as well, with \code{NA} filled in the
  corresponding columns of \code{y};  analogously for \code{all.y}.

  If the columns in the data frames not used in merging have any common
  names, these have \code{suffixes} (\code{".x"} and \code{".y"} by
  default) appended to try to make the names of the result unique.  If
  this is not possible, an error is thrown.

  If a \code{by.x} column name matches one of \code{y}, and if
  \code{no.dups} is true (as by default), the y version gets suffixed as
  well, avoiding duplicate column names in the result.

  The complexity of the algorithm used is proportional to the length of
  the answer.

  % Terminology follows https://en.wikipedia.org/wiki/Join_(SQL)
  In SQL database terminology, the default value of \code{all = FALSE}
  gives a \emph{natural join}, a special case of an \emph{inner
  join}. Specifying \code{all.x = TRUE} gives a \emph{left (outer)
  join}, \code{all.y = TRUE} a \emph{right (outer) join}, and both
  (\code{all = TRUE}) a \emph{(full) outer join}.  \abbr{DBMS}es do not match
  \code{NULL} records, equivalent to \code{incomparables = NA} in \R.
}

\note{
  This is intended to work with data frames with vector-like columns:
  some aspects work with data frames containing matrices, but not all.

  Currently long vectors are not accepted for inputs, which are thus
  restricted to less than 2^31 rows. That restriction also applies to
  the result for 32-bit platforms.
}

\value{
  A data frame.  The rows are by default lexicographically sorted on the
  common columns, but for \code{sort = FALSE} are in an unspecified order.
  The columns are the common columns followed by the
  remaining columns in \code{x} and then those in \code{y}.  If the
  matching involved row names, an extra character column called
  \code{Row.names} is added at the left, and in all cases the result has
  \sQuote{automatic} row names.
}
\seealso{
  \code{\link{data.frame}},
  \code{\link{by}},
  \code{\link{cbind}}.

  \code{\link{dendrogram}} for a class which has a \code{merge} method.
}

\examples{
authors <- data.frame(
    surname = c("Tukey", "Venables", "Tierney", "Ripley", "McNeil"),
    nationality = c("US", "Australia", "US", "UK", "Australia"),
    deceased = c("yes", rep("no", 4)))
authorN <- within(authors, { name <- surname; rm(surname) })
books <- data.frame(
    name = c("Tukey", "Venables", "Tierney",
             "Ripley", "Ripley", "McNeil", "R Core"),
    title = c("Exploratory Data Analysis",
              "Modern Applied Statistics ...",
              "LISP-STAT",
              "Spatial Statistics", "Stochastic Simulation",
              "Interactive Data Analysis",
              "An Introduction to R"),
    other.author = c(NA, "Ripley", NA, NA, NA, NA,
                     "Venables & Smith"))

(m0 <- merge(authorN, books))
(m1 <- merge(authors, books, by.x = "surname", by.y = "name"))
 m2 <- merge(books, authors, by.x = "name", by.y = "surname")
stopifnot(exprs = {
   identical(m0, m2[, names(m0)])
   as.character(m1[, 1]) == as.character(m2[, 1])
   all.equal(m1[, -1], m2[, -1][ names(m1)[-1] ])
   identical(dim(merge(m1, m2, by = NULL)),
             c(nrow(m1)*nrow(m2), ncol(m1)+ncol(m2)))
})

## "R core" is missing from authors and appears only here :
merge(authors, books, by.x = "surname", by.y = "name", all = TRUE)


## example of using 'incomparables'
x <- data.frame(k1 = c(NA,NA,3,4,5), k2 = c(1,NA,NA,4,5), data = 1:5)
y <- data.frame(k1 = c(NA,2,NA,4,5), k2 = c(NA,NA,3,4,5), data = 1:5)
merge(x, y, by = c("k1","k2")) # NAs match
merge(x, y, by = "k1") # NAs match, so 6 rows
merge(x, y, by = "k2", incomparables = NA) # 2 rows
}
\keyword{array}
\keyword{manip}