ffbase (version 0.12.7)

merge.ffdf: Merge two ffdf by common columns, or do other versions of database join operations.

Description

Merge two ffdf by common columns, or do other versions of database join operations. This method is similar to merge in the base package but only allows inner and left outer joins. Note that joining is done based on ffmatch or ffdfmatch: only the first element in y will be added to x; and since ffdfmatch works by paste-ing together a key, this might not be suited if your key contains columns of vmode double.

Usage

# S3 method for ffdf
merge(x, y, by = intersect(names(x), names(y)),
  by.x = by, by.y = by, all = FALSE, all.x = all, all.y = all,
  sort = FALSE, suffixes = c(".x", ".y"), incomparables = NULL,
  trace = FALSE, ...)

Arguments

x

an ffdf

y

an ffdf

by

specifications of the common columns. Columns can be specified by name, number or by a logical vector.

by.x

specifications of the common columns of the x ffdf, overruling the by parameter

by.y

specifications of the common columns of the y ffdf, overruling the by parameter

all

see merge in R base

all.x

if TRUE, then extra rows will be added to the output, one for each row in x that has no matching row in y. These rows will have NAs in those columns that are usually filled with values from y. The default is FALSE, so that only rows with data from both x and y are included in the output.

all.y

similar as all.x

sort

logical, currently not used yet, defaults to FALSE.

suffixes

character(2) specifying the suffixes to be used for making non-by names() unique.

incomparables

values which cannot be matched. See match. Currently not used.

trace

logical indicating to show on which chunk the function is computing

...

other options passed on to ffdfindexget

Value

an ffdf

Details

If a left outer join is performed and no matching record in x is found in y, columns with vmodes 'boolean', 'quad', 'nibble', 'ubyte', 'ushort' are coerced respectively to vmode 'logical', 'byte', 'byte', 'short', 'integer' to allow NA values.

See Also

merge

Examples

# NOT RUN {
authors <- data.frame(
    surname = c("Tukey", "Venables", "Tierney", "Ripley", "McNeil"),
    nationality = c("US", "Australia", "US", "UK", "Australia"),
    deceased = c("yes", rep("no", 4)))
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"))
books <- lapply(1:100, FUN=function(x, books){
	books$price <- rnorm(nrow(books))
	books
}, books=books)
books <- do.call(rbind, books)
authors <- as.ffdf(authors)                
books <- as.ffdf(books)

dim(books)
dim(authors)
## Inner join
oldffbatchbytes <- getOption("ffbatchbytes")
options(ffbatchbytes = 100)
m1 <- merge( books, authors, by.x = "name", by.y = "surname"
           , all.x=FALSE, all.y=FALSE, trace = TRUE)
dim(m1)
unique(paste(m1$name[], m1$nationality[]))
unique(paste(m1$name[], m1$deceased[]))
m2 <- merge( books[,], authors[,], by.x = "name", by.y = "surname"
           , all.x=FALSE, all.y=FALSE, sort = FALSE)
dim(m2)
unique(paste(m2$name[], m2$nationality[]))
unique(paste(m2$name[], m2$deceased[]))
## Left outer join
m1 <- merge( books, authors, by.x = "name", by.y = "surname"
           , all.x=TRUE, all.y=FALSE, trace = TRUE)
class(m1)
dim(m1)
names(books)
names(m1)
unique(paste(m1$name[], m1$nationality[]))
unique(paste(m1$name[], m1$deceased[]))

authors$test <- ff(TRUE, length=nrow(authors), vmode = "logical")
m1 <- merge( books, authors, by.x = "name", by.y = "surname"
           , all.x=TRUE, all.y=FALSE, trace = TRUE)
vmode(m1$test)
table(m1$test[], exclude=c())
options(ffbatchbytes = oldffbatchbytes)
# }