Create a table


Creates a Pandoc's markdown style table with optional caption and some other tweaks. See 'Details' below.


library("pander") pandoc.table.return(t, caption, digits = panderOptions("digits"), decimal.mark = panderOptions("decimal.mark"), big.mark = panderOptions("big.mark"), round = panderOptions("round"), missing = panderOptions("missing"), justify, style = c("multiline", "grid", "simple", "rmarkdown"), split.tables = panderOptions("table.split.table"), split.cells = panderOptions("table.split.cells"), keep.trailing.zeros = panderOptions("keep.trailing.zeros"), keep.line.breaks = panderOptions("keep.line.breaks"), plain.ascii = panderOptions("plain.ascii"), use.hyphening = panderOptions("use.hyphening"), emphasize.rownames = panderOptions("table.emphasize.rownames"), emphasize.rows, emphasize.cols, emphasize.cells, emphasize.strong.rows, emphasize.strong.cols, emphasize.strong.cells, emphasize.italics.rows, emphasize.italics.cols, emphasize.italics.cells, emphasize.verbatim.rows, emphasize.verbatim.cols, emphasize.verbatim.cells, ...)



data frame, matrix or table


caption (string) to be shown under the table


passed to format. Can be a vector specifying values for each column (has to be the same length as number of columns).


passed to format


passed to format


passed to round. Can be a vector specifying values for each column (has to be the same length as number of columns). Values for non-numeric columns will be disregarded.


string to replace missing values


defines alignment in cells passed to format. Can be left, right or centre, which latter can be also spelled as center. Defaults to centre. Can be abbreviated to a string consisting of the letters l, c and r (e.g. 'lcr' instead of c('left', 'centre', 'right').


which Pandoc style to use: simple, multiline, grid or rmarkdown


where to split wide tables to separate tables. The default value (80) suggests the conventional number of characters used in a line, feel free to change (e.g. to Inf to disable this feature) if you are not using a VT100 terminal any more :)


where to split cells' text with line breaks. Default to 30, to disable set to Inf. Can be also supplied as a vector, for each cell separately (if length(split.cells) == number of columns + 1, then first value in split.cells if for row names, and others are for columns). Supports relative (percentage) parameters in combination with split.tables.


to show or remove trailing zeros in numbers on a column basis width


(default: FALSE) if to keep or remove line breaks from cells in a table


(default: FALSE) if output should be in plain ascii (without markdown markup) or not


boolean (default: FALSE) if try to use hyphening when splitting large cells according to table.split.cells. Requires koRpus package.


boolean (default: TRUE) if row names should be highlighted


deprecated for emphasize.italics.rows argument


deprecated for emphasize.italics.cols argument


deprecated for emphasize.italics.cells argument


see emphasize.italics.rows but in bold


see emphasize.italics.cols but in bold


see emphasize.italics.cells but in bold


a vector for a two dimensional table specifying which rows to emphasize


a vector for a two dimensional table specifying which cols to emphasize


a vector for one-dimensional tables or a matrix like structure with two columns for row and column indexes to be emphasized in two dimensional tables. See e.g. which(..., arr.ind = TRUE)


see emphasize.italics.rows but in verbatim


see emphasize.italics.cols but in verbatim


see emphasize.italics.cells but in verbatim


unsupported extra arguments directly placed into /dev/null


By default this function outputs (see: cat) the result. If you would want to catch the result instead, then call pandoc.table.return instead.

See also

set.caption, set.alignment


This function takes any tabular data as its first argument and will try to make it pretty like: rounding and applying digits and custom decimal.mark to numbers, auto-recognizing if row names should be included, setting alignment of cells and dropping trailing zeros by default.

pandoc.table also tries to split large cells with line breaks or even the whole table to separate parts on demand. Other arguments lets the use to highlight some rows/cells/cells in the table with italic or bold text style.

For more details please see the parameters above and passed arguments of panderOptions.


If caption is missing, then the value is first checked in t object's caption attribute and if not found in an internal buffer set by link{set.caption}. justify parameter works similarly, see set.alignment for details.


library("pander") pandoc.table(mtcars) ## caption pandoc.table(mtcars, 'Motor Trend Car Road Tests') ## other input/output formats pandoc.table(mtcars[, 1:3], decimal.mark = ',') pandoc.table(mtcars[, 1:3], decimal.mark = ',', justify = 'right') pandoc.table(matrix(sample(1:1000, 25), 5, 5)) pandoc.table(matrix(runif(25), 5, 5)) pandoc.table(matrix(runif(25), 5, 5), digits = 5) pandoc.table(matrix(runif(25),5,5), round = 1) pandoc.table(table(mtcars$am)) pandoc.table(table(mtcars$am, mtcars$gear)) pandoc.table(table(state.division, state.region)) pandoc.table(table(state.division, state.region), justify = 'centre') m <- data.frame(a = c(1, -500, 10320, 23, 77), b = runif(5), c = c('a', 'bb', 'ccc', 'dddd', 'eeeee')) pandoc.table(m) pandoc.table(m, justify = c('right', 'left', 'centre')) pandoc.table(m, justify = 'rlc') # Same as upper statement ## splitting up too wide tables pandoc.table(mtcars) pandoc.table(mtcars, caption = 'Only once after the first part!') ## tables with line breaks in cells ## NOTE: line breaks are removed from table content in case keep.line.breaks is set to FALSE ## and added automatically based on "split.cells" parameter! t <- data.frame(a = c('hundreds\nof\nmouses', '3 cats'), b=c('FOO is nice', 'BAR\nBAR2')) pandoc.table(t) pandoc.table(t, split.cells = 5) ## exporting tables in other Pandoc styles pandoc.table(m) pandoc.table(m, style = "grid") pandoc.table(m, style = "simple") pandoc.table(t, style = "grid") pandoc.table(t, style = "grid", split.cells = 5) tryCatch(pandoc.table(t, style = "simple", split.cells = 5), error = function(e) 'Yeah, no newline support in simple tables') ## highlight cells t <- mtcars[1:3, 1:5] pandoc.table(t$mpg, emphasize.italics.cells = 1) pandoc.table(t$mpg, emphasize.strong.cells = 1) pandoc.table(t$mpg, emphasize.italics.cells = 1, emphasize.strong.cells = 1) pandoc.table(t$mpg, emphasize.italics.cells = 1:2) pandoc.table(t$mpg, emphasize.strong.cells = 1:2) pandoc.table(t, emphasize.italics.cells = which(t > 20, arr.ind = TRUE)) pandoc.table(t, emphasize.italics.cells = which(t == 6, arr.ind = TRUE)) pandoc.table(t, emphasize.verbatim.cells = which(t == 6, arr.ind = TRUE)) pandoc.table(t, emphasize.verbatim.cells = which(t == 6, arr.ind = TRUE), emphasize.italics.rows = 1) ## with helpers emphasize.cols(1) emphasize.rows(1) pandoc.table(t) emphasize.strong.cells(which(t > 20, arr.ind = TRUE)) pandoc.table(t) ### plain.ascii pandoc.table(mtcars[1:3, 1:3], plain.ascii = TRUE) ### keep.line.breaks x <- data.frame(a="Pandoc\nPackage") pandoc.table(x) pandoc.table(x, keep.line.breaks = TRUE) ## split.cells x <- data.frame(a = "foo bar", b = "foo bar") pandoc.table(x, split.cells = 4) pandoc.table(x, split.cells = 7) pandoc.table(x, split.cells = c(4, 7)) pandoc.table(x, split.cells = c("20%", "80%"), split.tables = 30) y <- c("aa aa aa", "aaa aaa", "a a a a a", "aaaaa", "bbbb bbbb bbbb", "bb bbb bbbb") y <- matrix(y, ncol = 3, nrow = 2) rownames(y) <- c("rowname one", "rowname two") colnames(y) <- c("colname one", "colname two", "colname three") pandoc.table(y, split.cells = 2) pandoc.table(y, split.cells = 6) pandoc.table(y, split.cells = c(2, 6, 10)) pandoc.table(y, split.cells = c(2, Inf, Inf)) ## first value used for rownames pander(y, split.cells = c(5, 2, Inf, Inf)) pandoc.table(y, split.cells = c(5, 2, Inf, 5, 3, 10)) ## when not enough reverting to default values pandoc.table(y, split.cells = c(5, 2)) ## split.cells with hyphenation x <- data.frame(a = "Can be also supplied as a vector, for each cell separately", b = "Can be also supplied as a vector, for each cell separately") pandoc.table(x, split.cells = 10, use.hyphening = TRUE)


John MacFarlane (2012): _Pandoc User's Guide_.

Documentation reproduced from package pander, version 0.6.0, License: AGPL-3 | file LICENSE