The function dataTableProxy()
creates a proxy object that can be used
to manipulate an existing DataTables instance in a Shiny app, e.g. select
rows/columns, or add rows.
dataTableProxy(
outputId,
session = shiny::getDefaultReactiveDomain(),
deferUntilFlush = TRUE
)selectRows(proxy, selected, ignore.selectable = FALSE)
selectColumns(proxy, selected, ignore.selectable = FALSE)
selectCells(proxy, selected, ignore.selectable = FALSE)
addRow(proxy, data, resetPaging = TRUE)
clearSearch(proxy)
selectPage(proxy, page)
updateCaption(proxy, caption)
updateSearch(proxy, keywords = list(global = NULL, columns = NULL))
showCols(proxy, show, reset = FALSE)
hideCols(proxy, hide, reset = FALSE)
colReorder(proxy, order, origOrder = FALSE)
reloadData(
proxy,
resetPaging = TRUE,
clearSelection = c("all", "none", "row", "column", "cell")
)
the id of the table to be manipulated (the same id as the one
you used in dataTableOutput()
)
the Shiny session object (from the server function of the Shiny app)
whether an action should be carried out right away, or should be held until after the next time all of the outputs are updated
a proxy object returned by dataTableProxy()
an integer vector of row/column indices, or a matrix of two
columns (row and column indices, respectively) for cell indices; you may
use NULL
to clear existing selections
when FALSE
(the default), the "non-selectable"
range specified by selection = list(selectable= )
is respected, i.e.,
you can't select "non-selectable" range. Otherwise, it is ignored.
a single row of data to be added to the table; it can be a matrix
or data frame of one row, or a vector or list of row data (in the latter
case, please be cautious about the row name: if your table contains row
names, here data
must also contain the row name as the first
element)
whether to reset the paging position
a number indicating the page to select
a new table caption (see the caption
argument of
datatable()
)
a list of two components: global
is the global search
keyword of a single character string (ignored if NULL
);
columns
is a character vector of the search keywords for all columns
(when the table has one column for the row names, this vector of keywords
should contain one keyword for the row names as well)
a vector of column positions to show (the indexing starts at 0, but if row.names are visible, they are the first column).
if TRUE
, will only show/hide the columns indicated.
a vector of column positions to hide
A numeric vector of column positions, starting from 0, and including the row.names as a column, if they are include. Must contain a value for all columns, regardless of whether they are visible or not. Also for column reordering to work, the datatable must have extension 'ColReorder' set as well as option 'colReordoer' set to TRUE).
Whether column reordering should be relative to the original order (the default is to compare to current order)
which existing selections to clear: it can be any
combinations of row
, column
, and cell
, or all
for all three, or none
to keep current selections (by default, all
selections are cleared after the data is reloaded)