`xl_filter_keep()` answers, for a data frame and a set of [xl_filter()] rules, the question `xl_sheet(filter =)` has to answer internally: which rows does Excel leave visible? It writes nothing — it is the matching rule on its own, exported because reproducing Excel's filter semantics is hard to get right and useful outside writing a file.
The rules are described in detail under [xl_filter()], and were established by measurement rather than from documentation: a workbook was written with criteria set and no rows hidden, opened in Excel, and Data > Reapply pressed so that Excel computed each match itself.
Accuracy
This function tracks Excel's *observed* behaviour, so a case found to disagree with Excel is treated as a bug and fixed, which may change the rows it returns. One limitation is known: a filter written as a value list (`"=="` without a wildcard, or `list`) matches the text a cell **displays**, which writexl can only predict for the General format. A numeric column carrying a custom number format may therefore display differently from what is compared here — prefer a comparison such as `">="` on such a column. Dates are refused outright for the same reason. See [xl_filter()].
See also
[xl_filter], [xl_sheet]
Other worksheet features:
xl_conditional,
xl_filter(),
xl_merge(),
xl_table(),
xl_table_column(),
xl_validation()
Examples
sales <- data.frame(fruit = c("apple", "banana", "cherry"),
qty = c(5, 150, 300))
xl_filter_keep(sales, xl_filter("qty", ">", 100))
#> [1] FALSE TRUE TRUE
sales[xl_filter_keep(sales, xl_filter("fruit", "==", "*a*")), ]
#> fruit qty
#> 1 apple 5
#> 2 banana 150
# filters on different columns combine with AND
xl_filter_keep(sales, list(xl_filter("qty", ">", 100),
xl_filter("fruit", "==", "b*")))
#> [1] FALSE TRUE FALSE