matsbyname (version 0.6.14)

rename_via_pattern_byname: Rename row or column names via regexp pattern

Description

It is sometimes helpful to rename row or column names for a list of matrices via a regexp_pattern. When a is a matrix or a list of matrices, regexp_pattern indicates which characters are replaced by replacement.

Usage

rename_via_pattern_byname(
  a,
  margin = list(c(1, 2)),
  regexp_pattern = "$^",
  replacement,
  pieces = "all",
  prepositions = RCLabels::prepositions_list,
  notation = RCLabels::bracket_notation,
  ...
)

Value

A modified version of a.

Arguments

a

A matrix or list of matrices.

margin

The margin on which replacements are performed. Default is c(1, 2), meaning both row (1) and column (2) names will be replaced.

regexp_pattern

The regular expression pattern that will be replaced in the row or column labels. Default is "$^", meaning nothing will be matched.

replacement

The string to replace the regexp_pattern.

pieces

The pieces of labels to be searched for regexp_pattern. See RCLabels::replace_by_pattern for details. Default is "all".

prepositions

Prepositions to use while searching for pieces. Default is RCLabels::prepositions_list.

notation

The notation used for for searching pieces. Default is RCLabels::bracket_notation.

...

Other arguments passed to gsub(), such as ignore.case, perl, fixed, or useBytes. Arguments in ... apply to all matrices in a. See examples.

Details

Note that margin can be a rowtype or coltype string which will be de-referenced to the integer margin (1 for rows or 2 for columns).

Internally, this function calls RCLabels::replace_by_pattern.

Examples

ma <- matrix(c(1, 2), nrow = 2,
             dimnames = list(c("Natural gas [from Supply]",
                               "row2"), 
                             "col")) |> 
  setrowtype("Product") |> setcoltype("Industry")
mb <- matrix(c(1, 2), nrow = 2,
             dimnames = list(c("Natural gas [from Supply]",
                               "Fuel oil [from Supply]"), 
                             "col")) |> 
  setrowtype("Product") |> setcoltype("Industry")
ma |> 
  rename_via_pattern_byname(regexp_pattern = " [from Supply]",
                            replacement = " bogus", 
                            fixed = TRUE)
list(ma, mb) |> 
  rename_via_pattern_byname(margin = 1,
                            regexp_pattern = " [from Supply]",
                            replacement = " from Supply", 
                            fixed = TRUE)
res1 <- tibble::tibble(m = list(ma, mb)) |> 
  dplyr::mutate(
    m1 = .data[["m"]] |> 
      rename_via_pattern_byname(regexp_pattern = " [from Supply]", 
                                replacement = "", 
                                fixed = TRUE)
  )
res1$m1
# Transpose mb and use a string for the margin.
# The string (in this case "Product")
# is dereferenced to an integer margin.
# In this case, the rownames of the first matrix
# and the colnames of the second matrix are replaced,
# because those are on the "Product" margin.
res2 <- tibble::tibble(m = list(ma, 
                                transpose_byname(mb))) |> 
  dplyr::mutate(
    m2 = .data[["m"]] |> 
      rename_via_pattern_byname(margin = "Product", 
                                regexp_pattern = " [from Supply]", 
                                replacement = "", 
                                fixed = TRUE)
  )
rowtype(res2$m2[[1]])
coltype(res2$m2[[2]])
res2$m2