将 R 数据框导出到 Excel 命名范围

Question

我正在将来自不同来源的数据整理到多个数据框中。每个数据帧都有维度 R_k X C_k。然后，我想将每个数据帧（由唯一标签（如 Object1_Time1）标识）导出到 Excel 中相应的命名范围（对应于唯一标签，即 Object1_Time1）。使用 R 可以实现类似的功能吗？

我研究了 openxlsx 和 xlsx，它们似乎都不支持本机。使用 R 以某种方式隔离 Excel 命名范围的单元格范围，然后在导出数据帧时将它们用作开始和结束列，这可能是一个想法吗？

我在下面提供了一个带有数据的可重现示例 - 我只是还没有成功。

范围值的计算和存储如下：

# Load the Excel workbook
wb <- loadWorkbook("...\\Test_Excel.xlsx")

# Get information about all named ranges
range_info <- getNamedRegions(wb)

# Extract position attribute as a vector
positions <- attr(range_info, "position")

# Drop values after (inclusive) the ":" character
positions_cleaned <- sub(":.*", "", positions)

col_start <- gsub("\\d", "", positions_cleaned)
row_start <- as.numeric(gsub("\\D", "", positions_cleaned))

# Function to convert Excel column letters to numeric values
excel_col_to_numeric <- function(col) {
  col = toupper(col)
  result = 0
  
  for (i in 1:nchar(col)) {
    result = result * 26 + (as.numeric(charToRaw(substr(col, i, i))) - as.numeric(charToRaw("A")) + 1)
  }
  
  return(result)
}


# Convert to numeric values
column_start_numeric <- lapply(col_start, excel_col_to_numeric)

代码主体如下：

library(writexl)
library(openxlsx)
library(dplyr)
library(purrr)
library(MASS)
library(tidyr)

# Create some sample dataframes
set.seed(123)

# Number of time periods and countries
num_time_periods <- 10
num_countries <- 10

# Generate panel data for the first data set
panel_data_1 <- expand.grid(TimePeriod = 1:num_time_periods, Country = 1:num_countries) %>%
  mutate(Value = rnorm(n()))

# Reshape the data to wide format for the first set
panel_data_wide_1 <- spread(panel_data_1, key = TimePeriod, value = Value)

# Generate panel data for the second data set
panel_data_2 <- expand.grid(TimePeriod = 1:num_time_periods, Country = 1:num_countries) %>%
  mutate(Value = rnorm(n()))

# Reshape the data to wide format for the second set
panel_data_wide_2 <- spread(panel_data_2, key = TimePeriod, value = Value)

# Store dataframes in a list with names based on named ranges
df_list <- list(NamedRange1 = panel_data_wide_1, NamedRange2 = panel_data_wide_2)

# Export each dataframe to an EXISTING named range in Excel
write_to_excel <- function(df, sheet_name = "Panel1", range_name) {
  wb <- loadWorkbook("..\\Test_Excel.xlsx")
  # Check if the sheet exists before writing data
  if (sheet_name %in% getSheetNames(wb)) {
    writeData(wb, sheet = sheet_name, x = df, startCol = 1, startRow = 1)
    defineName(wb, name = range_name, formula = sheet_name)
    saveWorkbook(wb, "..\\Test_Excel.xlsx", overwrite = TRUE)
  } else {
    warning(paste("Sheet", sheet_name, "does not exist. Data not exported."))
  }
}

# Iterate over the list and export each dataframe
walk(df_list, ~write_to_excel(.x, range_name = .y))

这是我想要粘贴到的 Excel 工作表示例，其中命名范围延伸到绿色阴影区域。

Excel表格

Answer 1

使用

openxlsx

，我确定了以下内容：

# define function to extract named range sheet/row/col
fn_namedRng <- function(wb,rname,part) {

  if(!part %in% c("sh","col","row","xy")) {
    stop("Argument 'part' must be one of: 'sh','col','row' or 'xy'.")
  }
          
  i = which(c(getNamedRegions(wb)==rname))
  sheet = c(attr(getNamedRegions(wb), "sheet"))[i]
  rng = c(attr(getNamedRegions(wb), "position"))[i]
          
  col1 = col2int(gsub("[0-9]","",gsub(":.*$", "", rng)))
  col2 = col2int(gsub("[0-9]","",gsub( ".*:", "", rng)))
  row1 = as.integer(gsub("[A-Z]","",gsub(":.*$", "", rng)))
  row2 = as.integer(gsub("[A-Z]","",gsub( ".*:", "", rng)))
        
  if(part == "sh"){
    result = sheet
  } else if (part == "col") {
    result = col1
  } else if (part == "row") {
    result = row1
  } else if (part == "xy") {
    result = c(col1,row1)
  }
return(result)
}

然后使用通常的

writeData

并引用自定义函数进行调用。 它将把框架转储到指定范围的左上角。 您不需要指定开始和结束列/行。

# send column names
nRng = "YourNamedRange"
writeData(
  wb    = wb, 
  sheet = fn_namedRng(wb,nRng,"sh"), 
  xy    = fn_namedRng(wb,nRng,"xy"),
  x     = yourDataFrame, colNames = FALSE
  )

您可以类似地为

writeData

构建一个环绕函数，以最大限度地减少调用函数来获取各个地址元素的需要。