R 提取同一工作表中的多个数据文件

更新： 手动剪切并粘贴到多张图纸中。如果能找到一个解决办法，那就太好了

问题： 给定以下虚拟数据集：

structure(list(V1 = structure(c(8L, 6L, 2L, 4L, 1L, 1L, 1L, 1L, 
9L, 5L, 2L, 1L, 1L, 1L, 1L, 10L, 7L, 3L), .Label = c("", "1", 
"12", "5", "Age", "Class A", "Height", "Number of Boys", "More Boys", 
"More Girls"), class = "factor"), V2 = structure(c(1L, 5L, 3L, 
4L, 1L, 1L, 1L, 1L, 1L, 6L, 3L, 1L, 1L, 1L, 1L, 1L, 7L, 2L), .Label = c("", 
"12", "2", "6", "Class B", "Time", "Weight"), class = "factor"), 
    V3 = structure(c(1L, 5L, 3L, 4L, 1L, 1L, 1L, 1L, 1L, 6L, 
    3L, 1L, 1L, 1L, 1L, 1L, 7L, 2L), .Label = c("", "13", "3", 
    "7", "Class C", "Next", "Time"), class = "factor"), V4 = structure(c(1L, 
    5L, 3L, 4L, 1L, 1L, 1L, 1L, 1L, 6L, 3L, 1L, 1L, 1L, 1L, 1L, 
    6L, 2L), .Label = c("", "14", "4", "8", "Class D", "Day"), class = "factor"), 
    V5 = c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA), V6 = c(NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), V7 = c(NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
    ), V8 = c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA), V9 = c(NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), V10 = structure(c(5L, 
    4L, 3L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L), .Label = c("", "1", "8", "Class E", "Number of Girls"
    ), class = "factor"), V11 = structure(c(1L, 4L, 3L, 2L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("", 
    "2", "8", "Class F"), class = "factor"), V12 = structure(c(1L, 
    4L, 3L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L), .Label = c("", "3", "9", "Class G"), class = "factor"), 
    V13 = structure(c(1L, 4L, 2L, 3L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("", "0", "4", 
    "Class Q"), class = "factor"), V14 = c(NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA)), class = "data.frame", row.names = c(NA, 
-18L))

这看起来像（被截断的）：

正如我们所希望看到的，这些是同一工作表上的独立“文件”。我一直在寻找一种快速的方法来挖掘不同的数据集，但还没有。附近有工作吗？ 我的想法是使用基于序列的选择，比如说每20行选择一次，但如果有数百万行，这显然会失败

预期输出（及其各自的行）

---

提前感谢。

使用@alexis_laz的解决方案

库（矩阵）
x[[抑制14个列名'V1'，'V2'，'V3'…]]
#>                                  
#>  [1,] | . . . . . . . . | . . . .
#>  [2,] | | | | . . . . . | | | | .
#>  [3,] | | | | . . . . . | | | | .
#>  [4,] | | | | . . . . . | | | | .
#>  [5,] . . . . . . . . . . . . . .
#>  [6,] . . . . . . . . . . . . . .
#>  [7,] . . . . . . . . . . . . . .
#>  [8,] . . . . . . . . . . . . . .
#>  [9,] | . . . . . . . . . . . . .
#> [10,] | | | | . . . . . . . . . .
#> [11,] | | | | . . . . . . . . . .
#> [12,] . . . . . . . . . . . . . .
#> [13,] . . . . . . . . . . . . . .
#> [14,] . . . . . . . . . . . . . .
#> [15,] . . . . . . . . . . . . . .
#> [16,] | . . . . . . . . . . . . .
#> [17,] | | | | . . . . . . . . . .
#> [18,] | | | | . . . . . . . . . .
sm=作为矩阵（汇总（m））
d=距离（sm，“曼哈顿”）
gr=cutree（hclust（d，“单个”），h=1）
res 18 x 13类“dgCMatrix”的稀疏矩阵
#>                                
#>  [1,] 1 . . . . . . . . 4.
#>  [2,] 1 1 1 1 . . . . . 4 4 4 4
#>  [3,] 1 1 1 1 . . . . . 4 4 4 4
#>  [4,] 1 1 1 1 . . . . . 4 4 4 4
#>  [5,] . . . . . . . . . . . . .
#>  [6,] . . . . . . . . . . . . .
#>  [7,] . . . . . . . . . . . . .
#>  [8,] . . . . . . . . . . . . .
#>  [9,] 2 . . . . . . . . . . . .
#> [10,] 2 2 2 2 . . . . . . . . .
#> [11,] 2 2 2 2 . . . . . . . . .
#> [12,] . . . . . . . . . . . . .
#> [13,] . . . . . . . . . . . . .
#> [14,] . . . . . . . . . . . . .
#> [15,] . . . . . . . . . . . . .
#> [16,] 3 . . . . . . . . . . . .
#> [17,] 3 3 3 3 . . . . . . . . .
#> [18,] 3 3 3 3 . . . . . . . . .
res2$`1`
#>V1 V2 V3 V4
#>1ï？男孩人数
#>2 A类B类C类D类
#> 3                 1       2       3       4
#> 4                 5       6       7       8
#> 
#> $`2`
#>V1 V2 V3 V4
#>还有9个男孩
#>第二天是10岁
#> 11         1    2    3   4
#> 
#> $`3`
#>V1 V2 V3 V4
#>还有16个女孩
#>17身高体重时间-天
#> 18         12     12   13  14
#> 
#> $`4`
#>V10 V11 V12 V13
#>1女孩人数
#>2 E类F类G类Q类
#> 3               8       8       9       0
#> 4               1       2       3       4

---

于2019年4月10日由（v0.2.1）

使用@alexis_laz在

库（矩阵）
x[[抑制14个列名'V1'，'V2'，'V3'…]]
#>                                  
#>  [1,] | . . . . . . . . | . . . .
#>  [2,] | | | | . . . . . | | | | .
#>  [3,] | | | | . . . . . | | | | .
#>  [4,] | | | | . . . . . | | | | .
#>  [5,] . . . . . . . . . . . . . .
#>  [6,] . . . . . . . . . . . . . .
#>  [7,] . . . . . . . . . . . . . .
#>  [8,] . . . . . . . . . . . . . .
#>  [9,] | . . . . . . . . . . . . .
#> [10,] | | | | . . . . . . . . . .
#> [11,] | | | | . . . . . . . . . .
#> [12,] . . . . . . . . . . . . . .
#> [13,] . . . . . . . . . . . . . .
#> [14,] . . . . . . . . . . . . . .
#> [15,] . . . . . . . . . . . . . .
#> [16,] | . . . . . . . . . . . . .
#> [17,] | | | | . . . . . . . . . .
#> [18,] | | | | . . . . . . . . . .
sm=作为矩阵（汇总（m））
d=距离（sm，“曼哈顿”）
gr=cutree（hclust（d，“单个”），h=1）
res 18 x 13类“dgCMatrix”的稀疏矩阵
#>                                
#>  [1,] 1 . . . . . . . . 4.
#>  [2,] 1 1 1 1 . . . . . 4 4 4 4
#>  [3,] 1 1 1 1 . . . . . 4 4 4 4
#>  [4,] 1 1 1 1 . . . . . 4 4 4 4
#>  [5,] . . . . . . . . . . . . .
#>  [6,] . . . . . . . . . . . . .
#>  [7,] . . . . . . . . . . . . .
#>  [8,] . . . . . . . . . . . . .
#>  [9,] 2 . . . . . . . . . . . .
#> [10,] 2 2 2 2 . . . . . . . . .
#> [11,] 2 2 2 2 . . . . . . . . .
#> [12,] . . . . . . . . . . . . .
#> [13,] . . . . . . . . . . . . .
#> [14,] . . . . . . . . . . . . .
#> [15,] . . . . . . . . . . . . .
#> [16,] 3 . . . . . . . . . . . .
#> [17,] 3 3 3 3 . . . . . . . . .
#> [18,] 3 3 3 3 . . . . . . . . .
res2$`1`
#>V1 V2 V3 V4
#>1ï？男孩人数
#>2 A类B类C类D类
#> 3                 1       2       3       4
#> 4                 5       6       7       8
#> 
#> $`2`
#>V1 V2 V3 V4
#>还有9个男孩
#>第二天是10岁
#> 11         1    2    3   4
#> 
#> $`3`
#>V1 V2 V3 V4
#>还有16个女孩
#>17身高体重时间-天
#> 18         12     12   13  14
#> 
#> $`4`
#>V10 V11 V12 V13
#>1女孩人数
#>2 E类F类G类Q类
#> 3               8       8       9       0
#> 4               1       2       3       4

---

由（v0.2.1）创建于2019-04-10以下代码创建了一个列名称设置正确的data.frames列表。然而，它依赖于这样一个事实：在您的工作表中，“表的列”至少由一个空列分隔

df <- apply(df, 2, function(x) gsub("^$|^ $", NA, x))
empty_cols <- sapply(1:ncol(df), function(i){length(which(is.na(df[, i])))==nrow(df)})
start_cols <- c(1, which(diff(empty_cols)==-1)+1)
if (is.na(df[1, 1])) start_cols <- start_cols[-1]
start_rows <- lapply(start_cols, function(i){
  start_rows <- c(1, which(diff(is.na(df[, i]))==-1)+1)
  if (is.na(df[1, i])) start_rows <- start_rows[-1]
  start_rows})

end_rows <- lapply(start_cols, function(i){
  end_rows <- c(1, which(diff(is.na(df[, i]))==1))
  if (!is.na(df[nrow(df), i])) end_rows <- c(end_rows, nrow(df))
  end_rows[-1]})

data.sets <- list()
for (i in 1:length(start_cols)) {
    for (j in 1:length(start_rows[[i]])){

      col <- start_cols[i]
      row <- start_rows[[i]][j]
      start_row <- row+1
      end_row <- end_rows[[i]][j]
      name <- df[row, col]
      ncol <- which(diff(is.na(df[row+1, col:ncol(df)]))==1)[1]
      end_col <- col+ncol-1
      column_names <- df[start_row, col:end_col]
      data <- df[(start_row+1):end_row, col:end_col]
      data <- matrix(data, ncol = length(col:end_col))
      data <- as.data.frame(data)
      names(data) <- column_names
      data.sets[[name]] <- data
    }
}


> data.sets
$`Number of Boys`
  Class A Class B Class C Class D
1       1       2       3       4
2       5       6       7       8

$`More Boys`
  Age Time Next Day
1   1    2    3   4

$`More Girls`
  Height Weight Time Day
1     12     12   13  14

$`Number of Girls`
  Class E Class F Class G Class Q
1       8       8       9       0
2       1       2       3       4

---

df以下代码创建了列名称设置正确的data.frames列表。然而，它依赖于这样一个事实：在您的工作表中，“表列”是separa
Three data sets: 
A: Number of Boys
B: Number of Girls
C: More Boys

df <- apply(df, 2, function(x) gsub("^$|^ $", NA, x))
empty_cols <- sapply(1:ncol(df), function(i){length(which(is.na(df[, i])))==nrow(df)})
start_cols <- c(1, which(diff(empty_cols)==-1)+1)
if (is.na(df[1, 1])) start_cols <- start_cols[-1]
start_rows <- lapply(start_cols, function(i){
  start_rows <- c(1, which(diff(is.na(df[, i]))==-1)+1)
  if (is.na(df[1, i])) start_rows <- start_rows[-1]
  start_rows})

end_rows <- lapply(start_cols, function(i){
  end_rows <- c(1, which(diff(is.na(df[, i]))==1))
  if (!is.na(df[nrow(df), i])) end_rows <- c(end_rows, nrow(df))
  end_rows[-1]})

data.sets <- list()
for (i in 1:length(start_cols)) {
    for (j in 1:length(start_rows[[i]])){

      col <- start_cols[i]
      row <- start_rows[[i]][j]
      start_row <- row+1
      end_row <- end_rows[[i]][j]
      name <- df[row, col]
      ncol <- which(diff(is.na(df[row+1, col:ncol(df)]))==1)[1]
      end_col <- col+ncol-1
      column_names <- df[start_row, col:end_col]
      data <- df[(start_row+1):end_row, col:end_col]
      data <- matrix(data, ncol = length(col:end_col))
      data <- as.data.frame(data)
      names(data) <- column_names
      data.sets[[name]] <- data
    }
}


> data.sets
$`Number of Boys`
  Class A Class B Class C Class D
1       1       2       3       4
2       5       6       7       8

$`More Boys`
  Age Time Next Day
1   1    2    3   4

$`More Girls`
  Height Weight Time Day
1     12     12   13  14

$`Number of Girls`
  Class E Class F Class G Class Q
1       8       8       9       0
2       1       2       3       4