如何将创建列的函数应用到 R 中的数据框
How to apply a function that creates columns to dataframe in R
我尝试在 R 中对数据框进行一些数据整理，但我可以找出我的解决方案不起作用的原因。
auditlog <- data.frame(X_ID=c(1,2,4,5,6,7,8,9,10),
EVENT=c('create', 'delete', 'update', 'update', 'delete', 'delete', 'create',
'create', 'update'),
UNIT=c('30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx'),
CREATED=c('2015-12-01', '2015-12-01', '2015-12-02', '2015-12-04',
'2015-12-05', '2015-12-06', '2015-12-10', '2015-12-10',
'2015-12-10'),
R1=c('xxxxxxxx11', 'title', 'xxxxxxxx25', 'xxxxxxxx11', 'new_title',
'_title', 'xxxxxxxx12', 'xxxxxxxx87', 'xxxxxxxx87'),
R2=c('my_title', 'xxxxxxxx47', 'titleA', 'my_title', 'xxxxxxxx11',
'xxxxxxxx64', 'my_title_24', 'my_title_2', 'my_title_2'),
R3=c('red', '', 'title_42', 'new_title', '', '', 'green', 'blue', '_title_'),
R4=c('note', '', '', '', '', '', 'my_important_note', 'yet another note', ''))
这是我的审计日志的摘录：
X_ID EVENT UNIT CREATED R1 R2 R3 R4
1 create 30xx 2015-12-01 xxxxxxxx11 my_title red 'note'
2 delete 30xx 2015-12-01 title xxxxxxxx47
4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
6 delete 30xx 2015-12-05 new_title xxxxxxxx11
7 delete 30xx 2015-12-06 _title xxxxxxxx64
8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green 'my_important_note'
9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue 'yet another note'
10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
我想通过将 r1 到 r4 列重新分配给更清晰的列来构建此日志，如下所示：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 create 30xx 2015-12-01 xxxxxxxx11 my_title red 'note'
2 delete 30xx 2015-12-01 xxxxxxxx47 title
4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
7 delete 30xx 2015-12-06 xxxxxxxx64 _title
8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green 'my_important_note'
9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue 'yet another note'
10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
这是我在 R 中的方法，我尝试对每一行应用一个带条件的函数：
struct_log <- apply(auditlog, 1, function(row) {
if (row['EVENT'] == 'create') { row['ELEMNT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['COLOR'] <- row['R3']; row['COMMENT'] <- row['R4'] }
else if (row['EVENT'] == 'delete') { row['TITLE'] <- row['R1']; row['ELEMNT_ID'] <- row['R2'] }
else if (row['EVENT'] == 'update') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- row['R3'] } }
})
现在我不知道为什么我没有取回数据框。
使用索引更容易做到这一点。方法如下，但首先您还应该在 auditlog 定义中包含一个 stringsAsFactors=FALSE 参数：
k <- nrow(auditlog)
# build an empty dataframe:
struct_log <- data.frame(X_ID=numeric(k),
EVENT=character(k),
UNIT=character(k),
CREATED=character(k),
ELEMENT_ID=character(k),
TITLE=character(k),
NEW_TITLE=character(k),
COLOR=character(k),
COMMENT=character(k),
stringsAsFactors=FALSE)
struct_log[,1:4] <- auditlog[,1:4] # keep first 4 columns as-is
ind <- which(auditlog['EVENT']=='create')
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R1']
struct_log[ind,'TITLE'] <- auditlog[ind,'R2']
struct_log[ind,'COLOR'] <- auditlog[ind,'R3']
struct_log[ind,'COMMENT'] <- auditlog[ind,'R4']
ind <- which(auditlog['EVENT']=='delete')
struct_log[ind,'TITLE'] <- auditlog[ind,'R1']
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R2']
ind <- which(auditlog['EVENT']=='update')
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R1']
struct_log[ind,'TITLE'] <- auditlog[ind, 'R2']
struct_log[ind,'NEW_TITLE'] <- auditlog[ind,'R3']
你想要的结果：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 1 create 30xx 2015-12-01 xxxxxxxx11 my_title red note
2 2 delete 30xx 2015-12-01 xxxxxxxx47 title
3 4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
4 5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
5 6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
6 7 delete 30xx 2015-12-06 xxxxxxxx64 _title
7 8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green my_important_note
8 9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue yet another note
9 10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
不要忘记 将 stringsAsFactors=FALSE 包含到 auditlog 定义中（类似于我在上面对 struct_log 所做的） , 不然不行...
（好的，评论后的第二个答案：）
我修改了你最初的方法。首先，您需要在 if 语句中设置 all 列（即包括空白列）；其次，您需要 return 函数结果。这是整个代码（包括 auditlog 定义中的修改）：
auditlog <- data.frame(X_ID=c(1,2,4,5,6,7,8,9,10),
EVENT=c('create', 'delete', 'update', 'update', 'delete', 'delete', 'create',
'create', 'update'),
UNIT=c('30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx'),
CREATED=c('2015-12-01', '2015-12-01', '2015-12-02', '2015-12-04',
'2015-12-05', '2015-12-06', '2015-12-10', '2015-12-10',
'2015-12-10'),
R1=c('xxxxxxxx11', 'title', 'xxxxxxxx25', 'xxxxxxxx11', 'new_title',
'_title', 'xxxxxxxx12', 'xxxxxxxx87', 'xxxxxxxx87'),
R2=c('my_title', 'xxxxxxxx47', 'titleA', 'my_title', 'xxxxxxxx11',
'xxxxxxxx64', 'my_title_24', 'my_title_2', 'my_title_2'),
R3=c('red', '', 'title_42', 'new_title', '', '', 'green', 'blue', '_title_'),
R4=c('note', '', '', '', '', '', 'my_important_note', 'yet another note', ''),
stringsAsFactors = FALSE # added
)
ff <- function(row) {
if (row['EVENT'] == 'create') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- ''; row['COLOR'] <- row['R3']; row['COMMENT'] <- row['R4'] }
else if (row['EVENT'] == 'delete') { row['ELEMENT_ID'] <- row['R2']; row['TITLE'] <- row['R1']; row['NEW_TITLE'] <- row['COLOR'] <- row['COMMENT'] <- '' }
else if (row['EVENT'] == 'update') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- row['R3']; row['COLOR'] <- row['COMMENT'] <- '' }
return(row) # added
}
struct_log <- t(apply(auditlog,1,ff))
struct_log <- as.data.frame(struct_log)
new_names <- c("ELEMENT_ID", "TITLE", "NEW_TITLE", "COLOR", "COMMENT")
names(struct_log) <- c(names(auditlog), new_names)
struct_log[c('R1', 'R2', 'R3', 'R4')] <- list(NULL) # drop 'R' columns
之后你又得到了：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 1 create 30xx 2015-12-01 xxxxxxxx11 my_title red note
2 2 delete 30xx 2015-12-01 xxxxxxxx47 title
3 4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
4 5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
5 6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
6 7 delete 30xx 2015-12-06 xxxxxxxx64 _title
7 8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green my_important_note
8 9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue yet another note
9 10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
我尝试在 R 中对数据框进行一些数据整理，但我可以找出我的解决方案不起作用的原因。
auditlog <- data.frame(X_ID=c(1,2,4,5,6,7,8,9,10),
EVENT=c('create', 'delete', 'update', 'update', 'delete', 'delete', 'create',
'create', 'update'),
UNIT=c('30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx'),
CREATED=c('2015-12-01', '2015-12-01', '2015-12-02', '2015-12-04',
'2015-12-05', '2015-12-06', '2015-12-10', '2015-12-10',
'2015-12-10'),
R1=c('xxxxxxxx11', 'title', 'xxxxxxxx25', 'xxxxxxxx11', 'new_title',
'_title', 'xxxxxxxx12', 'xxxxxxxx87', 'xxxxxxxx87'),
R2=c('my_title', 'xxxxxxxx47', 'titleA', 'my_title', 'xxxxxxxx11',
'xxxxxxxx64', 'my_title_24', 'my_title_2', 'my_title_2'),
R3=c('red', '', 'title_42', 'new_title', '', '', 'green', 'blue', '_title_'),
R4=c('note', '', '', '', '', '', 'my_important_note', 'yet another note', ''))
这是我的审计日志的摘录：
X_ID EVENT UNIT CREATED R1 R2 R3 R4
1 create 30xx 2015-12-01 xxxxxxxx11 my_title red 'note'
2 delete 30xx 2015-12-01 title xxxxxxxx47
4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
6 delete 30xx 2015-12-05 new_title xxxxxxxx11
7 delete 30xx 2015-12-06 _title xxxxxxxx64
8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green 'my_important_note'
9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue 'yet another note'
10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
我想通过将 r1 到 r4 列重新分配给更清晰的列来构建此日志，如下所示：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 create 30xx 2015-12-01 xxxxxxxx11 my_title red 'note'
2 delete 30xx 2015-12-01 xxxxxxxx47 title
4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
7 delete 30xx 2015-12-06 xxxxxxxx64 _title
8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green 'my_important_note'
9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue 'yet another note'
10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
这是我在 R 中的方法，我尝试对每一行应用一个带条件的函数：
struct_log <- apply(auditlog, 1, function(row) {
if (row['EVENT'] == 'create') { row['ELEMNT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['COLOR'] <- row['R3']; row['COMMENT'] <- row['R4'] }
else if (row['EVENT'] == 'delete') { row['TITLE'] <- row['R1']; row['ELEMNT_ID'] <- row['R2'] }
else if (row['EVENT'] == 'update') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- row['R3'] } }
})
现在我不知道为什么我没有取回数据框。
使用索引更容易做到这一点。方法如下，但首先您还应该在 auditlog 定义中包含一个 stringsAsFactors=FALSE 参数：
k <- nrow(auditlog)
# build an empty dataframe:
struct_log <- data.frame(X_ID=numeric(k),
EVENT=character(k),
UNIT=character(k),
CREATED=character(k),
ELEMENT_ID=character(k),
TITLE=character(k),
NEW_TITLE=character(k),
COLOR=character(k),
COMMENT=character(k),
stringsAsFactors=FALSE)
struct_log[,1:4] <- auditlog[,1:4] # keep first 4 columns as-is
ind <- which(auditlog['EVENT']=='create')
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R1']
struct_log[ind,'TITLE'] <- auditlog[ind,'R2']
struct_log[ind,'COLOR'] <- auditlog[ind,'R3']
struct_log[ind,'COMMENT'] <- auditlog[ind,'R4']
ind <- which(auditlog['EVENT']=='delete')
struct_log[ind,'TITLE'] <- auditlog[ind,'R1']
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R2']
ind <- which(auditlog['EVENT']=='update')
struct_log[ind,'ELEMENT_ID'] <- auditlog[ind,'R1']
struct_log[ind,'TITLE'] <- auditlog[ind, 'R2']
struct_log[ind,'NEW_TITLE'] <- auditlog[ind,'R3']
你想要的结果：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 1 create 30xx 2015-12-01 xxxxxxxx11 my_title red note
2 2 delete 30xx 2015-12-01 xxxxxxxx47 title
3 4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
4 5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
5 6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
6 7 delete 30xx 2015-12-06 xxxxxxxx64 _title
7 8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green my_important_note
8 9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue yet another note
9 10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_
不要忘记 将 stringsAsFactors=FALSE 包含到 auditlog 定义中（类似于我在上面对 struct_log 所做的） , 不然不行...
（好的，评论后的第二个答案：）
我修改了你最初的方法。首先，您需要在 if 语句中设置 all 列（即包括空白列）；其次，您需要 return 函数结果。这是整个代码（包括 auditlog 定义中的修改）：
auditlog <- data.frame(X_ID=c(1,2,4,5,6,7,8,9,10),
EVENT=c('create', 'delete', 'update', 'update', 'delete', 'delete', 'create',
'create', 'update'),
UNIT=c('30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx', '30xx'),
CREATED=c('2015-12-01', '2015-12-01', '2015-12-02', '2015-12-04',
'2015-12-05', '2015-12-06', '2015-12-10', '2015-12-10',
'2015-12-10'),
R1=c('xxxxxxxx11', 'title', 'xxxxxxxx25', 'xxxxxxxx11', 'new_title',
'_title', 'xxxxxxxx12', 'xxxxxxxx87', 'xxxxxxxx87'),
R2=c('my_title', 'xxxxxxxx47', 'titleA', 'my_title', 'xxxxxxxx11',
'xxxxxxxx64', 'my_title_24', 'my_title_2', 'my_title_2'),
R3=c('red', '', 'title_42', 'new_title', '', '', 'green', 'blue', '_title_'),
R4=c('note', '', '', '', '', '', 'my_important_note', 'yet another note', ''),
stringsAsFactors = FALSE # added
)
ff <- function(row) {
if (row['EVENT'] == 'create') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- ''; row['COLOR'] <- row['R3']; row['COMMENT'] <- row['R4'] }
else if (row['EVENT'] == 'delete') { row['ELEMENT_ID'] <- row['R2']; row['TITLE'] <- row['R1']; row['NEW_TITLE'] <- row['COLOR'] <- row['COMMENT'] <- '' }
else if (row['EVENT'] == 'update') { row['ELEMENT_ID'] <- row['R1']; row['TITLE'] <- row['R2']; row['NEW_TITLE'] <- row['R3']; row['COLOR'] <- row['COMMENT'] <- '' }
return(row) # added
}
struct_log <- t(apply(auditlog,1,ff))
struct_log <- as.data.frame(struct_log)
new_names <- c("ELEMENT_ID", "TITLE", "NEW_TITLE", "COLOR", "COMMENT")
names(struct_log) <- c(names(auditlog), new_names)
struct_log[c('R1', 'R2', 'R3', 'R4')] <- list(NULL) # drop 'R' columns
之后你又得到了：
X_ID EVENT UNIT CREATED ELEMENT_ID TITLE NEW_TITLE COLOR COMMENT
1 1 create 30xx 2015-12-01 xxxxxxxx11 my_title red note
2 2 delete 30xx 2015-12-01 xxxxxxxx47 title
3 4 update 30xx 2015-12-02 xxxxxxxx25 titleA title_42
4 5 update 30xx 2015-12-04 xxxxxxxx11 my_title new_title
5 6 delete 30xx 2015-12-05 xxxxxxxx11 new_title
6 7 delete 30xx 2015-12-06 xxxxxxxx64 _title
7 8 create 30xx 2015-12-10 xxxxxxxx12 my_title_24 green my_important_note
8 9 create 30xx 2015-12-10 xxxxxxxx87 my_title_2 blue yet another note
9 10 update 30xx 2015-12-10 xxxxxxxx87 my_title_2 _title_