R 标签长度2应为1或0

我有以下数据：

library(dplyr)
dframefull <- data.frame(id = c(1,1,1,1,1,1,1,1), name = c("Google", "Google", "Google", "Google", "Google", "Google", "Google", "Google"), date = c("12/8/2014 19:30:57", "26/8/2014 19:30:57", "27/8/2014 10:12:01", "27/8/2014 14:10:29", "27/8/2014 14:10:32", "27/8/2014 14:10:33", "3/9/2014 14:10:32", "14/9/2014 19:30:57"), mytext = c("out text", "text", "another", "text", "here", "other text", "text more", "out text 2"))
dframekeep <- data.frame(id = c(1), name = c("Google"), date = c("27/8/2014 14:10:32"))

你知道我该怎么修吗

预期结果是在考虑秒数的情况下，在第二个数据帧的时间戳之前和之后分别保持10天和10天

data.frame(id = c(1,1), label = c("before", "after"), name = c("Google", "Google"), mytext = c("text another text here", "other text text more"))
  id  label   name                 mytext
1  1 before Google text another text here
2  1  after Google   other text text more

---

我对你的代码做了两处修改。首先，我在

data.frame

中使用了

stringsAsFactors=FALSE

参数。其次，我将

date

列转换为

POSIXct

类（日期/时间类）

这允许示例代码无错误地运行。我希望您的生产数据中的解决方案与此类似

library(dplyr)
dframefull <- data.frame(id = c(1,1,1,1,1,1,1,1), 
                         name = c("Google", "Google", "Google", "Google", 
                                  "Google", "Google", "Google", "Google"), 
                         date = c("12/8/2014 19:30:57", "26/8/2014 19:30:57", 
                                  "27/8/2014 10:12:01", "27/8/2014 14:10:29", 
                                  "27/8/2014 14:10:32", "27/8/2014 14:10:33", 
                                  "3/9/2014 14:10:32",  "14/9/2014 19:30:57"), 
                         mytext = c("out text", "text", "another", "text", 
                                    "here", "other text", "text more", 
                                    "out text 2"),
                         stringsAsFactors = FALSE) %>% 
  mutate(date = as.POSIXct(date, 
                           format = "%d/%m/%Y %H:%M:%S"))
dframekeep <- data.frame(id = c(1), 
                         name = c("Google"), 
                         date = c("27/8/2014 14:10:32"),
                         stringsAsFactors = FALSE) %>% 
  mutate(date = as.POSIXct(date, format = "%d/%m/%Y %H:%M:%S"))

b <- with(dframefull, 
          aggregate(list(mytext=mytext), 
                    by=list(id=id, 
                            label=factor(I(date > dframekeep$date), labels=c("before", "after")), 
                            name=name), 
                    FUN=paste))

库（dplyr）
dframefull%
突变（日期=as.POSIXct（日期，
format=“%d/%m/%Y%H:%m:%S”））
数据帧保留%
变异（日期=as.POSIXct（日期，格式=“%d/%m/%Y%H:%m:%S”））
b dframekeep$date），labels=c（“之前”、“之后”），
名称=名称），
乐趣=粘贴）

预期结果是什么？在您的代码中，

dframefull

和

dframekeep

的

date

列是因素。也许你需要先将它们转换成

Date

对象？@Benjamin我使用了Date对象，但它没有考虑小时分钟seconds@Bruno更新了预期结果，我很抱歉。您需要使它们成为

POSIXct

对象。我很快就会提交一份答复。非常感谢。是的。还有一个问题，您知道如何在第二个数据帧的日期之前和之后保留10天吗？请尝试在标签之前保留之前和之后的10天。我尝试的是问题中的预期输出

data.frame(id = c(1,1), label = c("before", "after"), name = c("Google", "Google"), mytext = c("text another text here", "other text text more"))
  id  label   name                 mytext
1  1 before Google text another text here
2  1  after Google   other text text more

library(dplyr)
dframefull <- data.frame(id = c(1,1,1,1,1,1,1,1), 
                         name = c("Google", "Google", "Google", "Google", 
                                  "Google", "Google", "Google", "Google"), 
                         date = c("12/8/2014 19:30:57", "26/8/2014 19:30:57", 
                                  "27/8/2014 10:12:01", "27/8/2014 14:10:29", 
                                  "27/8/2014 14:10:32", "27/8/2014 14:10:33", 
                                  "3/9/2014 14:10:32",  "14/9/2014 19:30:57"), 
                         mytext = c("out text", "text", "another", "text", 
                                    "here", "other text", "text more", 
                                    "out text 2"),
                         stringsAsFactors = FALSE) %>% 
  mutate(date = as.POSIXct(date, 
                           format = "%d/%m/%Y %H:%M:%S"))
dframekeep <- data.frame(id = c(1), 
                         name = c("Google"), 
                         date = c("27/8/2014 14:10:32"),
                         stringsAsFactors = FALSE) %>% 
  mutate(date = as.POSIXct(date, format = "%d/%m/%Y %H:%M:%S"))

b <- with(dframefull, 
          aggregate(list(mytext=mytext), 
                    by=list(id=id, 
                            label=factor(I(date > dframekeep$date), labels=c("before", "after")), 
                            name=name), 
                    FUN=paste))