在dplyr中使用filter()来查找grepl值的第一个匹配项,并将其和所有后续行返回

时间:2018-10-23 10:55:16

标签: r dataframe filter dplyr grepl

我正在尝试使用dplyr按组过滤数据帧,其中捕获了字符串'ReadingOnset'在一行中的首次出现,并将其和所有后续行传递到新的数据帧中。

Text_Stimuli <- structure(list(Name = c("Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", 
"Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1", "Sub1"
), StimulusName = c("GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", "GenLie20", 
"GenLie20"), StimuliBlock = c("Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4", "Block_4", "Block_4", "Block_4", "Block_4", 
"Block_4", "Block_4"), Reading_Onset = c("", "", "", "", "", 
"", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", 
"", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", 
"", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", 
"", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", 
"", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", 
"", "", "", "", "", "", "", "", "ReadingOnset", "", "", "", "", 
"", "", "")), row.names = c(NA, -101L), vars = c("Name", "StimulusName", 
"StimuliBlock"), drop = TRUE, indices = list(0:100), group_sizes = 101L, biggest_group_size = 101L, labels = structure(list(
    Name = "Innocent Subject 15", StimulusName = "GenLie20", 
    StimuliBlock = "Block_4"), row.names = c(NA, -1L), class = "data.frame", vars = c("Name", 
"StimulusName", "StimuliBlock"), drop = TRUE), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"))

这是我一直尝试开始工作的解决方案类型的一个例子,但没有成功。

Test <- Text_Stimuli %>% 
  group_by(Name, StimulusName, StimuliBlock)%>%   
  filter(!lead(cumsum(grepl("ReadingOnset", Reading_Onset)), default = 0))

如您所见,我正在尝试按NameStimulusNameStimuliBlock进行分组。然后,我试图在列Reading_Onset中找到“ ReadingOnset”的第一个匹配项,并从中返回所有前导行(包括带有“ ReadingOnset”的行)。

我一直在尝试使该解决方案适应问题的反面:https://stackoverflow.com/a/37922522/2653210

2 个答案:

答案 0 :(得分:2)

您可以尝试使用数据集进行此操作:

library(dplyr)
library(stringr)
library(zoo)
df %>% filter(ifelse(str_detect(Reading_Onset,"ReadingOnset"),TRUE,NA) %>%
              na.locf(na.rm=FALSE))

## A tibble: 8 x 4
## Groups:   Name, StimulusName, StimuliBlock [1]
#  Name                StimulusName StimuliBlock Reading_Onset
#  <chr>               <chr>        <chr>        <chr>        
#1 Innocent Subject 15 GenLie20     Block_4      ReadingOnset 
#2 Innocent Subject 15 GenLie20     Block_4      ""           
#3 Innocent Subject 15 GenLie20     Block_4      ""           
#4 Innocent Subject 15 GenLie20     Block_4      ""           
#5 Innocent Subject 15 GenLie20     Block_4      ""           
#6 Innocent Subject 15 GenLie20     Block_4      ""           
#7 Innocent Subject 15 GenLie20     Block_4      ""           
#8 Innocent Subject 15 GenLie20     Block_4      ""

答案 1 :(得分:1)

我无法确定您是否想要包括ReadingOnset之前的内容,还是想要包括ReadingOnset的所有内容,所以我将同时显示两者。

之前和包括的所有内容:

library(dplyr)

Text_Stimuli %>% 
  filter(row_number() <= which(grepl("ReadingOnset", Reading_Onset)))

包括以下所有内容:

Text_Stimuli %>% 
  filter(row_number() >= which(grepl("ReadingOnset", Reading_Onset)))

我们正在做的只是基于找到“ ReadingOnset”的行号进行过滤。