在 dplyr 中使用 filter() 查找 grepl 值的第一次出现并返回它和所有后续行

Question

我正在尝试使用 dplyr 按组过滤数据帧，其中捕获了一行中第一次出现的字符串“ReadingOnset”，并将它和所有后续行传递到一个新的数据帧中。

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"))

这是我一直在尝试的解决方案类型的示例，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Name、StimulusName 和StimuliBlock 进行分组。然后我试图在Reading_Onset 列中找到第一次出现的“ReadingOnset”，并从中返回所有前导行（包括带有“ReadingOnset”的行）。

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

Answer 1

你可以用你的数据集试试这个：

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      ""

Answer 2

我不知道您是想要在 ReadingOnset 之前（包括 ReadingOnset）之前的所有内容，还是想要在 ReadingOnset 之后（包括 ReadingOnset）之后的所有内容，所以我将两者都显示。

之前和包括的所有内容：

library(dplyr)

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

之后的所有内容，包括：

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

我们所做的只是根据“ReadingOnset”所在的行号进行过滤。