R数据表：如何在特定单元格的正下方找到未知数量的空单元格，并用编号的字符串填充它们

Question

I know how to find empty cells in a data.table in general, but this is a little trickier and I haven't quite figured out how to manage this.

Lets say I have a data.table which in df[2,1] contains my keyword string i need to target: "Smart triggered!" in this case.

The amount of smart triggers used by the user of the instrument can vary, but they result in 1 empty cell below df[2,1] for each trigger, and their name in the 2nd column

I'm looking for a way to figure out how many empty cells are directly below "Smart triggered!" and fill them sequentially with "Smart trigger 1", Smart trigger 2"... until we hit the first next cell in column 1 that contains something ('Instrument'). There are potential other empty cells further down in the table that I do not want to alter.

In this case my info files read into a data.table looking like this:

df <- data.frame(name = c("Trigger", "Smart Triggered!", "", "", "Instrument", "Beam", "Core speed", "Channel1", "Channel2", "Channel3", "Channel4", ""),
values = c("SWS", "", "FLRED", "FLORANGE", "Demo", "5um", "2.2", "FWS", "SWS", "FLRED", "FLORANGE", "x"))

I suspect perhaps a while loop, but there are probably better data table solutions than that.

Answer 1

see if this works for you:

library(tidyverse)

df <- data.frame(name = c("Trigger", "Smart Triggered!", "", "", "Instrument", "Beam", "Core speed", "Channel1", "Channel2", "Channel3", "Channel4", ""),
                 values = c("SWS", "", "FLRED", "FLORANGE", "Demo", "5um", "2.2", "FWS", "SWS", "FLRED", "FLORANGE", "x"))

df %>%
  mutate(new_name = ifelse(as.character(name) == "", NA, as.character(name))) %>%
  tidyr::fill(new_name) %>%
  split(.$new_name) %>%
  map_df(., ~.x %>% mutate(row_no = row_number(),
                           row_no = lag(row_no),
                           new_name1 = ifelse(is.na(row_no), 
                                              as.character(new_name), 
                                              paste0(as.character(new_name), "_", row_no)))) %>%
  select(name, new_name, new_name1, values) %>%
  full_join(df, .) %>%
  mutate(name = as.character(name)) %>%
  mutate(name = ifelse(new_name == "Smart Triggered!", new_name1, name)) %>%
  select(name, values)

Answer 2

You can use a lapply function. I dont know how is your for loop solution, but mine requiere using <<- which is not ideal so maybe you prefer to stick to a for loop

library(data.table)
df <- data.frame(name = c("Trigger", "Smart Triggered!", "", "", "Instrument", "Beam", "Core speed", "Channel1", "Channel2", "Channel3", "Channel4", ""),
                 values = c("SWS", "", "FLRED", "FLORANGE", "Demo", "5um", "2.2", "FWS", "SWS", "FLRED", "FLORANGE", "x"))

df <- as.data.table(df)

df$name <- as.character(df$name)
counting <<- 1
df$name[2:nrow(df)] <- unlist(lapply(2:nrow(df), function(x){
  if((df[x,]$name=="") && (df[x-counting,]$name=="Smart Triggered!")){
    counting <<- counting + 1
    return(paste0("Smart trigger ", counting))
  }
  else{
    counting <<- 1
    return(df[x,]$name)
  }
}
))
df