如何按R中的第一个和最后一个字符过滤消息？

如何解决如何按R中的第一个和最后一个字符过滤消息？

我绝对是R语言的初学者，但我必须尽快解决一个简单的任务

我必须分析电报中的消息。我已经做了什么：我从电报中解析了提取的jstor文件，并得到了一个表，其中1行是一条包含ID，日期，文本的消息。

telegram.to.data.frame <- function(path) {
  
  file <- list.files(path,pattern = "result*",full.names = T) # returns all files that located in a specified path
  data <- fromJSON(file = file) # parse JSON file to R list
  messages <- lapply(data$messages,function(message) { # label service messages
    if (message$type == "message") message else NULL
  })
  messages <- messages[unlist(lapply(messages,function(message) !is.null(message)))] # delete service messages
  df <- lapply(messages,function(message) { # transform JSON list to data frame
    if (!is.list(message$text)) { 
      text <- message$text # if not list than just string
    } else { # if list then process list of characters
      text <- lapply(message$text,function(submessage) {
        if (is.list(submessage)) submessage$text else submessage
      }) %>%
        c(sep = "\n") %>%
        do.call("paste",args = .)
    }
    c(
      "id" = message$id,"date" = message$date,"from" = if (data$type == "public_supergroup") {
        if (!is.null(message$from)) {
          message$from
        } else {
          NA
        }
      } else {
        data$name
      },"from_id" = if (data$type == "public_supergroup") {
        message$from_id
      } else {
        data$id
      },"text" = text
    )
  }) %>%
    c(stringsAsFactors = FALSE) %>%
    do.call("rbind.data.frame",args = .)
  names(df) <- c("id","date","from","from_id","text")
  df$date <- ymd_hms(df$date) # coerce character to date type
  
  
  text_list <- list() 
  time_limit <- 5 # minutes
  i <- 1
  while (i != nrow(df)) {
    j <- i + 1
    text_list[[paste0(i)]] <- matrix(
      c(as.character(df$id[i]),as.character(df$date[i]),as.character(df$from[i]),as.character(df$from_id[i]),as.character(df$text[i])),nrow = 1,ncol = 5)
    while (df$from_id[i] == df$from_id[j] & j != nrow(df)) {
      if (abs(difftime(df$date[j - 1],df$date[j],units = "mins")) < time_limit) {
        j <- j + 1
      } else {
        break
      }
    }
    if (length(i:(j - 1)) > 1) {
      text_list[[paste0(i)]][1,5] <- df[i:(j - 1),5] %>% 
        paste(collapse = "\n")
    }
    i <- j
  }
  
  df <- do.call("rbind",text_list) %>% 
    as.data.frame(stringsAsFactors=FALSE) %>% 
    `names<-` (c("id","text"))
  df$date <- ymd_hms(df$date)
  df$group_name <- data$name
  df$group_type <- data$type
  df$messenger <- "telegram"
  
  return(df)
}

现在，我需要根据单词“ wechat”的前10个字符和后10个字符的存在性来过滤邮件。。 在这里，我已经从一位专家朋友那里获得了该功能，但是我不知道现在必须使用哪些参数。我应该在这里放哪个令牌向量？

find_token_on_edges <- function(tokens,token = "wechat",edge_window = 5) {
  library(stringr)
  num_tokens <- length(tokens)
  tokens_left <- tokens[1:edge_window]
  if (edge_window * 2 <= num_tokens) {
    tokens_right <- tokens[(num_tokens - edge_window + 1):num_tokens]
  } else if (edge_window < num_tokens) {
    tokens_right <- tokens[(edge_window + 1):num_tokens]
  } else {
    tokens_right <- ""
  }    
  match_token_left <- str_extract(tokens_left,token) %>% 
    na.omit()
  match_token_right <- str_extract(tokens_right,token) %>% 
    na.omit()
  return(c(match_token_left = match_token_left,match_token_right = match_token_right))
}

在此先感谢您的帮助！！！！