4.12 高级的替换

相比于 sprintf() 格式化输出字符串的方式替换，它的优势在于提示性，或者说代码的可读性

glue_data <- function(param, text) {
  idx <- gregexpr('\\{[^}]*\\}', text)[[1L]]
  keys <- substring(text, idx, idx + attr(idx, 'match.length') - 1L)
  for (key in keys) {
    text <- gsub(key, param[[gsub('[{}]', '', key)]], text, fixed = TRUE)
  }
  text
}
cat(glue_data(
  param = list(table = 'flights', origin = 'JFK'),
  text = "
  select count(*) as n
  from {table}
  where origin = '{origin}'
  "
))

## 
##   select count(*) as n
##   from flights
##   where origin = 'JFK'
##