我使用regex和stringr和dplyr的解决方案

library(stringr)
library(dplyr)

df <- read.table(
  text = "No; Memo
  1; Date: 2020/10/22 City: UA Note: True mastery of any skill takes a lifetime.
  2; Date: 2022/11/01 City: CH Note: Sweat is the lubricant of success.
  3; Date: 2022y11m1d City: UA Note: Every noble work is at first impossible.
  4; Date: 2022y2m15d City: AA Note: Live beautifully, dream passionately, love completely.",
  sep = ";",
  header = T
)

df_test <- df %>% mutate(date = str_extract(Memo, "(?<=Date: )(.*)(?= City)"),
                         city = str_extract(Memo, "(?<=City: )(.*)(?= Note)"),
                         note = str_extract(Memo, "(?<=Note: ).*")) %>%
  select(-Memo)

> df_test
  No       date city                                                   note
1  1 2020/10/22   UA            True mastery of any skill takes a lifetime.
2  2 2022/11/01   CH                     Sweat is the lubricant of success.
3  3 2022y11m1d   UA               Every noble work is at first impossible.
4  4 2022y2m15d   AA Live beautifully, dream passionately, love completely.

正则表达式匹配使用正向lookahead和loohbehind指定的组之间的所有内容。