根据匹配的字符串分配站点名称（不分隔地址）-解网

问：

我正在尝试根据一些匹配条件分配 SiteName。我已经使用以下步骤完成了任务，我想知道如果没有“分开”和“pivot_longer”功能，我是否可以获得相同的结果

法典：


df <- tibble::tribble(
  ~Author,                       ~"Author Address",
     "BA",        "METRO NORTH; DALLAS",
     "OA",       "RBWH; TPCH; TOPEND",
     "OB", "CABOOLTURE HOSPITAL; CALIFORNIA UNI",
    "AOS",           "CABOOLTURE HOSPITAL; RBWH",
       "KB",                     "WOMENS HOSPITAL"
  )


df


df %>% 
  separate (`Author Address`, into = c("Site1", "Site2", "Site3"), sep = ";", remove = F ) %>%
  rename (Address = `Author Address` ) %>% 
  pivot_longer (cols = - c(Author, Address), names_to = "Site", values_to = "Author Address" ) %>% 
  mutate (`Author Address`  = str_to_upper(`Author Address`),
          Site = case_when (grepl ("METRO NORTH", `Author Address` ) ~ "Metro North Health",
                            grepl ("CABOOLTURE HOSPITAL", `Author Address` ) ~ "Caboolture Hospital",
                            grepl ("TPCH", `Author Address` ) ~ "Prince charles hospital",
                            grepl ("RBWH", `Author Address` ) ~ "RBWH")) %>% 
  #filter (!is.na (Site)) %>% 
  select (Author, Address, Site ) %>% 
  distinct () %>% 
  group_by (Author, Address) %>% 
  summarise (Site = paste0(Site, collapse = ", "  )) %>% 
  ungroup () %>% 
  mutate (Site = str_trim (str_remove_all (Site, pattern = ", NA$"  ), "right"  ) )

输出：

我已经完成了任务，但希望没有“分离”和“pivot_longer”函数，可能还有一个短代码

r 字符串操作数据整理

感谢您的反馈。我正在尝试消除该功能，因为我必须提供一个分隔符。不幸的是，数据非常肮脏（比我的样本数据复杂得多），而且我无法预测分号总是落在正确的位置，这使得分离非常棘手。我认为在不分离字段的情况下匹配字符串将是最好的方法。separate

答： 暂无答案

上一个：极坐标中的字符串操作

下一个：如何删除出现在另一列中的特定行的第一个单词？

根据匹配的字符串分配站点名称（不分隔地址）

Assign site name based on matching strings (without separating the address)

评论