如何用一个表达式替换regexp载体？

Question

我的材料里有一些类似于100+不同村庄的东西。为了让我的想象有意义，我需要把它们组合成22个城市，如下所示：

TROLLHÄTTAN<-toupper(c("Trollhättan","Sjuntorp","Velanda","Åsaka","Upphärad"))
UDDEVALLA<-toupper(c("UDDEVALLA","KURVERÖD","AMMENÄS","FAGERHULT","LANESUND OCH ÖVERBY",
"LANESUND","ÖVERBY","RESTENÄS OCH ULVESUND","RESTENÄS","ULVESUND","STRAND","UTBY","HOGSTORP","SUND","SMEDSERÖD"))
VÄNERSBORG<-toupper(c("Vänersborg","Vargön","Brålanda","Frändefors","Nordkroken","Katrinedal"))
LYSEKIL<-toupper(c("Lysekil", "Brastad", "Grundsund", "Fiskebäckskil"))
FÄRGELANDA<-toupper(c("Färgelanda","Högsäter","Ödeborg","Stigen"))
MELLERUD<-toupper(c("Mellerud","Dals Rostock","Åsensbruk"))
ED<-toupper(c("Ed"))
BENGTSFORS<-toupper(c("Bengtsfors","Dals Långed","Billingsfors","Bäckefors","Skåpafors"))
ÅMÅL<-toupper(c("Åmål","Tösse","Fengersfors"))
STRÖMSTAD<-toupper(c("Strömstad","Skee","Kebal","Stare"))
TANUM<-toupper(c("Grebbestad","Tanumshede","Fjällbacka","Hamburgsund","Rabbalshede"))
SOTENÄS<-toupper(c("Hunnebostrand","Kungshamn","Smögen","Malmön","Bovallstrand"))
MUNKEDAL<-toupper(c("Munkedal","Dingle","Hällevadsholm","Hedekas","Torreby"))
ORUST<-toupper(c("Svanesund","Ellös","Hälleviksstrand","Mollösund","Henån","Höggeröd","Vindön och Töllås","Varekil","Vindön","Töllås"))
LILLA_EDET<-toupper(c("Lilla Edet","Lödöse","Lilla Edet västra","Göta","Nygård","Hjärtum"))
ALE<-toupper(c("Ale","Nödinge-Nol","Surte","Älvängen","Skepplanda","Alvhem"))
STENUNGSUND<-toupper(c("Jörlanda","Stora Höga","Timmervik","Spekeröd","Stenungsund","Stenungsön","Svartehallen","Svenshögen","Ucklum","Ödsmål"))
TJÖRN<-toupper(c("Bleket","Djupvik och Fagerfjäll","Höviksnäs","Klövedal","Kyrkesund och Bö","Kållekärr","Myggenäs","Rönnäng","Skärhamn","Stora Dyrön",
"Djupvik","Fagerfjäll","Kyrkesund","Bö"))
KUNGÄLV<-toupper(c("Aröd och Timmervik","Diseröd","Duvesjön","Harestad och Nereby","Kareby","Kode","Kovikshamn","Kungälv","Kärna",
"Lundby","Marstrand","Marstrand", "Arvidsvik","Risby","Rishammar","Signehög och Norrmannebo","Solberga","Tjuvkil","Ödsmål och Åseby",
"Ödsmåls mosse och Rörtången","Aröd","Timmervik","Harestad","Nereby","Signehög","Norrmannebo","Ödsmål","Åseby","Ödsmåls mosse","Rörtången"))
ALINGSÅS<-toupper(c("Alingsås","Ingared","Sollebrunn","Västra Bodarna","Gräfsnäs","Hemsjö","Stora Mellby","Hjälmared","Långared","Svanvik",
"Ryd","Magra"))
VARA<-toupper(c("Vara","Kvänum","Tråvad","Jung","Vedum","Larv","Stora Levene","Emtunga","Arentorp"))
ESSUNGA<-toupper(c("Nossebro","Främmestad","Jonslund"))
VÅRGÅRDA<-toupper(c("Vårgårda","Östadkulle","Horla"))
GRÄSTORP<-toupper(c("GRÄSTORP"))
LIDKÖPING<-toupper(c("Lidköping","Lidköping norra","Vinninga","Järpås","Filsbäck","Örslösa","Saleby"))
GÖTEBORG<-toupper(c("Göteborg","Gunnared och Hammarkullen","Torslanda","Billdal","Olofstorp","Donsö","Nolvik","Styrsö","Angered",
"Brännö","Säve","Helgered","Tumlehed","Asperö","Stenared","Vrångö","Gundal och Högås","Gunnared","Hammarkullen","Gundal","Högås"))

我试图通过mgsub (文本清洁版)将村名向量替换为市政名称，但我遇到了一些问题。例如，"strand“(一个村庄的名字)也是一个流行的后缀。这意味着"HUNNEBOSTRAND“被转换为"HUNNEBOUDDEVALLA”，这当然不是最优的。

我尝试用正则表达式来编写向量：

LYSEKIL<-toupper(c("^Lysekil$", "^Brastad$", "^Grundsund$", "^Fiskebäckskil$"))

我发现文本干净的mgsub版本无法处理正则表达式。我换到mgsub包，它希望我的向量长度相同，这不是我想要的。mgsub的qdap版本似乎以类似的方式运行。

有办法绕过这件事吗？

原始数据(删除敏感部分)

结构(城市=c(“HEN”、NA、“HEN”、“LV NGEN”、NA、“TROLLH”)、ZIPCODE =c(47395 L、NA、47332 L、44636L、NA、46157 L)、COURSEOFFERING_ID =c(97113 L、97113 L)、row.names = c(1L、5L、9L、12L、15L、18L)，类= "data.frame")

Answer 1

为了避免村庄名称也是后缀的问题，您可以用^和$锚定这些名称的开头和结尾(您有正确的想法)。但是，要将村庄名称替换为相应市政当局的名称，您需要使用函数gsub() (或stringr::str_replace_all())。为了安全起见，不必担心哪些名字会成为问题，只需用^和$锚定所有的村庄名称。

以下是一个选择：

1. 创建一个带有100+村庄名称的向量(我使用的是前两个向量)：

all_village_names <- c(
  "Trollhättan", "Sjuntorp", "Velanda", "Åsaka", "Upphärad",  "UDDEVALLA",
  "KURVERÖD", "AMMENÄS", "FAGERHULT", "LANESUND OCH ÖVERBY", "LANESUND",
  "ÖVERBY", "RESTENÄS OCH ULVESUND", "RESTENÄS", "ULVESUND", "STRAND",
  "UTBY", "HOGSTORP", "SUND", "SMEDSERÖD"
)

1. 使用一系列的gsub()替换它们对应的市政名称(这里，我再次使用您的前2行代码)：

library(dplyr)

all_village_names %>%
  gsub("^Trollhättan$|^Sjuntorp$|^Velanda$|^Åsaka$|^Upphärad$", "TROLLHÄTTAN", .) %>%
  gsub("^UDDEVALLA$|^KURVERÖD$|^AMMENÄS$|^FAGERHULT$|^LANESUND OCH ÖVERBY$|^LANESUND$|^ÖVERBY$|^RESTENÄS OCH ULVESUND$|^RESTENÄS$|^ULVESUND$|^STRAND$|^UTBY$|^HOGSTORP$|^SUND$|^SMEDSERÖD$", "UDDEVALLA", .)

您将得到一个100+元素的向量(每个初始村庄都有一个)，但是该向量将仅由22个市政名称组成。

对于我使用的示例数据，这将给您提供：

[1] "TROLLHÄTTAN" "TROLLHÄTTAN" "TROLLHÄTTAN" "TROLLHÄTTAN" "TROLLHÄTTAN"
 [6] "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"  
[11] "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"  
[16] "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"   "UDDEVALLA"  

在没有锚定的情况下，LANESUND被转换为LANEUDDEVALLA (因为SUND被转换为UDDEVALLA)。但锚定阻止了这一点。