如何从地址字符串提取含数字的门牌号并拆分楼宇名称？

To split your address strings into door number and building name columns, we can use a targeted regular expression to capture the two components—even when door numbers include non-numeric characters like letters, prefix words, or hyphens. Here's how to do it in R:

Step-by-Step Explanation #

The core idea is to write a regex pattern that matches all the varied door number formats present in your data, then extract those matches along with the remaining building name. The pattern accounts for:

• Simple numeric door numbers (e.g., "5")
• Alphanumeric numbers (e.g., "22A")
• Hyphenated ranges (e.g., "2-3")
• Prefixed numbers (e.g., "Flat 19", "Room 1", "No 17")

Code Implementation #

First, let's define our input data and regex pattern:

# Input address vector
add <- c("5 Ark Royal House" , "22A Blington Garden Lincoln Street", "Flat 19 PICTON HOUSE" , "2-3 Royal Albert Court" , "Room 1 Grand Hall", "No 17 The Dell Alpha House")

# Regex pattern to capture door number (group 1) and building name (group 2)
pattern <- "^((?:Flat|Room|No)\\s+\\d+|\\d+(?:-\\d+)?|\\d+[A-Za-z])\\s+(.*)$"

Using stringr (Tidyverse)

If you're using the stringr package (part of the tidyverse), str_match makes extracting groups straightforward:

library(stringr)

# Extract matches from each address
matches <- str_match(add, pattern)

# Convert matches to the desired data frame
aim <- data.frame(
  No = matches[, 2],
  Building = matches[, 3],
  stringsAsFactors = FALSE
)

# View the result
print(aim)

Using Base R

If you prefer base R, use regexec and regmatches to achieve the same result:

# Extract matches using base R functions
matches <- regmatches(add, regexec(pattern, add))

# Build the data frame by extracting groups from each match
aim <- data.frame(
  No = sapply(matches, function(x) x[2]),
  Building = sapply(matches, function(x) x[3]),
  stringsAsFactors = FALSE
)

# View the result
print(aim)

Output #

Both methods will produce exactly your expected result:

No                          Building
1       5                  Ark Royal House
2     22A Blington Garden Lincoln Street
3 Flat 19                    PICTON HOUSE
4     2-3              Royal Albert Court
5  Room 1                      Grand Hall
6   No 17            The Dell Alpha House

Customization Tips #

If you encounter other door number formats (like "Suite 45" or "12/34"), simply expand the regex pattern to include those cases. For example, to add support for "Suite" prefixes and slash-separated ranges:

pattern <- "^((?:Flat|Room|No|Suite)\\s+\\d+|\\d+(?:[-/]\\d+)?|\\d+[A-Za-z])\\s+(.*)$"

This updated pattern now handles additional formats like "Suite 45" and "12/34".

内容的提问来源于stack exchange，提问作者Phil Wu