如何在R语言中提取字符串匹配后的字符？附JSON字段提取实例

Got it, let's tackle this problem step by step. You're on the right track with str_extract from the stringr package—we just need to tweak the regex to target exactly what you want.

First, set up the stringr package #

Make sure you have it installed and loaded:

# Install if you haven't already
install.packages("stringr")
# Load the package
library(stringr)

Core Solution: Targeted Regex with str_extract #

Your goal is to extract the 14-digit number immediately following {"blur": ". We can use a positive lookbehind assertion to pinpoint this prefix, then match the 14-digit sequence.

Here's the regex you need:

'(?<={"blur": ")\\d{14}'

Let's break down the regex parts:

• (?<={"blur": "): This is a positive lookbehind—it checks that the text before our target number is exactly {"blur": " (note: in R's single-quoted strings, double quotes don't need escaping).
• \\d{14}: Matches exactly 14 consecutive digits (\\d represents a digit, {14} enforces the length).

Apply this to your data frame like so:

# Replace `df` with your data frame name, `json_column` with your JSON text column
df$extracted_blur <- str_extract(df$json_column, '(?<={"blur": ")\\d{14}')

Test Example to Verify #

Let's create a sample data frame to see how this works:

# Test data
df <- data.frame(json_column = c(
  '{"name": "Bob", "blur": "10010010010010", "age": 30}',
  '{"blur": "98765432109876", "id": 456}',
  '{"status": "active", "note": "No blur value here"}'
))

# Extract the numbers
df$extracted_blur <- str_extract(df$json_column, '(?<={"blur": ")\\d{14}')

# View results
print(df)

You'll get this output:

json_column extracted_blur
1 {"name": "Bob", "blur": "10010010010010", "age": 30} 10010010010010
2                {"blur": "98765432109876", "id": 456} 98765432109876
3           {"status": "active", "note": "No blur value here"}           <NA>

Alternative: Using Capture Groups #

If lookbehind assertions feel tricky, you can use a capture group instead. Wrap the 14-digit sequence in parentheses, then use str_match to pull out the captured value:

# Use str_match to get the captured group
matches <- str_match(df$json_column, '{"blur": "(\\d{14})')
df$extracted_blur <- matches[, 2] # The second column holds the captured number

Quick Notes #

• If your JSON uses single quotes instead of double quotes, adjust the regex to use ' instead of ".
• Observations without the {"blur": " prefix will return NA, which is expected behavior.

内容的提问来源于stack exchange，提问作者Martin M.