如何在R语言中查找字符串冒号前的n个单词？优先用正则表达式

Got it, let's break this down using stringr and regular expressions—super straightforward once you map out the pattern! Here's how to pull the n words before that colon in your example string.

First, let's set up our example string and load the package:

library(stringr)

input_str <- "Input on income economic activities: Small business, self-emp…"

Scenario 1: Extract the last n words before the colon #

This is probably what you're after if you want the closest n words to the colon. We'll use a positive lookahead ((?=:)) to anchor our match right before the colon, then build a pattern to capture n words.

For example, if you want the last 3 words:

n <- 3
# Dynamically build the regex pattern
pattern <- str_glue("(\\b\\w+\\b\\s+){{{n-1}}}\\b\\w+\\b(?=:)")

# Extract the match
str_extract(input_str, pattern)
# Output: "income economic activities"

Let's unpack the regex:

• \b : Word boundary, ensures we grab full words (not partial chunks)
• \w+ : Matches letters, numbers, or underscores (adjust to [a-zA-Z-]+ if your words have hyphens!)
• (?=:) : Positive lookahead—tells regex "only match this if a colon comes right after"
• str_glue : Lets us dynamically adjust the pattern based on n (no manual regex edits needed!)

Scenario 2: Extract the first n words before the colon #

If you want the earliest n words in the text before the colon, we can anchor to the start of the string instead:

n <- 2
pattern <- str_glue("^(\\b\\w+\\b\\s+){{{n-1}}}\\b\\w+\\b")

str_extract(input_str, pattern)
# Output: "Input on"

Bonus: Two-step approach (more readable!) #

If you prefer splitting the task into clear steps, first extract all text before the colon, then pull your n words from that substring:

# Step 1: Grab everything before the colon
before_colon <- str_extract(input_str, "^.+?(?=:)")

# Step 2: Extract last 2 words from that substring
n <- 2
last_n_words <- str_extract(before_colon, str_glue("(\\b\\w+\\b\\s+){{{n-1}}}\\b\\w+\\b$"))
# Output: "economic activities"

# Or extract first 4 words
first_n_words <- str_extract(before_colon, str_glue("^(\\b\\w+\\b\\s+){{{3}}}\\b\\w+\\b"))
# Output: "Input on income economic"

Just tweak the regex character class (\w+) if your words include special characters (like hyphens or apostrophes)—for example, use [a-zA-Z'-]+ to match words like "self-employed" or "Mary's".

内容的提问来源于stack exchange，提问作者elliot