如何提取含前置字符与单位的数值？R语言正则提取问题求助

Got it, let's fix this regex issue for you to extract the full desired strings ("2.6 cm" and "<2mm") instead of just the numeric part or getting NA.

问题根源分析 #

Your original regex had two key limitations:

1. It only targeted the numeric portion (using a positive lookahead for units, so units weren't included in the final match)
2. It didn't account for leading symbols like < that appear before the number, which is why the second string returned NA.

修改后的代码 #

First, ensure you have the stringr package loaded, then adjust the regex pattern to include leading symbols, the full numeric value, optional space, and units:

library(stringr)

res <- c("Diam MM = 2.6 cm.","Mild (<2mm).")
value <- str_extract(res, "[<>]?\\d+\\.?\\d*\\s?(cm|mm)")

# 查看输出结果
print(value)
# [1] "2.6 cm" "<2mm"

正则表达式详解 #

Let's break down the updated regex pattern to understand how it covers your cases:

• [<>]?: Matches an optional leading symbol (we included < and > here; you can add others like ≤, ≥, or ± if your data needs it)
• \\d+\\.?\\d*: Matches numeric values, covering both integers (like 2) and decimals (like 2.6):
  • \\d+: One or more digits before the decimal (if present)
  • \\.?: Optional decimal point
  • \\d*: Optional digits after the decimal point
• \\s?: Matches an optional space between the number and unit (handles both spaced formats like "2.6 cm" and unspaced formats like "<2mm")
• (cm|mm): Matches either the cm or mm unit (add more units here if your data includes others like m or in)

扩展说明 #

If you need to handle more edge cases (like numbers with thousands separators, or additional leading symbols), you can tweak the regex further. For example, to include ± or ≤, update the leading symbol segment to [<>±≤≥]?.

内容的提问来源于stack exchange，提问作者user1828605