正则表达式提取Java中不区分大小写的子字符串

Question

I am trying to extract the GRANT number from the a paragraph. The grant number are usually aplhanumeric, with capital alphabets and can have - in between, but they would all occur without any spaces.

Following are some examples of grants:

• W9124A-18-0001
• 007-FY2018
• W81XWH18PRMRPTTDA
• 07-544

Now I am not even certain if a paragraph would have a grant or not, so currently I am relying on the word grant to be present just before the grant number.

---

Example :

This research was supported by NIH/NHLBI Grant W9124A-18-0001(PI, Michael Brown)

I tried to use following regex

(?i)grant [A-Z0-9-]*

but its not perfect and matches Grant w9124A-18-0001 when it shouldn't (lowercase w ). How can I improve it?

Answer 1

You can use the expression:

(?i)(?<=Grant\s)(?-i)[A-Z0-9-]+\b

• (?i) Case insensitive.
• (?<=Grant\\s) Positive lookbehind for Grant followed by whitespace.
• [A-Z0-9-]+ Match digits, alphabetic characters and dashes - .
• (?-i) Turn off case insensitivity.
• \\b Word boundary.

You can try it live here .

Turning on the case sensitivity, instead of disabling insensitivity, with:

(?i)(?<=Grant\s)(?c)[A-Z0-9-]+\b

is only supported by Tcl .

Answer 2

Grant之后，您需要关闭不区分大小写的功能。

(?i)grant (?-i)[A-Z0-9-]*

Answer 3

Fundamentally speaking, you're not accounting for case. Your regex as it stands only looks for "grant" and would fail on "Grant". Worse, your grant identifiers also have mixed case, and your regex isn't checking that , either.

The simplest way to solve this would be to ensure that your regex actually supported those values. You don't need anything too fancy here; just perform a simple matching.

[Gg]rant [A-Za-z0-9\-]+

Fancier matching - such as specific subgroup matching with respect to the hyphen-deliminated parts of the grant ID - are left as an exercise for the reader.