Question

I’ve got a column that has some <p> tags like:

<p>test  test 2</p>    <p>this is a test 2</p>    <p>this is a test 3</p>

What is the best way to always remove the first <p> with corresponding </p>.

So that the result becomes:

test test 2 <p> this is a test 2 </p> <p> this is a test 3</p>

It should work for any case even <p>1</p> it should result in just 1.

I tried using CHARINDEX and SUBSTRING but I end up with a lot of hardcoded #’s :(.

Answer 1

It’s ugly, but it should work. The code basically finds the first instance of </p> and gets everything to the right of it. It also gets everything to the left of it, replacing the first <p> it finds.

DECLARE @x nvarchar(100)
SET @x = '<p>test test 1</p> <p>this is a test 2</p> <p>this is a test 3</p>'

SELECT REPLACE(LEFT(@x, charindex('</p>', @x) - 1), '<p>', '') +  
    RIGHT(@x, len(@x) - charindex('</p>', @x) - 3)

SET @x = '<p>1</p>'

SELECT REPLACE(LEFT(@x, charindex('</p>', @x) - 1), '<p>', '') +  
    RIGHT(@x, len(@x) - charindex('</p>', @x) - 3)

This should return:

test test 1 <p>this is a test 2</p> <p>this is a test 3</p>

And

1

EDIT:

Based on this question here, how about:

SELECT STUFF(STUFF(@x, CHARINDEX('</p>', @x), 4, ''), 
    CHARINDEX('<p>', @x), 3, '')