need a string for regular expression to preg_split records into array

Question

$source="<p><b>Lal, Vaninm</b></p>
<p><b>Vice President &amp;</b></p>
<p><b>General Manager</b></p>
<p>Company 1 Inc.</p>
<p>PO Box 123456</p>
<p>salt Lake1, 00111-3333</p>
<p>111-111-111 / F: 111-111-111</p>
<p>info1@site1.com</p>
<p><b>Andrus, Reed </b></p>
<p><b>Manager</b></p>
<p>Company 2 Inc.</p>
<p>Monada, Suite 222</p>
<p>J , Lousiana 2222</p>
<p>222-222-222 / F: 222-222-222</p>
<p>info2@site2.com</p>
<p><b>Sharma, John L.</b></p>
<p><b>Senior Property Manager</b></p>
<p>Company 3  Ltd.</p>
<p>PO Box 3333</p>
<p>Grand Cinema, Layman Islands</p>
<p>FGB 333</p>
<p>333-333-333</p>
<p>info3@site3.com</p>
<p><b>Lucky, Philip S</b></p>
<p>Life Member</p>
<p>Company 4 Inc.</p>
<p>Battelsville, Oklahoma 74000</p>
<p>444-444-444</p>
<p><b>Berry, Richard B, RPA, CPM</b></p>";
$records = preg_split ("@\<p\>\<b\>(.*?)(\<p\>(.*)\</p\>\<p\>\<b\>)@s", $source); 
var_dump($records);

The array must contain four records. The data contained inside tags are meaningless. I am new to regular expression. I tried as above. Please suggest regular expressions for this. Thanks in advance.

I think  ....... identifies a record. But I cant make the required expression.

If you're working with markup, don't use regexp, use [DOM](http://www.php.net/manual/en/book.dom.php) — Mark Baker, Jun 14 '14 at 11:31
Hi Mark, I tried simple html dom for this. But html file too large. — Romi, Jun 14 '14 at 11:38
http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags — dognose, Jun 14 '14 at 12:30

score 0 · Accepted Answer · answered Jun 14 '14 at 22:41

With all the disclaimers about parsing html with regex, the following regex will correctly split your input.

Version 1: file with only newlines (unix, osx)

(?=(?<=^|((?<!</b>)</p>\n))<p><b>)

Version 2: file with carriage returns and newlines (windows)

(?=(?<=^|((?<!</b>)</p>\r\n))<p><b>)

Therefore, if you were using the first, you could write:

$records = preg_split('~(?=(?<=^|((?<!</b>)</p>\n))<p><b>)~', $str);

Note that there are actually five records because of the last line:

<p><b>Berry, Richard B, RPA, CPM</b></p>";

How does it work?

With lookahead and lookbehind. This is a "zero-width" match that just looks for a certain position.

The (?= lookahead asserts that the current position that is followed by ...
as long as  is preceded by (lookbehind (?<= ) the beginning of the string ^ or \n that is not preceded by  (negative lookbehind (?<!))

Enjoy!

@Romi Glad it works, thanks for letting me know. :) FYI I already upvoted your interesting question when I answered and can't do it again, you can't see it but the question score is -4/+1 :( — zx81, Jun 16 '14 at 11:32

need a string for regular expression to preg_split records into array

1 Answers1