Preg_matching help needed...thanks

dbeck83

Hi guys, I need some help on preg_matching the following segment of data:

td class="ghr-c" nowrap>2004</td> <td class="ghr-c" nowrap>2003</td> <td class="ghr-c" nowrap>2002</td> <td class="ghr-c" nowrap>2001</td> <td class="ghr-c" nowrap>2001</td> <td class="ghr-c" nowrap>2000</td> <td class="ghr-c" nowrap>2000</td> <td class="ghr-c" nowrap>2000</td> <td class="ghr-c" nowrap>1999</td> <td class="ghr-c" nowrap>1999</td> <td class="ghr-c" nowrap>1999</td> <td class="ghr-c" nowrap>1998</td> <td class="ghr-c" nowrap>1998</td> <td class="ghr-c" nowrap>1997</td> <td class="ghr-c" nowrap>1997</td> <td class="ghr-c" nowrap>1996</td> <td class="ghr-c" nowrap>1996</td> <td class="ghr-c" nowrap>1995</td>
</tr>

I used the following function:

function scrape ($string){
$pattern = "/nowrap>(.*?)<\/td>/i";
preg_match($pattern, $string, $match);

print_r($match);

}

however, when I print_r the $match array, all I get is one match:

Array ( [0] => nowrap>2004[1] => 2004 )

I am just wondering how I can get all the dates into a $match array so that I can put each into a SQL table. Can you guys please tell me what's wrong with my code or can you guys please suggest me another approach to my problem?

Thanks a lot in advance!!

Andrew

subwayman

preg_match() only returns the first occurance. Either dump it into a loop or take a look at preg_match_all().

dbeck83

I'll look into preg_match_all, but regarding the looping approach, can you please give me a little bit more insights on how to do that please?

Trs2988

The loop approach would work like this: You use preg_match() and find the first occurance of the regular expression. Then you use strstr() to find where it occurs in the string, and then you can call preg_match() again with the offset defined as the value strstr() returned plus the length of the matched string. Repeat until there's no matches left.

My advice? Use preg_match_all(). Much, much, much simplier.

dbeck83

Yes, I've used preg_match_all and it does exactly what I wanted! Thanks a lot. However, I do have one more question regarding my regular expression. I am using:

$pattern = '/>(.*?)<\/td>/i';

since all the data I wanna extract are in the format of "nowap>XXX</td>"

I manage to pick up the single numbers (eg. 2004) and dates (eg. 2004/12/09). But I can't pick up long numbers like 123 456.90 or 123,456.90. Also, I can't pick up any number within brackets like (1234.99). Do I have to include more in my (.*?) expression? If so, can you guys please tell me what is the most generic expression that will ensure me getting all form of data between > and </td>. Thank you very much in advance!!

ANdrew

Trs2988

I don't see anything wrong with your regular expression. It should be picking up anything at all in between the > and the </td>. Could you possibly post a few examples of the larger numbers, with all of the code surrounding them (like you did for the single numbers in your original reply)? Maybe then I'll be able to spot the problem.

dbeck83

</div></td>

				<td style="" class="evenrow ndr nowrap">113 770.14	
				</td>
				<td style="" class="evenrow ndr nowrap">105 308.24	
				</td>
				<td style="" class="evenrow ndr nowrap">95 921.23	
				</td>
				<td style="" class="evenrow ndr nowrap">98 143.63	
				</td>
				<td style="" class="evenrow ndr nowrap">101 450.65	
				</td>
				<td style="" class="evenrow ndr nowrap">100 532.62	
				</td>
				<td style="" class="evenrow ndr nowrap">104 439.84	
				</td>
				<td style="" class="evenrow ndr nowrap">104 439.84	
				</td>
				<td style="" class="evenrow ndr nowrap">103 437.93	
				</td>
				<td style="" class="evenrow ndr nowrap">103 437.93	
				</td>
				<td style="" class="evenrow ndr nowrap">103 437.93	
				</td>
				<td style="" class="evenrow ndr nowrap">96 489.53	
				</td>
				<td style="" class="evenrow ndr nowrap">96 489.53	
				</td>
				<td style="" class="evenrow ndr nowrap">92 757.17	
				</td>
				<td style="" class="evenrow ndr nowrap">92 757.17	
				</td>
				<td style="" class="evenrow ndr nowrap">89 731.35	
				</td>
				<td style="" class="evenrow ndr nowrap">89 731.35	
				</td>
				<td style="" class="evenrow ndr nowrap">84 997.08	
				</td>
		</tr>

This is the result I get: Array ( [0] => Array ( [0] => > ) [1] => Array ( [0] => ) )

And this is the scraping code I used:

function scrape ($string){
$pattern = '/>(.*?)<\/td>/i';
preg_match_all($pattern, $string, $match);

print_r($match);

Thank you once again!

Trs2988

The problem is that there are line breaks after the numbers, before the </td>. In order to fix that, use a couple of str_replace's right before the preg_match_all line.

$string = str_replace("\n","",$string);
$string = str_replace("\r","",$string);

That should work.

dbeck83

Thank you very much...I must be blinded to have not spotted that myself 😃 I am such a noob to this hehe

I've added a str replace line to eliminate the space seperator in the number as well!

O, one very last question, do you know how I can export these into a date field in mySQL?

I believe the format for mySQl date is yyyy-mm-dd? Can I just use string replace or would that also replace the "/" in </td>? Any suggestions on what's the best way to convert this?

Thank you 🙂

dbeck83

I just got the solution myself...

I just str replaced all / to - and changed the pattern to > xxx <-td>

xblue

Trs2988 wrote:
In order to fix that, use a couple of str_replace's right before the preg_match_all line.

Or (might be simpler) set the s modifier. This makes the dot match newline characters which it does not do by default.

Trs2988

xblue wrote:
Or (might be simpler) set the s modifier. This makes the dot match newline characters which it does not do by default.

Well, the only problem with that is that he's trying to just match the numbers. Matching the numbers and a newline after them could cause problems.

dbeck83

Hey guys, is it possible to preg_match and get a specific number match? For example, can I look for the 2nd match, or the 5th match specifically by indicating some sort of offset?

Thanks 🙂

Trs2988

Well, if you use preg_match_all, then the first match would be in offset 1 in the $matches array, second would in $matches[2], etc.