Selecting words/patterns from sentences

[soltek]

Hello,

 

 

I'm looking for a way to filter a couple of words from one sentence for example:

«There was a man who had 13 legs and looked nice»

«My pocket had 41 coins before I spent them»

 

I want the sentences to be separated from before the «had [number]», so it becomes:

«There was a man»

«My pocket»

 

And also to select everything starting on the «had» to the number, so it'd become:

«had 13»

«had 41»

 

Any idea, mates?

Thank you.

[fugix]

there are several ways to do this..if you know what your string is going to be..you can use trim() to trim the words that you don't want

[Psycho]

You contradicted yourself. You said you wanted everything before the word "had", but your first expected result was "«There was a man»". What about the word "who"?

 

But based upon your explanation, what you want is fairly simple:

function splitString($string)
{
    if(preg_match("#(.*?)(had \d*)#", $string, $matches) > 0)
    {
        unset($matches[0]);
        return $matches;
    }
    //Return false if there was no match
    return false;
}

$parts = splitString('There was a man who had 13 legs and looked nice');
print_r($parts);

$parts = splitString('My pocket had 41 coins before I spent them');
print_r($parts);

 

Output

Array
(
    [1] => There was a man who 
    [2] => had 13
)
Array
(
    [1] => My pocket 
    [2] => had 41
)

[soltek]

You're right, mjdamato.

And thanks!

 

That function seems pretty awesome, can you give me some links that explain this part - "#(.*?)(had \d*)#" ?

I'd like to understand them for the next I need something similar.

[soltek]

Anyone about the last question? 

[Maq]

That function seems pretty awesome, can you give me some links that explain this part - "#(.*?)(had \d*)#" ?

I'd like to understand them for the next I need something similar.

Check out:

http://www.phpfreaks.com/tutorial/regular-expressions-part1---basic-syntax

[soltek]

Looks fun and useful, thank you both.

[Maq]

Looks fun and useful, thank you both.

Yes, regular expressions are very useful and something that you should understand the basics of.  We have a PHP Regex section in case you need help with them.

[Psycho]

That function seems pretty awesome, can you give me some links that explain this part - "#(.*?)(had \d*)#" ?

I'd like to understand them for the next I need something similar.

 

That is a "Regular Expression" I can explain THAT regular expression, but to really understand them will take quite a bit of time for you to research. In my opinion, Regular Expressions are one of the most powerful features in programming, but also one of the hardest to master.

 

In the above regular expression the hash marks (#) are just delimiters and do not define anything to match on. The "(.*?)" searches for ANY characters. The period is a wildcard for any character, the * means to match 0 to many occurrences of any character and the ? makes the match non-greedy. That means it will stop matching as soon as a match is found for the next parameter.

 

The next parameter is "(had \d*)" which will match the letters "had" followed by a space. Immediately following the space there must be o or more numbers (the wildcard d is used to match digits and the * matches 0 or more). On second though I should have used a + which will match one or more occurrences.

[soltek]

I went to a coffee shop and took the example given + the information on the first link to understand, or try to, and... jeez, I had no idea. Looks really powerful, organized and precise. I read the basic regular expressions and that helped to understand this part (had \d*). And yeah, I got kinda confused 'cause the + would have made more sense

If I have #had \d{2}+# -> It would match only «had [two_digits_number]», correct?

 

 

But I didnt get the first part: (.*?)

I mean, using the link you guys shared with me, as a guide, I understood what it means in word, but why do I need to select everything?

And also, how did the «There was a man who» got matched?

 

 

[Psycho]

If I have #had \d{2}+# -> It would match only «had [two_digits_number]», correct?

 

Actually, I'm not sure how that would be interpreted. You don't need to the '+'. The '+' as a modifier means to match one or more occurrences. But, you are already using '{2}' which means to match exactly two occurrences. So, if you removed the '+', then - yes - it would match 'had [two_digits_number]'

 

But I didnt get the first part: (.*?)

 

Hmm, I thought I explained that. The period is a wild card for any character. The asterisk is a modifier that say match 0 or more occurrences. By default the '*' is greedy which means it will match as many instances as it can. Even digits are "any character4" so the * would make it match everything to the end of the string. But, you only want to match to where the word 'had' starts. So we add the '?' to make the '*' non-greedy, so it will match as few characters as possible. In this case it will match every character up to the word 'had'.