regular expression question.

7khat

Ok, lets talk a little bit about eregi_replace:

eregi_replace("([[:punct:]][[:space:]]*)([[:alpha:]])", "\1\2", $string);

Now this part is easy. It gets each punc and space* and assigns it to 1, and alpha as two. In the replace parameter, is replaces them with itself, fine. Doesn't do anything exactly, but it's a leadin to what I want to do:

What if I want to parse these replacements through a function as such:
eregi_replace("([[:punct:]][[:space:]]*)([[:alpha:]])", strtoupper("\1\2"), $string);

But the function reads "\1\2", not their replacements. How do I get it to read their replacements?

What I want to do is just have a regular expression (or something) that capitalises first words in a sentence.

tomhath

What I want to do is just have a regular expression (or something) that capitalises first words in a sentence

Look in the manual at the string functions provided by php. One of them does exactly what you want )I'm too lazy to look up the name of the function right now).

7khat

Ok, will do. I looked through my text books and couldn't find anything there.

For the sake of future purposes, does anyone know if these \1 & \2 etc can be put into a function somehow?

TRJ

This one was quite a challenge! The code still needs a bit of work but it does the job. It assumes a fullstop at the end of the sentence, but you could easily take that out and add one at the end of the code instead.

Hope it helps.

<?php
$sentence = "this is a sentence without capital letters. the idea is to capitalise the first letter of a sentence.";

$splitsentences = explode(".", $sentence);
$count = count($splitsentences)-1; #assuming a fullstop at the end of $sentence.

for ($n=0; $n < $count; $n++) {
$splitsentences[$n] = trim($splitsentences[$n]);
$splitsentences[$n] = ucfirst($splitsentences[$n]);
}

$newsentence = implode(". ", $splitsentences);

print "Old sentence = $sentence<br>";
print "New sentence = $newsentence";
?>