This question already has answers here:
Warning: preg_replace(): Unknown modifier
(3 answers)
Closed 3 years ago.
Does anyone knows why i receive this error : preg_match() [function.preg-match]: Unknown modifier '('
using this method:
function checkFBDateFormat($date) {
if(preg_match ("/^([0-9]{2})/([0-9]{2})/([0-9]{4})$/", $date, $parts)){
if(checkdate($parts[2],$parts[1],$parts[3]))
return true;
else
return false;
} else {
return false;
}
}
You did not escape your "/" and you didn't complete your if statements either, please try this:
function checkFBDateFormat($date) {
if(preg_match("/^([0-9]{2})\/([0-9]{2})\/([0-9]{4})$/", $date, $parts)){
if(checkdate($parts[2],$parts[1],$parts[3])) {
return true;
} else {
return false;
}
} else {
return false;
}
}
echo var_dump(checkFBDateFormat('08/09/2012'));
If the first char is e.g. an slash / is detected as delimiter fro the regular expression. Thus your regex is only the part ^([0-9]{2}). And everything after the second slash is recognized as modifiers for the regex.
If you really want to match a slash, use \/ to escape it
Since you are using slash in regular expression, need use other delimiter, try:
preg_match ("#^([0-9]{2})/([0-9]{2})/([0-9]{4})$#", $date, $parts)
You need to escape your slash, like so:
if(preg_match ("/^([0-9]{2})\/([0-9]{2})\/([0-9]{4})$/", $date, $parts)){
You use / as delimiter for your expression. However, it's completely unnecessary anyway
$parts = explode('/', $date);
Even better: http://php.net/datetime.createfromformat
To give you an idea what happens: PCRE regular expression require a delimiter at the start and the end of the pattern itself. Everything after the second delimiter is treated as modifier. Thus you decided to use / as delimiter (it's always the first character), so your pattern ended right after /^([0-9]{2})/. Everything next (which is a ( at first) is treated as modifier, but ( is not an existing modifier.
If you want to stay with regular expression, I recommend to use another delimiter like
~^([0-9]{2})/([0-9]{2})/([0-9]{4})$~
#^([0-9]{2})/([0-9]{2})/([0-9]{4})$#
Just read the manual about the PCRE-extension
Two additional comments:
You should define $parts, before you use it
Remember, that the expression is quite inaccurate, because it allows dates like 33/44/5678, but denies 1/1/1970
You might want to consider not using regular expressions at all.
<?php
// simple example
$timestamp = strtotime('12/30/2012');
if ($timestamp) {
// valid date… Now do some magic
echo date('r', $timestamp);
}
Related
This question already has answers here:
My regex is matching too much. How do I make it stop? [duplicate]
(5 answers)
Closed 2 years ago.
I have to modify an URL like this:
$string = "/st:1/sc:RsrlYQhSQvs=/fp:1/g:3/start:2015-07-01/end:2015-07-30";
Namely, I want to delete st:1 with a regex. I used:
preg_replace("/\/st:(.*)\//",'',$string)
but I got
end:2015-07-30
while I would like to get:
/sc:RsrlYQhSQvs=/fp:1/g:3/start:2015-07-01/end:2015-07-30
Same if I would like to delete fp:1.
You can use:
$string = preg_replace('~/st:[^/]*~','',$string);
[^/]* will only match till next /
You are using greedy matching with . that matches any character.
Use a more restricted pattern:
preg_replace("/\/st:[^\/]*/",'',$string)
The [^\/]* negated character class only matches 0 or more characters other than /.
Another solution would be to use lazy matching with *? quantifier, but it is not that efficient as with the negated character class.
FULL REGEX EXPLANATION:
\/st: - literal /st:
[^\/]* - 0 or more characters other than /.
You need to add ? in your regex:-
<?php
$string = "/st:1/sc:RsrlYQhSQvs=/fp:1/g:3/start:2015-07-01/end:2015-07-30";
echo preg_replace("/\/st:(.*?)\//",'',$string)
?>
Output:- https://eval.in/397658
Based on this same you can do for next things also.
Instead of using regex here you should make parsing utility functions for your special format string, they are simple, they don't take to long to write and they will make your life a lot easier:
function readPath($path) {
$parameters = array();
foreach(explode('/', $path) as $piece) {
// Here we make sure we have something
if ($piece == "") {
continue;
}
// This here is just a fancy way of splitting the array returned
// into two variables.
list($key, $value) = explode(':', $piece);
$parameters[$key] = $value;
}
return $parameters;
}
function writePath($parameters) {
$path = "";
foreach($parameters as $key => $value) {
$path .= "/" . implode(":", array($key, $value));
}
return $path;
}
Now you can just work on it as a php array, in this case you would go:
$parameters = readPath($string);
unset($parameters['st']);
$string = writePath($parameters);
This makes for much more readable and reusable code, additionally since most of the time you are dealing with only slight variations of this format you can just change the delimiters each time or even abstract these functions to using different delimiters.
Another way to deal with this is to convert the string to conform to a normal path query, using something like:
function readPath($path) {
return parse_str(strtr($path, "/:", "&="));
}
In your case though since you are using the "=" character in a url you would also need to url encode each value so as to not conflict with the format, this would involve similarly structured code to above though.
I am building an application that processes some transactions with descriptions such as:
cash bnkm aug09 : bt phone / a#19:10
I need to extract the time from such strings using a regular expression using preg_match.
I have written a regex that works on this website:
http://www.switchplane.com/awesome/preg-match-regular-expression-tester?pattern=%2F%40%5Cd%5Cd.%5Cd%5Cd%2F&subject=cash+bnkm++++aug09+%3A+bt+phone+%2F+a%4019%3A10
But not in my code, as a, Warning: preg_match(): Unknown modifier '#' error is given.
The regex I am using is: /#\d\d.\d\d/ and the call to preg_match is as below:
$matches = array();
if (preg_match('/'.'/#\d\d.\d\d/'.'/', 'cash bnkm aug09 : bt phone / a#19:10', $matches) === 1) {
//Carry out trigger
echo 'yes';
} else {
echo 'no';
}
Essentially it seems to choke on the # symbol - which I can't seem to escape.
I am quite new to regular expressions but can't find any description of # being used for as a special character so would have assumed it would be used to test the string rather than alter how the string was tested.
Any help on how to fix this would be appreciated.
try this:
if (preg_match('/#\d\d:\d\d/', 'cash bnkm aug09 : bt phone / a#19:10', $matches) === 1) {
//Carry out trigger
echo 'yes';
print_r($matches);//Array ( [0] => #19:10 )
} else {
echo 'no';
}
in your pattern there are two extra slashes:
'/'.'/#\d\d.\d\d/'.'/'
^-- ^--
If # not used as delimiter, then no need to escape it
preg_match("/\#\d\d:\d\d/ims", "<input string>", $matches)
Use \ to escape special characters, the # is a delimiter.
My current regular expression should be correct, though I wouldn't expect so, it doesn't work properly. It won't return "Got Match"
My currrent code is as follows:
$id = "http://steamcommunity.com/id/TestID";
if (preg_match("^http://steamcommunity\.com/id/.*?\n$", $id)) {
print "Got match!\n";
}
You're missing delimiters on your regex:
if (preg_match("#^http://steamcommunity\.com/id/.*?\n$#", $id)) {
^--here ^--here
Note that I've used # as the delimiter here, since that saves you having to escape all of the internal / charrs, if you'd used the traditional / as the delimiter.
You need a delimiter, like this:
if (preg_match("#^http://steamcommunity\.com/id/.*?$#", $id)) {
^ ^
And what's with the newline at the end? Surely you don't need that.
You're missing delimiters. For example:
"#^http://steamcommunity\.com/id/.*?\n$#"
Also, you're trying to match a newline (\n) that isn't in your string.
You need to add the pattern delimiter:
$id = "http://steamcommunity.com/id/TestID";
if (preg_match("#^http://steamcommunity\.com/id/.*?(\n|$)#", $id)) {
print "Got match!\n";
}
There are a couple of things that are wrong with it. First of all, you need to delimit the start and end of your regex with a character. I used #. You're also matching for a new line at the end of your regex, which you don't have and likely won't ever have in your string.
<?php
$id = "http://steamcommunity.com/id/TestID";
if (preg_match("#^http://steamcommunity\.com/id/.*?$#", $id)) {
print "Got match!\n";
}
?>
http://codepad.viper-7.com/L7XctT
First of all, your regex shouldn't even compile because it's missing delimiters.
if (preg_match("~^http://steamcommunity\.com/id/.*?\n$~", $id)) {
^---- these guys here -----^
Second of all, why do you have a \n if your string doesn't contain a new line?
And finally, why are you using regex at all? Effectively, you are just trying to match a constant string. This should be equivalent to what you are trying to match:
if (strpos($id, 'http://steamcommunity.com/id/') === 0) {
You need to have starting and ending delimiter in your pattern like /pattern/ or #pattern# or with brackets (pattern). Why is that? To have some pattern modifiers after ending delimiter like #pattern#i (ignore case)
preg_match('(^http://steamcommunity\.com/id/.*?\n$)', $id)
As the say your patten is start and end wrong. (Delimiter)
But this will be a better match of a 64-bit Steam ID. (Minimum 17 and Maximum 25 numbers)
if( preg_match("#^http://steamcommunity\.com/id/([0-9]{17,25})#i", $id, $matches) )
{
echo "Got match! - ".$matches;
}
I believe that there is no need for you to require that the string must end with a line break.
Explanation.
http://steamcommunity\.com/id/([0-9]{17,25})
^--- string ---^^-- Regexp --^
[0-9] - Match a number between 0 to 9
{17,25} - Make 17 to 25 matches
() - Returns match
Or use pattern as those (It is the same):
/^http:\/\/steamcommunity\.com\/id\/([0-9]{17,25})/i
(^http://steamcommunity\.com/id/([0-9]{17,25}))i
Regular Expressions PHP Tutorial
Online regular expression testing <- Dont use delimiter.
<?php
# URL that generated this code:
# http://txt2re.com/index-php.php3?s=http://steamcommunity.com/id&-1
$txt='http://steamcommunity.com/id';
$re1='(http:\\/\\/steamcommunity\\.com\\/id)'; # HTTP URL 1
if ($c=preg_match_all ("/".$re1."/is", $txt, $matches))
{
$httpurl1=$matches[1][0];
print "($httpurl1) \n";
}
#-----
# Paste the code into a new php file. Then in Unix:
# $ php x.php
#-----
?>
Resorces:
http://txt2re.com/index.php3?s=http://steamcommunity.com/id&-1
I'm going through my code to replace any instances I'm using the ereg() function - which I was using for matching regex inside a string.
I could use a little direction, if someone has a better method than what I'm using.
Here's my old "currency validation" script:
function valid_currency($number){
if(ereg('^[0-9]+\.[0-9]{2}$', $number))
return true;
else
return false;
}
if(valid_currency(25.30)){
echo "valid currency";
}else{
echo "invalid currency string";
}
I replaced the ereg() with preg_match().
I'm getting this error now:
Warning: preg_match() [function.preg-match]: No ending delimiter '^'
I'm guessing the regular expression syntax isn't being recognized. From here I'm a little stuck.
preg requires delimiters around your regex. It can be almost anything though traditionally it's /. This should work:
preg_match('/^[0-9]+\.[0-9]{2}$/', $number)
You need the bounds for the regex statement, IE it's thinking you're starting the statement with ^ and saying there's no ending ^ indicating the end of the statement. Use /^[0-9]+\.[0-9]{2}$/ instead
in preg_match() or preg_match_all() uses "delimiters".examples:
/regEx/ or #regEx# or |regEx| or #regEx#.
I use: /regEx/
function valid_currency($n) {
return preg_match("/^\d+\.\d{2}$/", $n);
}
returns bool: 1(true) 0(false)
exmaple of use:
echo valid_currency("25.30") ? 'valid currency' : 'invalid currency string';
The code below gives me this mysterious error, and i cannot fathom it. I am new to regular expressions and so am consequently stumped. The regular expression should be validating any international phone number.
Any help would be much appreciated.
function validate_phone($phone)
{
$phoneregexp ="^(\+[1-9][0-9]*(\([0-9]*\)|-[0-9]*-))?[0]?[1-9][0-9\- ]*$";
$phonevalid = 0;
if (ereg($phoneregexp, $phone))
{
$phonevalid = 1;
}else{
$phonevalid = 0;
}
}
Hmm well the code you pasted isn't quite valid, I fixed it up by adding the missing quotes, missing delimiters, and changed preg to preg_match. I didn't get the warning.
Edit: after seeing the other comment, you meant "ereg" not "preg"... that gives the warning. Try using preg_match() instead ;)
<?php
function validate_phone($phone) {
$phoneregexp ='/^(\+[1-9][0-9]*(\([0-9]*\)|-[0-9]*-))?[0]?[1-9][0-9\- ]*$/';
$phonevalid = 0;
if (preg_match($phoneregexp, $phone)) {
$phonevalid = 1;
} else {
$phonevalid = 0;
}
}
validate_phone("123456");
?>
If this is PHP, then the regex must be enclosed in quotes. Furthermore, what's preg? Did you mean preg_match?
Another thing. PHP knows boolean values. The canonical solution would rather look like this:
return preg_match($regex, $phone) !== 0;
EDIT: Or, using ereg:
return ereg($regex, $phone) !== FALSE;
(Here, the explicit test against FALSE isn't strictly necessary but since ereg returns a number upon success I feel safer coercing the value into a bool).
It's the [0-9\\- ] part of your RE - it's not escaping the "-" properly. Change it to [0-9 -] and you should be OK (a "-" at the last position in a character class is treated as literal, not part of a range specification).
Just to provide some reference material please read
Regular Expressions (Perl-Compatible)
preg_match()
or if you'd like to stick with the POSIX regexp:
Regular Expression (POSIX Extended)
ereg()
The correct sample code has already been given above.