I'm having some difficulty with preg_match. I'm trying to match roman numerals, like this:
$string='This is roman XI and some other ones: XMCIII, like this.XXVIII'."\n";
preg_match('/(\s|\.)M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})\s/',$string,$matches);
print_r($matches);
It should match any roman numeral preceded with whitespace or period and ending with whitespace. But it returns the following:
Array
(
[0] => XI
[1] =>
[2] =>
[3] => X
[4] => I
)
You have {0, 4} or {0,3} ranges in regex which means that those parts are optional. You get spaces because space[nothing]space becomes a valid match.
You can simply filter out the empty space results from your array using array_filter
Related
I tried multiple time to make a pattern that can validate given string is natural number and split into single number.
..and lack of understanding of regex, the closest thing that I can imagine is..
^([1-9])([0-9])*$ or ^([1-9])([0-9])([0-9])*$ something like that...
It only generates first, last, and second or last-second split-numbers.
I wonder what I need to know to solve this problem.. thanks
You may use a two step solution like
if (preg_match('~\A\d+\z~', $s)) { // if a string is all digits
print_r(str_split($s)); // Split it into chars
}
See a PHP demo.
A one step regex solution:
(?:\G(?!\A)|\A(?=\d+\z))\d
See the regex demo
Details
(?:\G(?!\A)|\A(?=\d+\z)) - either the end of the previous match (\G(?!\A)) or (|) the start of string (^) that is followed with 1 or more digits up to the end of the string ((?=\d+\z))
\d - a digit.
PHP demo:
$re = '/(?:\G(?!\A)|\A(?=\d+\z))\d/';
$str = '1234567890';
if (preg_match_all($re, $str, $matches)) {
print_r($matches[0]);
}
Output:
Array
(
[0] => 1
[1] => 2
[2] => 3
[3] => 4
[4] => 5
[5] => 6
[6] => 7
[7] => 8
[8] => 9
[9] => 0
)
I want to use preg_match to parse '123,456,789,323' and only capture each number into arrray $m.
My php codes:
preg_match("/^(\d+)(?:,(\d+))*?$/",'123,456,789,323',$m));
print_r($m);
This is how I interpret my regexp:
^: Begin of line
1st (\d+): Capture 1st number
,(\d+): Match pattern 'a command then a number'.
(?:,(\d+))*?: Match zero or more [using *] of above pattern but don't
capture whole pattern [using ?:] instead only capture
the number [using (\d+)]. Lastly, match pattern
nongreedy [using last ?]
$: Match end of line.
But I get this output:
Array
(
[0] => 123,456,555,789,323
[1] => 123
[2] => 323
)
What I want is:
Array
(
[0] => 123,456,555,789,323
[1] => 123
[2] => 456
[3] => 789
[4] => 323
)
I thought (...)* is too greedy, so I use (...)*?. But it doesn't approve the output. What do I miss?
PS: I want to know how can regexp do this rather than use other way e.g. explode().
Let's take an example of following string:
$string = "length:max(260):min(20)";
In the above string, :max(260):min(20) is optional. I want to get it if it is present otherwise only length should be returned.
I have following regex but it doesn't work:
/(.*?)(?::(.*?))?/se
It doesn't return anything in the array when I use preg_match function.
Remember, there can be something else than above string. Maybe like this:
$string = "number:disallow(negative)";
Is there any problem in my regex or PHP won't return anything? Dumping preg_match returns int 1 which means the string matches the regex.
Fully Dumped:
int 1
array (size=2)
0 => string '' (length=0)
1 => string '' (length=0)
You're using single character (.) matching in the case of being lazy, at the very beginning. So it stops at the zero position. If you change your preg_match function to preg_match_all you'll see the captured groups.
Another problem is with your Regular Expression. You're killing the engine. Also e modifier is deprecated many many decades before!!! and yet it was used in preg_replace function only.
Don't use s modifier too! That's not needed.
This works at your case:
/([^:]+)(:.*)?/
Online demo
I tried to prepare a regex which can probably solve your issue and also add some value to it
this regex will not only match the optional elements but will also capture in key value pair
Regex
/(?<=:|)(?'prop'\w+)(?:\((?'val'.+?)\))?/g
Test string
length:max(260):min(20)
length
number:disallow(negative)
Result
MATCH 1
prop [0-6] length
MATCH 2
prop [7-10] max
val [11-14] 260
MATCH 3
prop [16-19] min
val [20-22] 20
MATCH 4
prop [24-30] length
MATCH 5
prop [31-37] number
MATCH 6
prop [38-46] disallow
val [47-55] negative
try demo here
EDIT
I think I understand what you meant by duplicate array with different key, it was due to named captures eg. prop & val
here is the revision without named capturing
Regex
/(?<=:|)(\w+)(?:\((.+?)\))?/
Sample code
$str = "length:max(260):min(20)";
$str .= "\nlength";
$str .= "\nnumber:disallow(negative)";
preg_match_all("/(?<=:|)(\w+)(?:\((.+?)\))?/",
$str,
$matches);
print_r($matches);
Result
Array
(
[0] => Array
(
[0] => length
[1] => max(260)
[2] => min(20)
[3] => length
[4] => number
[5] => disallow(negative)
)
[1] => Array
(
[0] => length
[1] => max
[2] => min
[3] => length
[4] => number
[5] => disallow
)
[2] => Array
(
[0] =>
[1] => 260
[2] => 20
[3] =>
[4] =>
[5] => negative
)
)
try demo here
I need to split my text into an array at every period, exclamation and question mark.
Example with a full-width period and exclamation mark:
$string = "日本語を勉強しているみんなを応援したいです。一緒に頑張りましょう！";
I am looking for the following output:
Array (
[0] => 日本語を勉強しているみんなを応援したいです。
[1] => 一緒に頑張りましょう！ )
I need the same code to work with half-width.
Example with a mix of full-width and half-width:
$string = "Hi. I am Bob! Nice to meet you. 日本語を勉強しています。Do you understand me?";
Output:
Array (
[0] => Hi.
[1] => I am Bob!
[2] => Nice to meet you.
[3] => 日本語を勉強しています。
[4] => Do you understand me? )
I suck at regular expressions and can't figure out a solution nor find one.
I tried:
$string = preg_split('(.*?[。？！])', $string);
First of all, you forgot your delimiters (most commonly a slash).
You can split on \pP (a unicode punctuation - remember the u modifier meaning unicode):
You can see the rest of the special unicode characters here.
<?php
$str = 'Hi. I am Bob! Nice to meet you. 日本語を勉強しています。Do you understand me?';
$array = preg_split('/(?<=\pP)\s*/u', $str, null, PREG_SPLIT_NO_EMPTY);
print_r($array);
The PREG_SPLIT_NO_EMPTY is there to make sure that we don't include an empty match if your last character is punctuation.
Output:
Array
(
[0] => Hi.
[1] => I am Bob!
[2] => Nice to meet you.
[3] => 日本語を勉強しています。
[4] => Do you understand me?
)
Regex autopsy:
/ - the start delimiter - this must also come at the end before our modifiers
(?<=\pP) - a positive lookbehind matching \pP (a unicode punctuation - we could just use \pP, but then the punctuation would not be included in our final string - a positive lookbehind includes it)
\s* - a white space character matched 0 to infinity times - this is to make sure that we don't include the white space after the punctuation
/u - the end delimiter (/) and our modifier (u meaning "unicode")
DEMO
Your first sentence would result in the following array:
Array
(
[0] => 日本語を勉強しているみんなを応援したいです。
[1] => 一緒に頑張りましょう！
)
Please note that this includes all punctuation including commas.
Array
(
[0] => This is my sentence,
[1] => and it is very nice.
)
This can be fixed by using a negative lookbehind in front of our positive lookbehind:
/(?<![,、;；"”\'’｀`])(?<=\pP)\s*/u
For the life of me, I can't figure out how to write the regex to split this.
Lets say we have the sample text:
15HGH(Whatever)ASD
I would like to break it down into the following groups (numbers, letters by themselves, and parenthesis contents)
15
H
G
H
Whatever
A
S
D
It can have any combination of the above such as:
15HGH
12ABCD
ABCD(Whatever)(test)
So far, I have gotten it to break apart either the numbers/letters or just the parenthesis part broken away. For example, in this case:
<?php print_r(preg_split( "/(\(|\))/", "5(Test)(testing)")); ?>
It will give me
Array
(
[0] => 5
[1] => Test
[2] => testing
)
I am not really sure what to put in the regex to match on only numbers and individual characters when combined. Any suggestions?
I don't know if preg_match_all satisfying you:
$text = '15HGH(Whatever)ASD';
preg_match_all("/([a-z]+)(?=\))|[0-9]+|([a-z])/i", $text, $out);
echo '<pre>';
print_r($out[0]);
Array
(
[0] => 15
[1] => H
[2] => G
[3] => H
[4] => Whatever
[5] => A
[6] => S
[7] => D
)
I've got this: Example (I don't know how is written the \n) but the substitution is working.
(\d+|\w|\([^)]++\)) Not too much to explain, first tries to get a number, then a char, and if there's nothing there, tries to get a whole word between parentheses. (They can't be nested)
Check this out using preg_match_all():
$string = '15HGH(Whatever)(Whatever)ASD';
preg_match_all('/\(([^\)]+)\)|(\d+)|([a-z])/i', $string, $matches);
$results = array_merge(array_filter($matches[1]),array_filter($matches[2]),array_filter($matches[3]));
print_r($results);
\(([^\)]+)\) --> Matches everything between parenthesis
\d+ --> Numbers only
[a-z] --> Single letters only
i --> Case insensitive