Special Expression not allowed - Regular Expression in PHP - php

I am trying match my String to not allow the case: for example 150x150 from the image name below:
test-string-150x150.png
I am using the following pattern to match this String:
/^([^0-9x0-9]+)\..+/
It works fine, Except in such a case:
teststring.com-150x150.jpg
What i need to get - the mask must disallow only dimensions in the end of string, here is some examples:
test-string-150x150.png > must disallow
any-string.png > allow
200x200-test.png > allow
1x1.png-100x100.jpg > disallow

You could use a negative lookahead to assert that the string does not contain the sizes followed by a dot and 1+ word characters till the end of the string.
^(?!.*\d+x\d+\.\w+$).+$
Explanation
^ Start of string
(?! Negative lookahead, assert what is on the right is not
.* Match 0+ occurrences of any char except a newline
\d+x\d+ Match the sizes format, where \d+ means 1 or more digits
\.\w+$ Match a dot, 1+ word characters and assert the end of the string $
) Close lookahead
.+ Match 1+ occurrences of any char except a newline
$ End of string
Regex demo

If I understand your question, you're trying to find image names that do not include the image dimensions. If so, try this:
/^(?![\w-\.]+(\d+x\d+))[\w-\.]+\.\w+$/gm
For details about this code, please see regexr.com/4tmd1. This site is a great place to play around with regexes to make sure you're getting the results you expect.
Be aware that the exact syntax of the regular expression depends on the regex engine used by whatever program you're running.

Related

Update a regex that matches twitter like mentions to allow for dots

I have already found helpful answers for a regex that matches twitter like username mentions in this answer and this answer
(?<=^|(?<=[^a-zA-Z0-9-_\.]))#([A-Za-z]+[A-Za-z0-9_]+)
(?<=^|(?<=[^a-zA-Z0-9-_\.]))#([A-Za-z]+[A-Za-z0-9-_]+)
However, I need to update this regex to also include usernames that has dots.
One or more dots are allowed in a username.
The username must not start or end with a dot.
No two consecutive dots are allowed.
Example of a matched string:
#valid.user.name
^^^^^^^^^^^^^^^^
Examples of non-matched strings:
#.user.name // starts with a dot
#user.name. // ends with a dot
#user..name // has two consecutive dots
You can use this refactored regex:
(?<=[^\w.-]|^)#([A-Za-z]+(?:\.\w+)*)$
RegEx Demo
RegEx Details:
(?<=[^\w.-]|^): Lookbehind to assert that we have start of line or any non-word, non-dot, non-hyphen character before current position
#: Match literal `#1
(: Start capture group
[A-Za-z]+: Match 1+ ASCII letters
(?:\.\w+)*: Match 0 or more instances of dot followed 1+ word characters
): End capture group
$: End
The (?<=^|(?<=[^a-zA-Z0-9-_\.])) is a positive lookbehind that requires a match to be at the start of the string or right after an alphanumeric, -, _, ., you may write it in a more compact way as (?<![\w.-]), a negative lookbehind.
Next, ([A-Za-z]+[A-Za-z0-9_]+) captures 1+ ASCII letters and then 1+ ASCII letters or/and underscores. You seem to make sure the first char is a letter, then any number of sequences of . and 1+ word chars are allowed, that is, you may use [A-Za-z]\w*(?:\.\w+)*.
As you do not want to match it if there is a . right after the expected match, you need to set a lookahead that will require a space or end of string, (?!\S).
So, combining it, you can use
'~(?<![\w.-])#([A-Za-z]\w*(?:\.\w+)*)(?!\S)~'
See the regex demo
Details
(?<![\w.-]) - no letters, digits, _, . and - immediately to the left of the current location are allowed
# - a # char
([A-Za-z]\w*(?:\.\w+)*) - Group 1:
[A-Za-z] - an ASCII letter
\w* - 0+ letters, digits, _
(?:\.\w+)* - 0+ sequences of
\. - dot
\w+ - 1+ letters, digits, _
(?!\S) - whitespace or end of string are required immediately to the right of the current location.
EDIT: Simpler version (same result)
^#[a-zA-Z](\.?[\w-]+)*$
Original
Another one:
^#[a-zA-Z][a-zA-Z_-]?(\.?[\w\d-]+){0,}$
^# starts with #
[a-zA-Z] first char
[a-zA-Z_-]? match a-zA-Z_- 0 or more times
( start group
\.? match . (optional)
[\w\d-]+ match a-zA-Z0-9-_ 1 or more times
) end group
{0,} repeat group 0 to infinite times
$ end
Tests
valid:
#validusername
#valid.user.name
#valid-user-name
#valid_user-name
#valid-user123_name
#a.valid-user123_name
not valid:
#-invalid.user
#_invalid.user
#1notvalid-user_123name33
#.user.name
#user.name.
#user..name

How do grab words that start and end with double percent signs?

I am really bad at regex and I am trying to do the following:
How do I get all strings that starts and end with %%.
If these words appear in a string I want to be able to grab them: %%HELLO_WOLD%%, %%STUFF%%
Here's what I came up with so far: %%[a-zA-Z0-9]\w+
You could use anchors to assert the start ^ and the end $ of the line and match zero or more times any character .* or if there must be at least one character your might use .+
^%%.*%%$
Or instead of .* you could add your character class [a-zA-Z0-9]+ which will match lower and uppercase characters and digits or use the \w+ which will match a word character.
Note that the character class [a-zA-Z0-9] does not match an underscore and \w does.
If you want to find multiple matches in a string you might use %%\w+%%. This will also match %%HELLO_WOLD%% in %%%%%HELLO_WOLD%%%.
If there should be only 2 percentage signs at the beginning and at the end, you could use a positive lookahead (?= and positive lookbehind (?<= to assert that what is before and after the 2 percentage signs is not a percentage sign or are the start ^ or end $ of the string.
(?<=^|[^%])%%\w+%%(?=[^%]|$)

PHP regex: zero or more whitespace not working

I'm trying to apply a regex constraint to a Symfony form input. The requirement for the input is that the start of the string and all commas must be followed by zero or more whitespace, then a # or # symbol, except when it's the empty string.
As far as I can tell, there is no way to tell the constraint to use preg_match_all instead of just preg_match, but it does have the ability to negate the match. So, I need a regular expression that preg_match will NOT MATCH for the given scenario: any string containing the start of the string or a comma, followed by zero or more whitespace, followed by any character that is not a # or # and is not the end of the string, but will match for everything else. Here are a few examples:
preg_match(..., ''); // No match
preg_match(..., '#yolo'); // No match
preg_match(..., '#yolo, #swag'); // No match
preg_match(..., '#yolo,#swag'); // No match
preg_match(..., '#yolo, #swag,'); // No match
preg_match(..., 'yolo'); // Match
preg_match(..., 'swag,#yolo'); // Match
preg_match(..., '#swag, yolo'); // Match
I would've thought for sure that /(^|,)\s*[^##]/ would work, but it's failing in every case with 1 or more spaces and it appears to be because of the asterisk. If I get rid of the asterisk, preg_match('/(^|,)\s[^##]/', '#yolo, #swag') does not match (as desired) when there's exactly once space, but as as soon as I reintroduce the asterisk it breaks for any quantity of spaces > 0.
My theory is that the regex engine is interpreting the second space as a character that is not in the character set [##], but that's just a theory and I don't know what to do about it. I know that I could create a custom constraint to use preg_match_all instead to get around this, but I'd like to avoid that if possible.
You may use
'~(?:^|,)\s*+[^##]~'
Here, the + symbol defines a *+ possessive quantifier matching 0 or more occurrences of whitespace chars, and disallowing the regex engine to backtrack into \s* pattern if [^##] cannot match the subsequent char.
See the regex demo.
Details
(?:^|,) - either start of string or ,
\s*+ - zero or more whitespace chars, possessively matched (i.e. if the next char is not matched with [^##] pattern, the whole pattern match will fail)
[^##] - a negated character class matching any char but # and #.

Regex to match numbers only if alphabets are present

I require a regex to match the string in the following way:
#1234abc : Should get matched
#abc123 : Should get matched
#123abc123 : Should get matched
#123 : Should not get matched
#123_ : Should not get matched
#123abc_ : Should get matched
This implies that it should only get matched if the string contains numbers or underscore along with alphabets. Only numbers/underscore should not get matched. Any other special characters should not get matched either.
This regex is basically to get hashtags from string. I have already tried the following but it didn't worked well for me.
preg_match_all('/(?:^|\s)#([a-zA-Z0-9_]+$)/', $text, $matches);
Please suggest something.
If you need to match hashtags in the format you specified in a larger string, use
(?<!\S)#\w*[a-zA-Z]\w*
See the regex demo
Details:
(?<!\S) - there must be a start of string or a whitespace before
# - a hash symbol
\w* - 0+ word chars (that is, letters, digits or underscore)
[a-zA-Z] - a letter (you may use \p{L} instead)
\w* - 0+ word chars.
Other alternatives (that may appear faster, but are a bit more complex):
(?<!\S)#(?![0-9_]+\b)\w+
(?<!\S)#(?=\w*[a-zA-Z])\w+
The point here is that the pattern basically matches 1+ word chars preceded with # that is either at the string start or after whitespace, but (?![0-9_]+\b) negative lookahead fails all matches where the part after # is all digits/underscores, and the (?=\w*[a-zA-Z]) positive lookahead requires that there should be at least 1 ASCII letter after 0+ word chars.
You can use this Regex:
((.*?(\d+)[a-zA-Z]+.*)|(.*[a-zA-Z]+(\d+).*)).
Access it here: http://regexr.com/3ef6q
see it working:
Do:
^(?=.*[A-Za-z])[\w_]+$
[\w_]+ matches one or more of letters, digits, _
The zero width positive lookahead pattern, (?=.*[A-Za-z]), makes sure the match contains at least one letter
Demo

Regular Expression to match ([^>(),]+) but include some \w's in it?

I'm using php's preg_replace function, and I have the following regex:
(?:[^>(),]+)
to match any characters but >(),. The problem is that I want to make sure that there is at least one letter in it (\w) and the match is not empty, how can I do that?
Is there a way to say what i DO WANT to match in the [^>(),]+ part?
You can add a lookahead assertion:
(?:(?=.*\p{L})[^>(),]+)
This makes sure that there will be at least one letter (\p{L}; \w also matches digits and underscores) somewhere in the string.
You don't really need the (?:...) non-capturing parentheses, though:
(?=.*\p{L})[^>(),]+
works just as well. Also, to ensure that we always match the entire string, it might be a good idea to surround the regex with anchors:
^(?=.*\p{L})[^>(),]+$
EDIT:
For the added requirement of not including surrounding whitespace in the match, things get a little more complicated. Try
^(?=.*\p{L})(\s*)((?:(?!\s*$)[^>(),])+)(\s*)$
In PHP, for example to replace all those strings we found with REPLACEMENT, leaving leading and trailing whitespace alone, this could look like this:
$result = preg_replace(
'/^ # Start of string
(?=.*\p{L}) # Assert that there is at least one letter
(\s*) # Match and capture optional leading whitespace (--> \1)
( # Match and capture... (--> \2)
(?: # ...at least one character of the following:
(?!\s*$) # (unless it is part of trailing whitespace)
[^>(),] # any character except >(),
)+ # End of repeating group
) # End of capturing group
(\s*) # Match and capture optional trailing whitespace (--> \3)
$ # End of string
/xu',
'\1REPLACEMENT\3', $subject);
You can just "insert" \w inside (?:[^>(),]+\w[^>(),]+). So it will have at least one letter and obviously not empty. BTW \w captures digits as well as letters. If you want only letters you can use unicode letter character class \p{L} instead of \w.
How about this:
(?:[^>(),]*\w[^>(),]*)

Categories