In my controller, I access the comment data with $this->request->data['Comment']['text']. I use CakePHP's formhelper to build the form, and a plugin called Summernote to transform the textarea into a WYSIWYG editor. I save the comment as HTML in my database.
In this case, I am trying to submit a comment with just '>'
$data = $this->request->data['Comment']['text'];
pr($data);
//returns >
pr(mb_strlen($data, utf-8));
//returns 4
pr(mb_strlen('>', utf-8));
//returns 1
//that is the one that confuses me the most,
//it seems that there's a difference between $data and '>'
mb_detect_encoding($data);
//returns ASCII
I'm already using jQuery to check the number of characters entered on the front-end, so I can deactivate the submit-button when the user goes over the limit. This uses .innerText.length and works like a charm, but if I make that the only check people can just go into the element editor and re-enable the submit button to send however long comments they like.
EDIT: var_dump($this->request->data['Comment']['text']) gave me the following result:
Note that unlike in the examples above, I am trying to send '>>>' here
array (size=1)
'text' => string '>>>' (length=12)
EDIT: Alex_Tartan figured out the problem: I needed to do html_entity_decode() on my string before counting it with mb_strlen()!
I've tested the case here: https://3v4l.org/VLr9e
What might be the case is an untrimmed $data (white spaces won't show up in regular print - you can use var_dump($data)).
The textarea tag will enclose formatting spaces into the value.
Check out Why is textarea filled with mysterious white spaces?
so for that, you can do:
$data = '> ';
$data = trim($data);
// var_dump(data) will output:
// string(4) "> "
echo $data."\n";
//returns >
echo mb_strlen($data, 'UTF-8')."\n";
//returns 1
echo mb_strlen('>', 'UTF-8')."\n";
//returns 1
Update (from comments):
The problem was encoded html characters which needed to be decoded:
$data = html_entity_decode($data);