I'm building an ExpressionEngine module in PHP.
In ExpressionEngine, one can access parameters passed to the module in a template using:
$my_param = $this->EE->TMPL->fetch_param('my_param');
However, when I fetch a string that way, explode does not work on it:
public function get_tyres()
{
$tyres = $this->EE->TMPL->fetch_param('tyres');
echo($tyres);
// this shows: '205/55R16M+S|205/55R16|205/55R16'
// now I want to split it into single tyres, using the pipe as a delimiter
$tyre_array = explode("|", $tyres);
foreach($tyre_array as $tyre)
{
echo($tyre . '<br>');
}
// the above produces: '205/55R16M+S|205/55R16|205/55R16',
// where I'd expect it to produce:
// 205/55R16M+S
// 205/55R16
// 205/55R16
}
I've tried to specifically cast to a string using $tyres = (string) $this->EE->TMPL->fetch_param('tyres');, with no luck.
I've also tried manually creating and exploding a string:
$tyres = '205/55R16M+S|205/55R16|205/55R16'; which worked, but obviously I need to get the param from the template, not hard code it.
Lastly, I tried using preg_split and a regex, with no luck either:
$tyre_array = preg_split('/\|/', $tyres); which also returned an array with the entire string in it.
What could be at work here? Is this a scope related thing? Is it an encoding-related thing? What to look for next?
Update
Okay, we're getting somewhere. I've added the following to the function:
for($i = 0; $i < strlen($tyres); $i++) {
echo substr($tyres, $i, 1) . ", ";
}
Which returns... {, v, e, r, s, i, o, n, :, t, y, r, e, s, }, and that is in fact the variable passed to PHP in the HTML template:
{exp:my_module:tyres tyres="{version:tyres}"}
<h1>{tyre:name}</h1>
... more irrelevant HTML
{/exp:my_module:tyres}
This means it has something to do with the parsing order of ExpressionEngine. Apparently, the variable {version:tyres} isn't parsed yet. So I pass that variable to the module, it tries to explode it by the pipe character, but the string {version:tyres} does not contain a pipe, meaning it can't be exploded. ExpressionEngine then returns {version:tyres} as a whole, passes it back in to the template and then the variable is parsed as 205/55R16M+S|205/55R16|205/55R16.
I've tested this, and can confirm that exploding by ':' returns the array:
array (size=2)
0 => string '{version' (length=8)
1 => string 'tyres}' (length=6)
I will now look in to ExpressionEngine parse order. If anyone has an idea on how to work around this, I'd be happy to know ;-).
I tested your code, copy and pasting your test string, and it works fine for me. Make sure your pipe ("|") is really the character you think it is, and not, say, some crazy unicode stuff that just looks like a pipe, but is actually Klingon for for 'staff', or something.
Try something like this as a reality check:
echo " the pipe is at " .strpos($tyres,"|"). " and I really hope this it says '12'";
The answer lies in the ExpressionEngine parse order, as outlined here: https://ellislab.com/expressionengine/user-guide/templates/template_engine.html#rendering-stages
Because the template tag {exp:my_module:tyres} used a variable passed to it by a template tag that it was nested in, the variable wasn't parsed yet as the innermost tags are parsed first.
Adding the parameter parse="inward" to the outer template tag makes ExpressionEngine parse that tag first, passing the correct variable to the inner template tag.
See https://ellislab.com/expressionengine/user-guide/templates/plugins.html#changing-parsing-order for more on changing the parsing order.
Related
Basically, what I want to achieve is dynamically replace {SOME_TAG} with "Text".
My idea was to read all tags like {SOME_TAG}, put them into array.
Then convert array keys into variables like $some_tag, and put them into array.
So, this is how far I got:
//Some code goes here
$some_tag = "Is defined somewhere else.";
$different_tag = 1 + $something;
Some text {SOME_TAG} appears in different file, which contents has been read earlier.
//Some code goes here
preg_match_all('/{\w+}/', $strings, $search);
$search = str_replace(str_split('{}'),"",$search[0]);
$search = array_change_key_case( array_flip($search), CASE_LOWER);
...some code missing here, which I cant figure out.
Replace array should look something like this
$replace = array($some_tag, $different_tag);
//Then comes replacing code and output blah blah blah..
How to make array $replace contain variables dynamically depending on $search array?
Why not something along the lines of:
<?php
$replace = array(
'{TAG_1}' => 'hello',
'{TAG_2}' => 'world',
'{TAG_3}' => '!'
);
$myString = '{TAG_1} {TAG_2}{TAG_3}{TAG_3}';
echo str_replace(array_keys($replace), array_values($replace), $myString);
If I understand correctly:
You're working on trying to create a customizable document, using {TAGS} in order to represent replaceable areas that can be filled in with dynamic information. At some point in time while replacing the {TAGS} with the dynamic information, you want the dynamic information to be stored in automatically generated basic variable names, as $tags.
I'm not sure why you want to convert these tags to basic variables instead using them entirely as array keys. I would like to point out that this represents a security or functionality hole - what happens if someone puts {REPLACE} in as a tag in your document? Your replace array would get overwritten with dynamic data, and your whole program would fall apart. Either that, or the whole replace array would get dumped in for {REPLACE}, making for a very messy document with perhaps data you don't WANT them to have in it. Perhaps you have this dealt with - I don't have all the context here - but I thought I'd point out the risk factor.
As for a better solution, unless there's some specific need that you're addressing by going through $tags instead of using using the $replace array directly, I like #Emissary's answer.
I'm writing a script (in PHP) which will go through a PHP file, find all instances of a function, replace the function name with another name, and manipulate the parameters. I'm using get_file_contents() then strpos() to find the positions of the function, but I'm trying to find a good way to extract the parameters once I know the position of the start of the function. Right now I'm just using loop which walks through the next characters in the file string and counts the number of opening and closing parentheses. Once it closes the function parameters it quits and passes back the string of parameters. Unfortunately, runs into trouble with quotes enclosing parentheses (i.e. function_name(')', 3)). I could just count quotes too, but then I have to deal with escaped quotes, different types of quotes, etc.
Is there a good way to, knowing the start of the function, to grab the string of parameters reliably? Thank you much!
EDIT:
In case i didn't read the question carefully, if you want to only get function parameters,you can see these example :
$content_file = 'function func_name($param_1=\'\',$param_2=\'\'){';
preg_match('/function func_name\((.*)\{/',$content_file,$match_case);
print_r($match_case);
but if you want to manipulate the function, read below.
How about these :
read file using file_get_contents();
use preg_match_all(); to get all function inside that file.
please not that i write /*[new_function]*/ inside that file to identify EOF.
I use this to dynamically add/ delete function without have to open that php files.
Practically, it should be like this :
//I use codeigniter read_file(); function to read the file.
//$content_file = read_file('path_to/some_php_file.php');
//i dont know whether these line below will work.
$content_file = file_get_content('path_to/some_php_file.php');
//get all function inside php file.
preg_match_all('/function (.*)\(/',$content_file,$match_case);
//
//function name u want to get
$search_f_name = 'some_function_name';
//
for($i=0;$i<count($match_case[1]);$i++){
if(trim($match_case[1][$i]) == $search_f_name){
break;
}
}
//get end position by using next function start position
if($i!=count($match_case[1])-1){
$next_function= $match_case[1][$i+1];
$get_end_pos = strripos($content_file,'function '.$next_function);
} else {
//Please not that i write /*[new_function]*/ at the end of my php file
//before php closing tag ( ?> ) to identify EOF.
$get_end_pos = strripos($content_file,'/*[new_function]*/');
}
//get start position
$get_pos = strripos($content_file,'function '.$search_f_name);
//get function string
$func_string = substr($content_file,$get_pos,$get_end_pos-$get_pos);
you can do echo $func_string; to know whether these code is running well or not.
Use a real parser, like this one:
https://github.com/nikic/PHP-Parser
Using this library, you can manipulate the source code as a tree of "node" objects, rather than as a string, and write it back out.
So a PHP file returns a string ( to an ajax call ) like this :
$output = $sessID."###".$sessEmail."###".$sessFirstName."###".$sessLanguage."###".$sessRememberMe;
and in javascript i do :
if (reply.indexOf("###") >= 0) {
arrayReply = reply.split("###");
user.ID = arrayReply[0];
user.Email = arrayReply[1];
user.FirstName = arrayReply[2];
user.Language = arrayReply[3];
user.RememberMe = arrayReply[4];
}
a problem can arise when parts of reply contain the the delimiter i use "###". What can I do in such a situation? Making the delimiter more complex/rare is not a solution in my opinion.
PS: I did try JSON but it's WAY SLOWER server side.
FINAL EDIT:
server side JSON is slower, and the same for client side, however it's not going to be a bottleneck ( 430ms for 100.000 calls ) and plus there is no need as Jules said below to re-invent the wheel. There was one more solution: bin2hex() in php [which reduced the time from 430ms to 240] and then get back the string in javascript with a hex2string function, however not worth the effort. JSON it is. Thank you all!
If as you say encoding as JSON is slower than you could try the following,
$output = '"' . some_kind_of_escape_function($sessID).'","'.some_kind_of_escape_function($sessEmail).'","'.some_kind_of_escape_function($sessFirstName).'","'.some_kind_of_escape_function($sessLanguage).'","'.$sessRememberMe.'"';
and of course replace some_kind_of_escape_function with the appropriate php function (e.g. addslashes or mysql_real_escape_string) it has been a while since I've done PHP development so choose the one that best suits your needs
Then it's a simple case of splitting by the comma and removing the quotes
One option is to use JSON object instead.
For PHP (using json_encode):
$output = json_encode(array(
"sessid" => $sessID,
"sessEmail" => $sessEmail,
"sessFirstName" => $sessFirstName,
"sessLanguage" => $sessLanguage,
"sessRememberMe" => $sessRememberMe
));
For JS (using jQuery method):
$.getJSON("/path/to/script.php", function(reply) {
user.ID = reply.sessid;
user.Email = reply.sessEmail;
user.FirstName = reply.sessFirstName;
user.Language = reply.sessLanguage;
user.RememberMe = reply.sessRememberMe;
});
Otherwise, you can use any other delimiter that possibly won't be found in the fields (or you can replace it throughout the fields). One of the examples is to use symbol of newline (\n).
Why develop your own format if there is already one?
use Json:
$output = json_encode(array('sessionID'=>$sessID,'sessionEmail'=>sessEmail,'sessionFirstName'=>$sessFirstName,'sessLanguage'=>$sessLanguage,'sessRememberMe'=>$sessRememberMe));
And for the Javsascript Side see
http://www.javascriptkit.com/dhtmltutors/ajaxgetpost4.shtml
or if your using JQuery etc. your Framework is much likely to have some kind of inbuild functionality such as http://api.jquery.com/jQuery.getJSON/
However if you want to use your ###-Delimiter i'd suggest you reduce it to just "#", for the sake of simplicity and space. After that introduce what is called an escape charater such as "\" So in a prepass you'll parse your input and replace all occurences of # with #, vice versa in the output. You can then Split your String using a special Regex, which only splits by # and not by "#"
You can use json.
http://php.net/manual/en/function.json-encode.php
How to JSON decode array elements in JavaScript?
This works up until the 13th character is hit. Once the str_ireplace hits "a" in the cyper array, the str_ireplace stops working.
Is there a limit to how big the array can be? Keep in mind if type "abgf" i get "nots", but if I type "abgrf" when I should get "notes" I get "notrs". Racked my brain cant figure it out.
$_cypher = array("n","o","p","q","r","s","t","u","v","w","x","y","z","a","b","c","d","e","f","g","h","i","j","k","l","m");
$_needle = array("a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y","z");
$_decryptedText = str_ireplace($_cypher, $_needle, $_text);
echo $_decryptedText;
Help?
Use strtrDocs:
$_text = 'abgrf';
$translate = array_combine($_cypher, $_needle);
$_decryptedText = strtr($_text, $translate);
echo $_decryptedText; # notes
Demo
But, was there something I was doing wrong?
It will replace each pair, one pair after the other on the already replaced string. So if you replace a character that you replace again, this can happen:
r -> e e -> r
abgrf -> notes -> notrs
Your e-replacement comes after your r-replacement.
Take a peak at the docs for str_replace. Namely the following line:
Because str_replace() replaces left to right, it might replace a previously inserted value when doing multiple replacements. See also the examples in this document.
So it's working as told. It's just doing a circular replacement (n -> a, then a -> n).
Use str_rot13
although it appears to be a straight rot13, if it is not, another option is to use strtr(). You provide a string and an array of replacement pairs and get the resulting translation back.
i have already succesfully translated some quotes via my translation function __(); and now I want to echo only one of those quotes at random. All quotes are separated in this string with a special character like a |
Sofar I only have this. What code could should go below this tackle my random echo?
$quotes =
__("IF YOU MAKE EVERYTHING BOLD, NOTHING IS BOLD") . "|" .
__("Quality of design is an indicator of credibility") . "|" .
__("People ignore design, that ignores people");
(An important restriction: it is essential that the quotes be exactly closed with __(" and "); sothat they can be checked and translated.) __($variable) doest not work with current clean up scripts that I have bought so these won't work.
You're already calling __() on each of your quotes individually, why not save all the extra translating and do something like:
$quotes = array('quote1', 'quote2', 'quote3');
$index = array_rand($quotes);
echo __($quotes[$index]);
Edit: To satisfy your other requirement, that the call to __() must immediately surround each string, you could do this:
$quotes = array(__('quote1'), __('quote2'), __('quote3'));
$index = array_rand($quotes);
echo $quotes[$index];
The big downside here is that you're now looking up a translation for every string in that array, even though only one is printed, but that's the same situation you had in the "one big string" solution.
Why don't you keep them in an array and translate only what is actually outputted?
$quotes = array(
"IF YOU MAKE EVERYTHING BOLD, NOTHING IS BOLD",
"Quality of design is an indicator of credibility",
"People ignore design, that ignores people",
);
$randomQuote = $quotes[ rand(0, count($quotes)-1)];
echo __($randomQuote);
Why the biscuits are they all in one string, and not an array? Your problem would be immediately solved if this was the case. As stands, split in | and index randomly into the array created to pick a random quote.