We Keep Coding

Regex check against referrer URL string

var orig = document.referrer; // Incoming URL
var check = new RegExp("boxes", "gi"); // Literal string, global + case insensitive.
// console.log(check);
if (orig.indexOf(check) > -1) {
console.log('you came from the box section');
} else {
console.log('you DIDNT come the box section');
}
Hi Guys,
I have a 'boxes' category on a site, where all box items have 'boxes' in the URL. A particular item from another category needs to be able to check whether or not the user came from a 'boxes' item. (This is an interim solution as I only have skin-level access).
When logging 'check', I get '/boxes/gi', which should be working when checking within indexOf, as a valid regex string.
I am not too sure why I can not get this to properly check, as the result is only ever that the user didn't come from the 'boxes' section.
I have a lot to learn, so in advance, I greatly appreciate any help.
Thanks!

You can use string variable instead of regex
var orig = document.referrer; // Incoming URL
// console.log(check);
if (orig.indexOf("boxes") > -1) {
console.log('you came from the box section');
} else {
console.log('you DIDNT come the box section');
}

indexOf does not accept a regex as argument. You either use your regex with search, or use indexOf with a string.
orig.toLowerCase().indexOf("box") > -1
// or
orig.search(check) > -1

You can parse the referrer URL into a link element and retrieve its pathname. You should also probably check the hostname to make sure it's from your own site:
var url = document.createElement('a');
url.href = document.referrer;
var comingFromBoxes = url.hostname === 'yoursite.com' && url.pathname.indexOf('/boxes') === 0;
Note: the referrer is not a reliable value by any means and should not be considered as such.

You can use match() with the regex to perform your logic.
$(document).ready(function(){
var url = "www.someurl.com/boxes/gi/abc";
var regex = /\/boxes\/gi/g;
var mtch = url.match(regex);
if(mtch !== null){
alert('url has the value');
}
else{
alert('url does not have the value');
}
})
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.min.js"></script>

Need to get indexes from array and output them to user in JavaScript

We're displaying five input fields to user. He can type some information in them. After that, we need to find out if his input is correct. For that purpose we use an array of possible correct values.
Like:
var input = document.getElementById("input").value;
input = input.toLowerCase();
inputPos = possibleInputs.indexOf(input);
inputPosArray.push(inputPos);
The code for analysis looks like that for now:
function arrayLookup() {
var inputCorrect = true;
inputPosArray.forEach(function(item, i, inputPosArray) {
if (inputPosArray[i] == -1) {
wrongInput = cardRPos.indexOf(cardRPos[i]) + 1;
wrongInputsArray.push(wrongInput);
inputCorrect = false;
} else {
null;
}
});
if (inputCorrect == false) {
alert("Wrong input! Check field " + wrongInputsArray);
} else {
nextStep();
}}
For now it correctly finds out if input is wrong and alerts user.
The problem is in "wrongInputsArray" - it doesn't display output correctly. E.g. if user has typed wrong information in 2nd field, it will print out "2".
But if he has made mistakes in 2nd and 5th field, he gets "Wrong input! Check field 2,2" alert.
Please show me what am I doing wrong.
Kindly yours,
Richard

You are using this code to insert the wrong asnwers:
wrongInput = cardRPos.indexOf(cardRPos[i]) + 1;
If two questions has the same answer, indexOf will return always the first match. Try just using this:
wrongInput = i + 1;

How to properly strip HTML tags dynamically using javascript?

I know this type of question has been asked before, which is how I came up with my regular expression in the first place, but my coding doesn't seem to be working.
I'm combining 2 things, firstly I'm trying to restrict a multiline textbox to 6000 characters and have this work on key up, which it does nicely. However, as part of this I also want to strip out HTML tags BEFORE checking the length, which is the bit that's not working. My code is below:
function TruncateNotes(text) {
var notesfield = document.getElementById(text.id);
//strip html tags such as < and > out of the text before checking length
stripHTML(text);
var maxlength = 6000;
if (notesfield.value.length > maxlength) {
notesfield.focus();
notesfield.value = text.value.substring(0, maxlength);
notesfield.scrolltop = notesfield.scrollHeight;
return false;
}
else {
return true;
}
}
function stripHTML(text) {
var notesfield = document.getElementById(text.id);
notesfield.value.replace(/<.*?>/g, "");
}
My feeling is that's something to do with the regular expression as I'm not very good with those. Any suggestions?

JavaScript '.replace' does not modify the original string, it returns a string with the values replaced. This means you'll have to assign it back to notesfield.value after the operation:
notesfield.value = notesfield.value.replace(/<.*?>/g, "");

ExtJS 4.1.1: Evaluating a field in a grid

I'm struggling with a ExtJS 4.1.1 grid that has editable cells (CellEditing plugin).
A person should be able to type a mathematic formula into the cell and it should generate the result into the field's value. For example: If a user types (320*10)/4 the return should be 800. Or similar if the user types (320m*10cm)/4 the function should strip the non-mathematical characters from the formula and then calculate it.
I was looking to replace (or match) with a RegExp, but I cannot seem to get it to work. It keeps returning NaN and when I do console.log(e.value); it returns only the originalValue and not the value that I need.
I don't have much code to attach:
onGridValidateEdit : function(editor,e,opts) {
var str = e.value.toString();
console.log(str);
var strCalc = str.match(/0-9+-*\/()/g);
console.log(strCalc);
var numCalc = Number(eval(strCalc));
console.log(numCalc);
return numCalc;
},
Which returns: str=321 strCalc=null numCalc=0 when I type 321*2.
Any help appreciated,
GR.
Update:
Based on input by Paul Schroeder, I created this:
onGridValidateEdit : function(editor,e,opts) {
var str = e.record.get(e.field).toString();
var strCalc = str.replace(/[^0-9+*-/()]/g, "");
var numCalc = Number(eval(strCalc));
console.log(typeof numCalc);
console.log(numCalc);
return numCalc;
},
Which calculates the number, but I am unable to print it back to the grid itself. It shows up as "NaN" even though in console it shows typeof=number and value=800.
Final code:
Here's the final code that worked:
onGridValidateEdit : function(editor,e,opts) {
var fldName = e.field;
var str = e.record.get(fldName).toString();
var strCalc = str.replace(/[^0-9+*-/()]/g, "");
var numCalc = Number(eval(strCalc));
e.record.set(fldName,numCalc);
},

Lets break this code down.
onGridValidateEdit : function(editor,e,opts) {
var str = e.value.toString();
What listener is this code being used in? This is very important for us to know, here's how I set up my listeners in the plugin:
listeners: {
edit: function(editor, e){
var record = e.record;
var str = record.get("your data_index of the value");
}
}
Setting it up this way works for me, So lets move on to:
var strCalc = str.match(/0-9+-*\/()/g);
console.log(strCalc);
at which point strCalc=null, this is also correct. str.match returns null because your regex does not match anything in the string. What I think you want to do instead is this:
var strCalc = str.replace(/[^0-9+*-]/g, "");
console.log(strCalc);
This changes it to replace all characters in the string that aren't your equation operators and numbers. After that I think it should work for whole numbers. I think that you may actually want decimal numbers too, but I can't think of the regex for that off the top of my head (the . needs to be escaped somehow), but it should be simple enough to find in a google search.

Jquery autocomplete, how to search on words instead of string

I have a textbox where i want to have an autocomplete that lets the user search through addresses. The user must be able to type in different words and the autocomplete must search through them to narrow its list.
I've been trying and reading the documentation, but nothing seems to do the trick as it always searches on the whole string instead of the words. Am i missing something?
Example:
When the user enters 'Mathias Antwerp' he must see all the addresses that contain those words. In the example it must show 1 row which is the second one.
<script>
var addresses = [
{ name: "Frederick Dereave Gentstreet 4 Gent" },
{ name: "Mathias Derian Meilaan 9 Antwerp" },
{ name: "Mathias Hors frelaan 5 Kortrijk" }
];
$(document).ready(SetAutoComplete);
function SetAutoComplete() {
$("#testveld").autocomplete(emails,
{
matchContains: "word"
}
);
}
</script>
<input type="text" id="testveld" style='width:300px'/>

I altered the code of matchSubset in jquery.autocomplete.js which enables the behavior i was looking for.
function matchSubset(s, sub) {
var arraySub=sub.split(" ");
if (!options.matchCase)
s = s.toLowerCase();
var i = s.indexOf(sub);
if (options.matchContains == "word"){
i = s.toLowerCase().search("\\b" + sub.toLowerCase());
}
//addition for split words
if (options.matchContains == "splittedword"){
for(itemindex=0;itemindex<arraySub.length;itemindex++){
i = s.toLowerCase().search(arraySub[itemindex].toLowerCase());
if(i==-1){
break;
}
}
}
if (i == -1) return false;
return i == 0 || options.matchContains;
};

AFAIK, you will have to to do some processing on your own to parse the string into words. You can do this using jquery or if you plan to get the addresses from server side then use some server side language.