We Keep Coding

Join three variables and later extract with regex

I am joining together 3 variables into a filename seperated by a -- for use in another Cloud function. I do this with the following code below:
const newFileName = size + "--" + bytes + "--" + name + ".txt";
How can I seperate these later with Regex?
size (extract before first --)
bytes (extract after first --)
name (extract after second --)

If you always use the same extension .txt you might remove that part from the string and then split on --
let str = "aaa--bbb--ccc.txt";
[size, bytes, name] = str.substring(0, str.length - 4).split("--");
console.log(size);
console.log(bytes);
console.log(name);
Another option could be to split on either -- or on a pattern that would match the last dot followed by matching not a dot until the end of the string.
let str = "aaa--bbb--ccc.txt";
[size, bytes, name] = str.split(/--|\.[^.]+$/);
console.log(size);
console.log(bytes);
console.log(name);

If you use this regex, it will pull the three parts into 3 different capture groups named "size", "bytes", and "name":
^(?<size>.*?)\-\-(?<bytes>.*?)\-\-(?<name>.*?[.txt])$
In javascript, you can use this code to pull the groups out afterwards:
var reg = /^(?<size>.*?)\-\-(?<bytes>.*?)\-\-(?<name>.*?[.txt])$/
var matches = "123--456--test.txt".match(reg);
var size = matches["groups"]["size"]
var bytes = matches["groups"]["bytes"]
var name = matches["groups"]["name"]
Where the "123--456--test.txt" string is instead the string you want to pull the data out of.
EDIT:
Like StackSlave pointed out, named capture groups like above are not always supported. If you need to support a browser or language that can't deal with named groups, you can instead do this:
var reg = /^(.*?)\-\-(.*?)\-\-(.*?)$/
var matches = "123--456--test.txt".match(reg);
var size = matches[1]
var bytes = matches[2]
var name = matches[3]
EDIT EDIT:
To get the file name without extension, use this regex instead:
var reg = /(.*?)\-\-(.*?)\-\-(.*?)\./
var matches = "123--456--test.txt".match(reg);
var size = matches[1]
var bytes = matches[2]
var name = matches[3]

Here's a numeric example:
var str = 'size--bytes--name.ext', m = str.match(/^(.+)--(.+)--(.+)(?:\..+)$/);
console.log(m); console.log('size:'+m[1]); console.log('bytes:'+m[2]); console.log('name:'+m[3]);

Update uri hash javascript

I find it hard to believe this hasn't been asked but I can find no references anywhere. I need to add a URI hash fragment and update the value if it already is in the hash. I can currently get it to add the hash but my regex doesn't appear to catch if it exists so it adds another instead of updating.
setQueryString : function() {
var value = currentPage;
var uri = window.location.hash;
var key = "page";
var re = new RegExp("([#&])" + key + "=.*#(&|$)", "i");
var separator = uri.indexOf('#') !== -1 ? "&" : "#";
if (uri.match(re)) {
return uri.replace(re, '$1' + key + "=" + value + '$2');
}
else {
return uri + separator + key + "=" + value;
}
},
Also if this can be made any cleaner while preserving other url values/hashes that would be great.
example input as requested
Starting uri value:
www.example.com#page=1 (or no #page at all)
then on click of "next page" setQueryString gets called so the values would equal:
var value = 2;
var uri = '#page1'
var key = 'page'
So the hopeful output would be '#page2'.

As to your regex question, testing if the pattern #page=(number) or &page=(number) is present combined with capturing the number, can be done with the regex /[#&]page\=(\d*)/ and the .match(regex) method. Note that = needs escaping in regexes.
If the pattern exists in the string, result will contain an array with the integer (as a string) at result[1]. If the pattern does not exist, result will be null.
//match #page=(integer) or &page=(integer)
var test = "#foo=bar&page=1";
var regex = /[#&]page\=(\d*)/;
var result = test.match(regex);
console.log(result);
If you want to dynamically set the key= to something other than "page", you could build the regex dynamically, like the following (note that backslashes needs escaping in strings, making the code a bit more convoluted):
//dynamically created regex
var test = "#foo=bar&page=1";
var key = "page"
var regex = new RegExp("[#&]" + key + "\\=(\\d*)");
var result = test.match(regex);
console.log(result);

Javascript string concatenate in the wrong order

I'm trying to generate a character from these three arrays. My code is as follows, but I can't figure out why the string doesn't concatenate as I've written. Some of it is missing, and the ending . even comes at the start at some times. Occasionally it works, but a lot of the times the output fails.
//Arrays
var character = ["man", "woman"];
var feature = ["one leg", "a drinking problem"];
var purpose = ["stay home", "leave home"];
function createSentence(){
var char = getRandomItem(character);
var feat = getRandomItem(feature);
var purp = getRandomItem(purpose);
var sentence = "A " + char + " with " + feat + ", that wants to " + purp + ".";
return sentence;
}
// Gets a random item from an input array
function getRandomItem(array){
var index = Math.floor(Math.random()*array.length);
return array[index];
}
Here's what I get in my console. . at the start and missing the "character"-part.
.with a drinking problem, that wants to leave home
EDIT: Sorry for not including this, as I guess this is very relevant here as well. I assumed the arrays were fine, but I might be missing something. When I print out the arrays after parsing them they look exactly like what I posted above.
// How i set up my arrays
var character = parseTextFile("character.txt");
var feature = parseTextFile("feature.txt");
var purpose = parseTextFile("purpose .txt");
// Parsing a file
function parseTextFile(filename){
var array = fs.readFileSync(filename).toString().split("\n");
return array;
}
/* --- character.txt ---*/
man
woman
Thank you.

// How i set up my arrays
var character = parseTextFile("character.txt");
var feature = parseTextFile("feature.txt");
var purpose = parseTextFile("purpose .txt");
// Parsing a file
function parseTextFile(filename){
//fs.read requires absolute path
var array = fs.readFileSync(__dirname + '/' + filename, 'utf8') // <-- specify format
.split('\n')
.filter(function(text) { return !!text; }); // <-- return only not-empty string
return array;
}

Regex mapping in Firefox context menu contentScript

I am developing context menu add-on for Firefox. I am trying to get the selectedText and validate if it is a number. If it is a number i am using that number value to process further.
But, i got stuck at a point where i am trying to replace [(,)] using regex in javascript replace method.
Following is the code which fails to map any number starting/ending with ( or ):
var menuItemLRG = contextMenu.Item({
label: "LRG",
data: "http://myurl/&val=:",
contentScript: 'self.on("click", function (node, data) {' +
' var selectedText = window.getSelection().toString();' +
' var formattedText1 = selectedText.trim();' +
' var formattedText2 = formattedText1.replace(/^[,\[\]()]*/g,"");' +
' var formattedText3 = formattedText2.replace(/[,\[\]()]*$/g,"");' +
' console.log(formattedText3); '+
' var regExp = new RegExp(/^[0-9]+$/);' +
' if (regExp.test(formattedText3) == true) {' +
' console.log("URL to follow :"+data+formattedText3);' +
' window.open(data+formattedText3);' +
' } '+
'});'
});
Above code fails to replace ( or ) in sample inputs: (5663812, 11620033).
But, a vanilla test like the following succeeds:
<script>
var str = "(2342423,])";
var tmpVal1 = str.replace(/^[,\[\]()]*/g,"");
var tmpVal2 = tmpVal1.replace(/[,\[\]()]*$/g,"");
var regExp = new RegExp(/^[0-9]+$/);
if (regExp.test(tmpVal2) == true) {
alert(tmpVal2);
}
</script>

After many trial and error found the issue. When we try to escape a character inside a single quotes we need to add one more escape for the escape character to get recognized, otherwise the single escape \] will be considered as ] which leads to abrupt ending of of the regex pattern.
In this case:
' var formattedText2 = formattedText1.replace(/^[,\[\]()]*/g,"");'
is decoded as :
var formattedText2 = formattedText1.replace(/^[,[]()]*/g,"");
instead of as:
var formattedText2 = formattedText1.replace(/^[,\[\]()]*/g,"");
So, by adding one more escape character for an escape character resolved the pattern correctly:
' var formattedText2 = formattedText1.replace(/^[,\\[\\]()]*/g,"");'
Sorry for wasting your time in analyzing the cause, if any.

Trimming a string from the end in Javascript

In Javascript, how can I trim a string by a number of characters from the end, append another string, and re-append the initially cut-off string again?
In particular, I have filename.png and want to turn it into filename-thumbnail.png.
I am looking for something along the lines of:
var sImage = "filename.png";
var sAppend = "-thumbnail";
var sThumbnail = magicHere(sImage, sAppend);

You can use .slice, which accepts negative indexes:
function insert(str, sub, pos) {
return str.slice(0, pos) + sub + str.slice(pos);
// "filename" + "-thumbnail" + ".png"
}
Usage:
insert("filename.png", "-thumbnail", -4); // insert at 4th from end

Try using a regular expression (Good documentation can be found at https://developer.mozilla.org/en/JavaScript/Guide/Regular_Expressions)
I haven't tested but try something like:
var re = /(.*)\.png$/;
var str = "filename.png";
var newstr = str.replace(re, "$1-thumbnail.png");
console.log(newstr);

I would use a regular expression to find the various parts of the filename and then rearrange and add strings as needed from there.
Something like this:
var file='filename.png';
var re1='((?:[a-z][a-z0-9_]*))';
var re2='.*?';
var re3='((?:[a-z][a-z0-9_]*))';
var p = new RegExp(re1+re2+re3,["i"]);
var m = p.exec(file);
if (m != null) {
var fileName=m[1];
var fileExtension=m[2];
}
That would give you your file's name in fileName and file's extension in fileExtension. From there you could append or prepend anything you want.
var newFile = fileName + '-thumbnail' + '.' + fileExtension;

Perhaps simpler than regular expressions, you could use lastindexof (see http://www.w3schools.com/jsref/jsref_lastindexof.asp) to find the file extension (look for the period - this allows for longer file extensions like .html), then use slice as suggested by pimvdb.

You could use a regular expression and do something like this:
var sImage = "filename.png";
var sAppend = "-thumbnail$1";
var rExtension = /(\.[\w\d]+)$/;
var sThumbnail = sImage.replace(rExtension, sAppend);
rExtension is a regular expression which looks for the extension, capturing it into $1. You'll see that $1 appears inside of sAppend, which means "put the extension here".
EDIT: This solution will work with any file extension of any length. See it in action here: http://jsfiddle.net/h4Qsv/