pass $` value to associated parameter function of replace - javascript

I have an expression say
log(1,3)+4,5+max(7,8,9)
where comma is being used two ways.
1- In "log(1,3)+4,5" comma is being used in place of dot(.) or decimal sign.i.e. "log(1,3)+4,5" is equivalent to "log(1.3)+4.5".
2- In max(7,8,9) it is being used as number separator. i.e. this outcome of this is 9 ; the maximum number.
My problem is to substitute comma; which is being used as decimal separator; with decimal but this should not affect max(7,8,9). i.e. I need to convert above expression to
log(1.3)+4.5+max(7,8,9)
What I tried-
function substitute(expr) {
expr.replace(/,/g, function ($`) {
/*some processing here to decide whether comma to be substituted with dot or not.On that basis I will return either dot or comma.*/
}
But how can I pass $` value to associated function
or
Is it possible to do this in javascript.
expr.replace(/,/g,function ($`) {
if yes then how?

Your language is ambiguous.
max(8,1,8,2)
Does this return 8, 8,1 or 8,2?
Your language also doesn't look regular, so you can't parse it with a regular expression, you need the context. If something like this is allowed:
max(1,max(2,3)) // 3?
Assuming you can get rid of the ambiguity, you could write a parser to do the context detection.

This could be a solution :
function myFilter(string) {
// save all functions and signs
var functions = [];
var regExp = /[+,-]max\(([^\)]+)\)/;
matches = true;
while (matches !== null) {
var matches = regExp.exec(string);
if (matches !== null) {
functions.push(matches[0]);
string = string.replace(matches[0], '');
}
}
// replace all remaining commas with dots
string = string.replace(/,/g , ".");
for (i in functions) {
string += functions[i];
}
return string;
}
var s = '1,3+4,5+max(7,8,9)-max(2,3,5)';
var filteredString = myFilter(s);
jsFiddle Demo
This currently works with multiple max functions but only + and - signs. It could be improved with *, / and more... You will have to find the good regex.

Try the below using Javascript. Hope this helps you in logic.
DEMO HERE
var value = "log(1,3)-4,5+max(7,8,9)";
var val = '';
var splitValue, appendSym;
if (value.indexOf("+") != -1)
{
splitValue = value.split("+");
appendSym = "+";
}
else if(value.indexOf("-") != -1)
{
splitValue = value.split("-");
appendSym = "-";
}
else if(value.indexOf("*") != -1)
{
splitValue = value.split("*");
appendSym = "*";
}
else
{
splitValue = value.split("/");
appendSym = "/";
}
var length = splitValue.length;
for (var i = 0; i < length; i++) {
if (val) val += appendSym;
var strrep = splitValue[i].replace(/,/g,".");
if (splitValue[i].indexOf("max") != -1 || splitValue[i].indexOf("min") != -1)
{
val+=splitValue[i];
}
else
{
val+=strrep;
}
}
alert(val);
The output for the above code is log(1.3)-4.5+max(7,8,9)

Related

Remove numbers that aren't within square brackets in JS

I have a generated a series of arrays in JavaScript but need to remove all numbers that are outside of square brackets.
Now I'm aware I need to use some form of Regex but I'm not sure how to use it in this instance.
Here's a sample of my arrays:
1[1,0.031410759078128],2[2,0.062790519529313],3[3,0.094108313318514],4[4,0.1253332335643],5[5,0.15643446504023]
etc.
How can I remove all the numbers that aren't in square brackets?
You will want to use a zero-width lookahead assertion. The Regex string below will select all numbers in sequence that come directly before a '['.
/([0-9]+)(?=\[)+/g
Rexexr.com is a great resource. http://regexr.com/3e63k
var stripNumsOutside = function(s) {
var result = "";
var insideBrackets = false;
var curChar;
for (var i = 0;
(i < s.length); i++) {
curChar = s[i];
if (curChar == "[") {
insideBrackets = true;
}
if ((insideBrackets) || (curChar == ",")) {
result += curChar;
};
if (curChar == "]") {
insideBrackets = false;
};
}
return result;
}
console.log(stripNumsOutside("1[1,0.031410759078128],2[2,0.062790519529313],3[3,0.094108313318514],4[4,0.1253332335643],5[5,0.15643446504023]"));

Javascript: Split a string by comma, except inside parentheses

Given string in the form:
'"abc",ab(),c(d(),e()),f(g(),zyx),h(123)'
How can I split it to get the below array format:
abc
ab()
c(d(),e())
f(g(),zyx)
h(123)
I have tried normal javascript split, however it doesn't work as desired. Trying Regular Expression but not yet successful.
You can keep track of the parentheses, and add those expressions when the left and right parens equalize.
For example-
function splitNoParen(s){
var left= 0, right= 0, A= [],
M= s.match(/([^()]+)|([()])/g), L= M.length, next, str= '';
for(var i= 0; i<L; i++){
next= M[i];
if(next=== '(')++left;
else if(next=== ')')++right;
if(left!== 0){
str+= next;
if(left=== right){
A[A.length-1]+=str;
left= right= 0;
str= '';
}
}
else A=A.concat(next.match(/([^,]+)/g));
}
return A;
}
var s1= '"abc",ab(),c(d(),e()),f(g(),zyx),h(123)';
splitNoParen(s1).join('\n');
/* returned value: (String)
"abc"
ab()
c(d(),e())
f(g(),zyx)
h(123)
*/
This might be not the best or more refined solution, and also maybe won't fit every single possibility, but based on your example it works:
var data = '"abc",ab(),c(d(),e()),f(g(),zyx),h(123)';
// Create a preResult splitting the commas.
var preResult = data.replace(/"/g, '').split(',');
// Create an empty result.
var result = [];
for (var i = 0; i < preResult.length; i++) {
// Check on every preResult if the number of parentheses match.
// Opening ones...
var opening = preResult[i].match(/\(/g) || 0;
// Closing ones...
var closing = preResult[i].match(/\)/g) || 0;
if (opening != 0 &&
closing != 0 &&
opening.length != closing.length) {
// If the current item contains a different number of opening
// and closing parentheses, merge it with the next adding a
// comma in between.
result.push(preResult[i] + ',' + preResult[i + 1]);
i++;
} else {
// Leave it as it is.
result.push(preResult[i]);
}
}
Demo
For future reference, here's another approach to top-level splitting, using string.replace as a control flow operator:
function psplit(s) {
var depth = 0, seg = 0, rv = [];
s.replace(/[^(),]*([)]*)([(]*)(,)?/g,
function (m, cls, opn, com, off, s) {
depth += opn.length - cls.length;
var newseg = off + m.length;
if (!depth && com) {
rv.push(s.substring(seg, newseg - 1));
seg = newseg;
}
return m;
});
rv.push(s.substring(seg));
return rv;
}
console.log(psplit('abc,ab(),c(d(),e()),f(g(),zyx),h(123)'))
["abc", "ab()", "c(d(),e())", "f(g(),zyx)", "h(123)"]
Getting it to handle quotes as well would not be too complicated, but at some point you need to decide to use a real parser such as jison, and I suspect that would be the point. In any event, there's not enough detail in the question to know what the desired handling of double quotes is.
You can't use .split for this, but instead you'll have to write a small parser like this:
function splitNoParen(s){
let results = [];
let next;
let str = '';
let left = 0, right = 0;
function keepResult() {
results.push(str);
str = '';
}
for(var i = 0; i<s.length; i++) {
switch(s[i]) {
case ',':
if((left === right)) {
keepResult();
left = right = 0;
} else {
str += s[i];
}
break;
case '(':
left++;
str += s[i];
break;
case ')':
right++;
str += s[i];
break;
default:
str += s[i];
}
}
keepResult();
return results;
}
var s1= '"abc",ab(),c(d(),e()),f(g(),zyx),h(123)';
console.log(splitNoParen(s1).join('\n'));
var s2='cats,(my-foo)-bar,baz';
console.log(splitNoParen(s2).join('\n'));
Had a similar issue and existing solutions were hard to generalize. So here's another parser that's a bit more readable and easier to extend to your personal needs. It'll also work with curly braces, brackets, normal braces, and strings of any type. License is MIT.
/**
* This function takes an input string and splits it by the given token, but only if the token is not inside
* braces of any kind, or a string.
* #param {string} input The string to split.
* #param {string} split_by Must be a single character.
* #returns {string[]} An array of split parts without the split_by character.
*/
export function parse_split(input:string, split_by:string = ",") : string[]
{
// Javascript has 3 types of strings
const STRING_TYPES = ["'","`","\""] as const;
// Some symbols can be nested, like braces, and must be counted
const state = {"{":0,"[":0,"(":0};
// Some cannot be nested, like a string, and just flip a flag.
// Additionally, once the string flag has been flipped, it can only be unflipped
// by the same token.
let string_state : (typeof STRING_TYPES)[number] | undefined = undefined
// Nestable symbols come in sets, usually in pairs.
// These sets increase or decrease the state, depending on the symbol.
const pairs : Record<string,[keyof typeof state,number]> = {
"{":["{",1],
"}":["{",-1],
"[":["[",1],
"]":["[",-1],
"(":["(",1],
")":["(",-1]
}
let start = 0;
let results = [];
let length = input.length;
for(let i = 0; i < length; ++i)
{
let char = input[i];
// Backslash escapes the next character. We directly skip 2 characters by incrementing i one extra time.
if(char === "\\")
{
i++;
continue;
}
// If the symbol exists in the single/not nested state object, flip the corresponding state flag.
if(char == string_state)
{
string_state = undefined;
console.log("Closed string ", string_state);
}
// if it's not in a string, but it's a string opener, remember the string type in string_state.
else if(string_state === undefined && STRING_TYPES.includes(char as typeof STRING_TYPES[number]))
{
string_state = char as typeof STRING_TYPES[number];
}
// If it's not in a string, and if it's a paired symbol, increase or decrease the state based on our "pairs" constant.
else if(string_state === undefined && (char in pairs) )
{
let [key,value] = pairs[char];
state[key] += value;
}
// If it's our split symbol...
else if(char === split_by)
{
// ... check whether any flags are active ...
if(Object.entries(state).every(([k,v])=>v == 0) && (string_state === undefined))
{
// ... if not, then this is a valid split.
results.push(input.substring(start,i))
start = i+1;
}
}
}
// Add the last segment if the string didn't end in the split_by symbol, otherwise add an empty string
if(start < input.length)
{
results.push(input.substring(start,input.length))
}
else
results.push("");
return results;
}
With this regex, it makes the job:
const regex = /,(?![^(]*\))/g;
const str = '"abc",ab(),c(d(),e()),f(g(),zyx),h(123)';
const result = str.split(regex);
console.log(result);
Javascript
var str='"abc",ab(),c(d(),e()),f(g(),zyx),h(123)'
str.split('"').toString().split(',').filter(Boolean);
this should work

Javascript in list

What's the easiest way to check to see if a number is in a comma delimited list?
console.log(provider[cardType]);
//returns: Object { name="visa", validLength="16,13", prefixRegExp=}
if (ccLength == 0 || (cardType > 0 && ccLength < provider[cardType].validLength)) {
triggerNotification('x', 'Your credit card number isn\'t long enough');
return false;
} else {
if ($('.credit-card input[name="cc_cvv"]').val().length < 3) {
triggerNotification('x', 'You must provide a CCV');
return false;
}
Seems similar to this SO question.
Just .split() the CSV and use inArray.
Not sure how your sample code relates to checking to see if a number is in a comma delimited list...
Also not sure if this is the easiest way, but it's what springs to mind:
<script type="text/javascript">
var myNumbers = "1,2,3,4,5";
var myArray = myNumbers.split( ',' );
// looking for "4"
for ( var i=0; i<myArray.length; i++ ) {
if (myArray[i] == 4) {
alert('Found it!');
break;
}
}
I do not see where you have a significant comma delimited list in the script you posted.
The fastest way could be something like
var csvList ="a,b,c,d,e,f,g,h";
var testList = ","+csvList+",";
var needle = "f";
alert(testList.indexOf(","+needle+",")!=-1)
just to be different ;)
If it's just a list of comma separated numbers with nothing fancy, you can just use the split method:
var numbers = list.split(",");
This will give you an array of all of the numbers in the list. Checking whether a number is in an array is trivial.
Native JavaScript and therefore cross-browser compliant. Some frameworks provide functions that do this for you, but you don't get more basic than the following.
var numbers = list.split(",");
var count = numbers.length;
var exists = false;
for (var i = 0; i < count; ++i) {
if (numbers[i] == anumber) {
exists = true;
break;
}
}
From your sample, I assume your question was "How do I see if a number is within a range of two values specified by a single-comma-delimited string?":
function inRange( number, stringRange ){
var minmax = stringRange.split(',');
minmax[0] = minmax[0]*1; //convert to number
minmax[1] = minmax[1]*1; //convert to number
minmax.sort(); // Ensure [0] is the min
return number>=minmax[0] && number<=minmax[1];
}
Try this one...
console.log(provider[cardType]); //returns: Object { name="visa", validLength="16,13", prefixRegExp=}
var regExp = new RegExp(",?" + ccLength + ",?");
if (ccLength == 0 || (cardType > 0 && !regExp.test(provider[cardType].validLength)))
{
triggerNotification('x', 'Your credit card number isn\'t long enough');
return false;
}
else
{
if ($('.credit-card input[name="cc_cvv"]').val().length < 3)
{
triggerNotification('x', 'You must provide a CCV');
return false;
}
}

Parse query string in JavaScript [duplicate]

This question already has answers here:
How can I get query string values in JavaScript?
(73 answers)
Closed 3 years ago.
The community reviewed whether to reopen this question 1 year ago and left it closed:
Duplicate This question has been answered, is not unique, and doesn’t differentiate itself from another question.
I need to parse the query string www.mysite.com/default.aspx?dest=aboutus.aspx.
How do I get the dest variable in JavaScript?
Here is a fast and easy way of parsing query strings in JavaScript:
function getQueryVariable(variable) {
var query = window.location.search.substring(1);
var vars = query.split('&');
for (var i = 0; i < vars.length; i++) {
var pair = vars[i].split('=');
if (decodeURIComponent(pair[0]) == variable) {
return decodeURIComponent(pair[1]);
}
}
console.log('Query variable %s not found', variable);
}
Now make a request to page.html?x=Hello:
console.log(getQueryVariable('x'));
function parseQuery(queryString) {
var query = {};
var pairs = (queryString[0] === '?' ? queryString.substr(1) : queryString).split('&');
for (var i = 0; i < pairs.length; i++) {
var pair = pairs[i].split('=');
query[decodeURIComponent(pair[0])] = decodeURIComponent(pair[1] || '');
}
return query;
}
Turns query string like hello=1&another=2 into object {hello: 1, another: 2}. From there, it's easy to extract the variable you need.
That said, it does not deal with array cases such as "hello=1&hello=2&hello=3". To work with this, you must check whether a property of the object you make exists before adding to it, and turn the value of it into an array, pushing any additional bits.
You can also use the excellent URI.js library by Rodney Rehm. Here's how:-
var qs = URI('www.mysite.com/default.aspx?dest=aboutus.aspx').query(true); // == { dest : 'aboutus.aspx' }
alert(qs.dest); // == aboutus.aspx
And to parse the query string of current page:-
var $_GET = URI(document.URL).query(true); // ala PHP
alert($_GET['dest']); // == aboutus.aspx
Me too! http://jsfiddle.net/drzaus/8EE8k/
(Note: without fancy nested or duplicate checking)
deparam = (function(d,x,params,p,i,j) {
return function (qs) {
// start bucket; can't cheat by setting it in scope declaration or it overwrites
params = {};
// remove preceding non-querystring, correct spaces, and split
qs = qs.substring(qs.indexOf('?')+1).replace(x,' ').split('&');
// march and parse
for (i = qs.length; i > 0;) {
p = qs[--i];
// allow equals in value
j = p.indexOf('=');
// what if no val?
if(j === -1) params[d(p)] = undefined;
else params[d(p.substring(0,j))] = d(p.substring(j+1));
}
return params;
};//-- fn deparam
})(decodeURIComponent, /\+/g);
And tests:
var tests = {};
tests["simple params"] = "ID=2&first=1&second=b";
tests["full url"] = "http://blah.com/?third=c&fourth=d&fifth=e";
tests['just ?'] = '?animal=bear&fruit=apple&building=Empire State Building&spaces=these+are+pluses';
tests['with equals'] = 'foo=bar&baz=quux&equals=with=extra=equals&grault=garply';
tests['no value'] = 'foo=bar&baz=&qux=quux';
tests['value omit'] = 'foo=bar&baz&qux=quux';
var $output = document.getElementById('output');
function output(msg) {
msg = Array.prototype.slice.call(arguments, 0).join("\n");
if($output) $output.innerHTML += "\n" + msg + "\n";
else console.log(msg);
}
var results = {}; // save results, so we can confirm we're not incorrectly referencing
$.each(tests, function(msg, test) {
var q = deparam(test);
results[msg] = q;
output(msg, test, JSON.stringify(q), $.param(q));
output('-------------------');
});
output('=== confirming results non-overwrite ===');
$.each(results, function(msg, result) {
output(msg, JSON.stringify(result));
output('-------------------');
});
Results in:
simple params
ID=2&first=1&second=b
{"second":"b","first":"1","ID":"2"}
second=b&first=1&ID=2
-------------------
full url
http://blah.com/?third=c&fourth=d&fifth=e
{"fifth":"e","fourth":"d","third":"c"}
fifth=e&fourth=d&third=c
-------------------
just ?
?animal=bear&fruit=apple&building=Empire State Building&spaces=these+are+pluses
{"spaces":"these are pluses","building":"Empire State Building","fruit":"apple","animal":"bear"}
spaces=these%20are%20pluses&building=Empire%20State%20Building&fruit=apple&animal=bear
-------------------
with equals
foo=bar&baz=quux&equals=with=extra=equals&grault=garply
{"grault":"garply","equals":"with=extra=equals","baz":"quux","foo":"bar"}
grault=garply&equals=with%3Dextra%3Dequals&baz=quux&foo=bar
-------------------
no value
foo=bar&baz=&qux=quux
{"qux":"quux","baz":"","foo":"bar"}
qux=quux&baz=&foo=bar
-------------------
value omit
foo=bar&baz&qux=quux
{"qux":"quux","foo":"bar"} <-- it's there, i swear!
qux=quux&baz=&foo=bar <-- ...see, jQuery found it
-------------------
Here's my version based loosely on Braceyard's version above but parsing into a 'dictionary' and support for search args without '='. In use it in my JQuery $(document).ready() function. The arguments are stored as key/value pairs in argsParsed, which you might want to save somewhere...
'use strict';
function parseQuery(search) {
var args = search.substring(1).split('&');
var argsParsed = {};
var i, arg, kvp, key, value;
for (i=0; i < args.length; i++) {
arg = args[i];
if (-1 === arg.indexOf('=')) {
argsParsed[decodeURIComponent(arg).trim()] = true;
}
else {
kvp = arg.split('=');
key = decodeURIComponent(kvp[0]).trim();
value = decodeURIComponent(kvp[1]).trim();
argsParsed[key] = value;
}
}
return argsParsed;
}
parseQuery(document.location.search);
Following on from my comment to the answer #bobby posted, here is the code I would use:
function parseQuery(str)
{
if(typeof str != "string" || str.length == 0) return {};
var s = str.split("&");
var s_length = s.length;
var bit, query = {}, first, second;
for(var i = 0; i < s_length; i++)
{
bit = s[i].split("=");
first = decodeURIComponent(bit[0]);
if(first.length == 0) continue;
second = decodeURIComponent(bit[1]);
if(typeof query[first] == "undefined") query[first] = second;
else if(query[first] instanceof Array) query[first].push(second);
else query[first] = [query[first], second];
}
return query;
}
This code takes in the querystring provided (as 'str') and returns an object. The string is split on all occurances of &, resulting in an array. the array is then travsersed and each item in it is split by "=". This results in sub arrays wherein the 0th element is the parameter and the 1st element is the value (or undefined if no = sign). These are mapped to object properties, so for example the string "hello=1&another=2&something" is turned into:
{
hello: "1",
another: "2",
something: undefined
}
In addition, this code notices repeating reoccurances such as "hello=1&hello=2" and converts the result into an array, eg:
{
hello: ["1", "2"]
}
You'll also notice it deals with cases in whih the = sign is not used. It also ignores if there is an equal sign straight after an & symbol.
A bit overkill for the original question, but a reusable solution if you ever need to work with querystrings in javascript :)
If you know that you will only have that one querystring variable you can simply do:
var dest = location.search.replace(/^.*?\=/, '');
The following function will parse the search string with a regular expression, cache the result and return the value of the requested variable:
window.getSearch = function(variable) {
var parsedSearch;
parsedSearch = window.parsedSearch || (function() {
var match, re, ret;
re = /\??(.*?)=([^\&]*)&?/gi;
ret = {};
while (match = re.exec(document.location.search)) {
ret[match[1]] = match[2];
}
return window.parsedSearch = ret;
})();
return parsedSearch[variable];
};
You can either call it once without any parameters and work with the window.parsedSearch object, or call getSearch subsequently.
I haven't fully tested this, the regular expression might still need some tweaking...
How about this?
function getQueryVar(varName){
// Grab and unescape the query string - appending an '&' keeps the RegExp simple
// for the sake of this example.
var queryStr = unescape(window.location.search) + '&';
// Dynamic replacement RegExp
var regex = new RegExp('.*?[&\\?]' + varName + '=(.*?)&.*');
// Apply RegExp to the query string
var val = queryStr.replace(regex, "$1");
// If the string is the same, we didn't find a match - return false
return val == queryStr ? false : val;
}
..then just call it with:
alert('Var "dest" = ' + getQueryVar('dest'));
Cheers
I wanted a simple function that took a URL as an input and returned a map of the query params.
If I were to improve this function, I would support the standard for array data in the URL, and or nested variables.
This should work back and for with the jQuery.param( qparams ) function.
function getQueryParams(url){
var qparams = {},
parts = (url||'').split('?'),
qparts, qpart,
i=0;
if(parts.length <= 1 ){
return qparams;
}else{
qparts = parts[1].split('&');
for(i in qparts){
qpart = qparts[i].split('=');
qparams[decodeURIComponent(qpart[0])] =
decodeURIComponent(qpart[1] || '');
}
}
return qparams;
};
I wanted to pick up specific links within a DOM element on a page, send those users to a redirect page on a timer and then pass them onto the original clicked URL. This is how I did it using regular javascript incorporating one of the methods above.
Page with links: Head
function replaceLinks() {
var content = document.getElementById('mainContent');
var nodes = content.getElementsByTagName('a');
for (var i = 0; i < document.getElementsByTagName('a').length; i++) {
{
href = nodes[i].href;
if (href.indexOf("thisurl.com") != -1) {
nodes[i].href="http://www.thisurl.com/redirect.aspx" + "?url=" + nodes[i];
nodes[i].target="_blank";
}
}
}
}
Body
<body onload="replaceLinks()">
Redirect page
Head
function getQueryVariable(variable) {
var query = window.location.search.substring(1);
var vars = query.split('&');
for (var i = 0; i < vars.length; i++) {
var pair = vars[i].split('=');
if (decodeURIComponent(pair[0]) == variable) {
return decodeURIComponent(pair[1]);
}
}
console.log('Query variable %s not found', variable);
}
function delayer(){
window.location = getQueryVariable('url')
}
Body
<body onload="setTimeout('delayer()', 1000)">

Javascript word-count for any given DOM element

I'm wondering if there's a way to count the words inside a div for example. Say we have a div like so:
<div id="content">
hello how are you?
</div>
Then have the JS function return an integer of 4.
Is this possible? I have done this with form elements but can't seem to do it for non-form ones.
Any ideas?
g
If you know that the DIV is only going to have text in it, you can KISS:
var count = document.getElementById('content').innerHTML.split(' ').length;
If the div can have HTML tags in it, you're going to have to traverse its children looking for text nodes:
function get_text(el) {
ret = "";
var length = el.childNodes.length;
for(var i = 0; i < length; i++) {
var node = el.childNodes[i];
if(node.nodeType != 8) {
ret += node.nodeType != 1 ? node.nodeValue : get_text(node);
}
}
return ret;
}
var words = get_text(document.getElementById('content'));
var count = words.split(' ').length;
This is the same logic that the jQuery library uses to achieve the effect of its text() function. jQuery is a pretty awesome library that in this case is not necessary. However, if you find yourself doing a lot of DOM manipulation or AJAX then you might want to check it out.
EDIT:
As noted by Gumbo in the comments, the way we are splitting the strings above would count two consecutive spaces as a word. If you expect that sort of thing (and even if you don't) it's probably best to avoid it by splitting on a regular expression instead of on a simple space character. Keeping that in mind, instead of doing the above split, you should do something like this:
var count = words.split(/\s+/).length;
The only difference being on what we're passing to the split function.
Paolo Bergantino's second solution is incorrect for empty strings or strings that begin or end with whitespaces. Here's the fix:
var count = !s ? 0 : (s.split(/^\s+$/).length === 2 ? 0 : 2 +
s.split(/\s+/).length - s.split(/^\s+/).length - s.split(/\s+$/).length);
Explanation: If the string is empty, there are zero words; If the string has only whitespaces, there are zero words; Else, count the number of whitespace groups without the ones from the beginning and the end of the string.
string_var.match(/[^\s]+/g).length
seems like it's a better method than
string_var.split(/\s+/).length
At least it won't count "word " as 2 words -- ['word'] rather than ['word', '']. And it doesn't really require any funny add-on logic.
Or just use Countable.js to do the hard job ;)
document.deepText= function(hoo){
var A= [];
if(hoo){
hoo= hoo.firstChild;
while(hoo!= null){
if(hoo.nodeType== 3){
A[A.length]= hoo.data;
}
else A= A.concat(arguments.callee(hoo));
hoo= hoo.nextSibling;
}
}
return A;
}
I'd be fairly strict about what a word is-
function countwords(hoo){
var text= document.deepText(hoo).join(' ');
return text.match(/[A-Za-z\'\-]+/g).length;
}
alert(countwords(document.body))
Or you can do this:
function CountWords (this_field, show_word_count, show_char_count) {
if (show_word_count == null) {
show_word_count = true;
}
if (show_char_count == null) {
show_char_count = false;
}
var char_count = this_field.value.length;
var fullStr = this_field.value + " ";
var initial_whitespace_rExp = /^[^A-Za-z0-9]+/gi;
var left_trimmedStr = fullStr.replace(initial_whitespace_rExp, "");
var non_alphanumerics_rExp = rExp = /[^A-Za-z0-9]+/gi;
var cleanedStr = left_trimmedStr.replace(non_alphanumerics_rExp, " ");
var splitString = cleanedStr.split(" ");
var word_count = splitString.length -1;
if (fullStr.length <2) {
word_count = 0;
}
if (word_count == 1) {
wordOrWords = " word";
} else {
wordOrWords = " words";
}
if (char_count == 1) {
charOrChars = " character";
} else {
charOrChars = " characters";
}
if (show_word_count & show_char_count) {
alert ("Word Count:\n" + " " + word_count + wordOrWords + "\n" + " " + char_count + charOrChars);
} else {
if (show_word_count) {
alert ("Word Count: " + word_count + wordOrWords);
} else {
if (show_char_count) {
alert ("Character Count: " + char_count + charOrChars);
}
}
}
return word_count;
}
The get_text function in Paolo Bergantino's answer didn't work properly for me when two child nodes have no space between them. eg <h1>heading</h1><p>paragraph</p> would be returned as headingparagraph (notice lack of space between the words). So prepending a space to the nodeValue fixes this. But it introduces a space at the front of the text but I found a word count function that trims it off (plus it uses several regexps to ensure it counts words only). Word count and edited get_text functions below:
function get_text(el) {
ret = "";
var length = el.childNodes.length;
for(var i = 0; i < length; i++) {
var node = el.childNodes[i];
if(node.nodeType != 8) {
ret += node.nodeType != 1 ? ' '+node.nodeValue : get_text(node);
}
}
return ret;
}
function wordCount(fullStr) {
if (fullStr.length == 0) {
return 0;
} else {
fullStr = fullStr.replace(/\r+/g, " ");
fullStr = fullStr.replace(/\n+/g, " ");
fullStr = fullStr.replace(/[^A-Za-z0-9 ]+/gi, "");
fullStr = fullStr.replace(/^\s+/, "");
fullStr = fullStr.replace(/\s+$/, "");
fullStr = fullStr.replace(/\s+/gi, " ");
var splitString = fullStr.split(" ");
return splitString.length;
}
}
EDIT
kennebec's word counter is really good. But the one I've found includes a number as a word which is what I needed. Still, that's easy to add to kennebec's. But kennebec's text retrieval function will have the same problem.
This should account for preceding & trailing whitespaces
const wordCount = document.querySelector('#content').innerText.trim().split(/\s+/).length;
string_var.match(/[^\s]+/g).length - 1;

Categories