Develop Reference

How to remove word in string based on array in Javascript when word's character length in string is fewer than in array?

I want to remove some word in string based on array. But the word's character length in string is fewer than in array. Is it possible to match it using regex and then replace it with empty string? If not, what is the alternatives?
I tried using regex to match the word, but i can't achieve it. I don't know how to make regex match minimum 3 character from the array.
array = ['reading', 'books'];
string = 'If you want to read the book, just read it.';
desiredOutput = 'If you want to the , just it.';
// Desired match
'reading' -> match for 'rea', 'read', 'readi', 'readin', 'reading'
'books' -> match for 'boo', 'book', 'books'

One option is to match 3 or more word characters starting at a word boundary, then use a replacer function to return the empty string if any of the words startsWith the word in question:
const array = ['reading', 'books'];
const string = 'If you want to read the book, just read it.';
const output = string.replace(
/\b\w{3,}/g,
word => array.some(item => item.startsWith(word)) ? '' : word
);
console.log(output);

The answer from CertainPerformance is better - easier to implement and to maintain but it's worth noting that - you can also generate a regex from the array.
The idea is simple enough - if you want to match r, re, rea, read, readi, readin, reading the regex for that is reading|readin|readi|read|rea|re|r. The reason you want the longest variation first is because otherwise the regex engine will stop at the first match in finds:
let regex = /r|re|rea|read/g
// ↑_________________
console.log( // |
"read".replace(regex, "")// |
// ↑___________________________|
)
So you can take a word and break it out in a this pattern to generate a regex from it
function allSubstrings(word) {
let substrings = [];
for (let i = word.length; i > 0; i--) {
let sub = word.slice(0, i);
substrings.push(sub)
}
return substrings;
}
console.log(allSubstrings("reading"))
With that you can simply generate the regex you need.
function allSubstrings(word) {
let substrings = [];
for (let i = word.length; i > 0; i--) {
let sub = word.slice(0, i);
substrings.push(sub)
}
return substrings;
}
function toPattern(word) {
let substrings = allSubstrings(word);
let pattern = substrings.join("|");
return pattern;
}
console.log(toPattern("reading"))
The final thing is to take an array and convert it to a regex. Which requires treating each word and then combining each individual regex into one that matches any of the words:
const array = ['reading', 'books'];
const string = 'If you want to read the book, just read it.';
//generate the pattern
let pattern = array
.map(toPattern) //first, for each word
.join("|"); //join patterns for all words
//convert the pattern to a regex
let regex = new RegExp(pattern, "g");
let result = string.replace(regex, "");
//desiredOutput: 'If you want to the , just it.';
console.log(result);
function allSubstrings(word) {
let substrings = [];
for (let i = word.length; i > 0; i--) {
let sub = word.slice(0, i);
substrings.push(sub)
}
return substrings;
}
function toPattern(word) {
let substrings = allSubstrings(word);
let pattern = substrings.join("|");
return pattern;
}
So, this is how you can generate a regular expression from that array. In this case, that works, but it's not guaranteed to, because there is a danger it could match something you don't want. For example, r will match any character, it doesn't necessarily need to be in a word that matches this.
const array = ['reading'];
const string = 'The quick brown fox jumps over the lazy dog';
// ^ ^
let pattern = array
.map(word => allSubstrings(word).join("|"))
.join("|");
let regex = new RegExp(pattern, "g");
let result = string.replace(regex, "");
console.log(result);
function allSubstrings(word) {
let substrings = [];
for (let i = word.length; i > 0; i--) {
let sub = word.slice(0, i);
substrings.push(sub)
}
return substrings;
}
Which is when it becomes more complicated, as you want to generate a more complicated pattern for each word. You generally want to match words, so you can use the word boundary character \b which means that the pattern for "reading" can now look like this:
\breading\b|\breadin\b|\breadi\b|\bread\b|\brea\b|\bre\b|\br\b
↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑ ↑↑
In the interest of keeping the output at least somewhat readable, it can instead be put in a group and the whole group made to match a single word:
\b(?:reading|readin|readi|read|rea|re|r)\b
↑↑
||____ non-capturing group
So, you have to generate this pattern
function toPattern(word) {
let substrings = allSubstrings(word);
//escape backslashes, because this is a string literal and we need \b as content
let pattern = "\\b(?:" + substrings.join("|") + ")\\b";
return pattern;
}
Which leads us to this
const array = ['reading', 'books'];
const string = 'The quick brown fox jumps over the lazy dog. If you want to read the book, just read it.';
let pattern = array
.map(toPattern)
.join("|");
let regex = new RegExp(pattern, "g");
let result = string.replace(regex, "");
console.log(result);
function allSubstrings(word) {
let substrings = [];
for (let i = word.length; i > 0; i--) {
let sub = word.slice(0, i);
substrings.push(sub)
}
return substrings;
}
function toPattern(word) {
let substrings = allSubstrings(word);
let pattern = "\\b(?:" + substrings.join("|") + ")\\b";
return pattern;
}
This will suffice to solve your task. So it's possible to generate a regex. The final one looks like this:
/\b(?:reading|readin|readi|read|rea|re|r)\b|\b(?:books|book|boo|bo|b)\b/g
But most of the generation of it is spent trying to generate something that works. It's not a necessarily complex solution but as mentioned, the one suggested by CertainPerformance is better because it's simpler which means less chance of it failing and it would be easier to maintain for the future.

I don't know of a straight way to do it, but you can create your own regexp pattern, like so:
// This function create a regex pattern string for each word in the array.
// The str is the string value (the word),
// min is the minimum required letters in eac h word
function getRegexWithMinChars(str, min) {
var charArr = str.split("");
var length = charArr.length;
var regexpStr = "";
for(var i = 0; i < length; i++){
regexpStr +="[" + charArr[i] + "]" + (i < min ? "" : "?");
}
return regexpStr;
}
// This function returns a regexp object with the patters of the words in the array
function getStrArrayRegExWithMinChars(strArr, min) {
var length = strArr.length;
var regexpStr = "";
for(var i = 0; i < length; i++) {
regexpStr += "(" + getRegexWithMinChars(strArr[i], min) + ")?";
}
return new RegExp(regexpStr, "gm");
}
var regexp = getStrArrayRegExWithMinChars(searchArr, 3);
// With the given regexp I was able to use string replace to
// find and replace all the words in the string
str.replace(regexp, "");
//The same can be done with one ES6 function
const getStrArrayRegExWithMinChars = (searchArr, min) => {
return searchArr.reduce((wordsPatt, word) => {
const patt = word.split("").reduce((wordPatt, letter, index) => {
return wordPatt + "[" + letter + "]" + (index < min ? "" : "?");
},"");
return wordsPatt + "(" + patt + ")?";
}, "");
}
var regexp = getStrArrayRegExWithMinChars(searchArr, 3);
// With the given regexp I was able to use string replace to
// find and replace all the words in the string
str.replace(regexp, "");

Finding the index to a non-specified character

Let's say for example I have a string
thisIsThisTuesday Day
I want to find the index of all the capital letters, test if there is a space before it, and if not insert one. I would need the index of each one.
At least from what I can see indexOf(String) will only produce the index of the first occurance of the character T/t
This :
for(i=0;i<str.length;i++){
let char=str[i];
if(isNaN(char*1)&&char==char.toUpperCase()){
y=str.indexOf(char);
console.log(char,y)
}
}
would produce the capital letters, and their indexes but will only display the first occurrence of the character in question. I feel pretty confident that the part I am missing is a for() loop in order to move the index iteration..but it escapes me.
Thank you in advance!

You can use a regex:
It matches any non-whitespace character followed by a capital letter and replaces it by the two characters with a space between.
const str = "thisIsThisTuesday Day";
const newstr = str.replace(/([^ ])([A-Z])/g, "$1 $2");
console.log(newstr);

You can use the following regular expression:
/(?<=\S)(?=[A-Z])/g
The replace will insert spaced between characters which are non-space followed by a capital letter.
See example below:
let str = "thisIsThisTuesday Day";
const res = str.replace(/(?<=\S)(?=[A-Z])/g, ' ');
console.log(res);
Note: As pointed out ?<= (positive lookbehind) is currently not be available in all browsers.

Actually, the String.indexOf function can take a second argument, specifying the character it should start searching from. Take a look at: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/indexOf
But, if you just want to find all capital letters and prefix them with a space character, if one is not found, there are many approaches, for example:
var str = "thisIsThisTuesday Day";
var ret = '';
for (var i=0; i<str.length; i++) {
if (str.substr(i, 1) == str.substr(i, 1).toUpperCase()) {
if ((i > 0) && (str.substr(i - 1,1) != " "))
ret += " ";
}
ret += str.substr(i,1);
}
After running this, ret will hold the value "this Is This Tuesday Day"

You could iterate over the string and check if each character is a capital. Something like this:
const s = 'thisIsThisTuesday Day';
const format = (s) => {
let string = '';
for (let c of s) {
if (c.match(/[A-Z]/)) string += ' ';
string += c;
}
return string;
};
console.log(format(s));
Or alternatively with reduce function:
const s = 'thisIsThisTuesday Day';
const format = (s) => s.split('').reduce((acc, c) => c.match(/[A-Z]/) ? acc + ` ${c}` : acc + c, '');
console.log(format(s));

Adding a whitespace in front of the first number in a string

I need a whitespace to be added in front of the first number of a string, unless there is one already and unless the number is the first character in the string.
So I wrote this JS code:
document.getElementById('billing:street1').addEventListener('blur', function() {
var value = document.getElementById('billing:street1').value;
var array = value.match(/\d{1,}/g);
if (array !== null) {
var number = array[0];
var index = value.indexOf(number);
if(index !== 0){
var street = value.substring(0, index);
var housenumber = value.substring(index);
if (street[street.length - 1] !== ' ') {
document.getElementById('billing:street1').value = street + ' ' + housenumber;
}
}
}
});
Fiddle
It works fine, but I feel like this can probably be done in a smarter, more compact way.
Also, JQuery suggestions welcome, I am just not very familiar with it.

Try this one :
const addSpace = (str) => {
return str.replace(/(\D)(\d)/, "$1 $2")
}
console.log(addSpace("12345")) // "12345"
console.log(addSpace("city12345")) // "city 12345"
console.log(addSpace("city")) // "city"
(\D) captures a non-digit
(\d) captures a digit
so (\D)(\d) means : non-digit followed by a digit
that we replace with "$1 $2" = captured1 + space + captured2

You can do it by using only regular expressions. For example:
var s = "abcde45";
if(!s.match(/\s\d/)){
s = s.replace(/(\d)/, ' $1');
}
console.log(s); // "abcde 45"
UPD : Of course, if you have string with a wrong syntax(e.g no numbers inside), that code wouldn't work.

check if any duplicate words exist in a string using regex

Using regex in javascript, how to check for the condition when duplicate words are present in a string?
The words could be located any where in the string:
Given: "but boys run fast boys are strong" or
"boys boys"
expected result: true, since there are 2 repeats of the word "boys"

\b matches word boundaries
\w+ will match 1 or more word characters
( ... ) creates a group for matches
\1 will match the contents of matching group #1.
Putting this together, you want a regex containing \b(\w+)\b.*\b\1\b
Plus appropriate backslash quoting, etc.
#guest27134 pointed out the above is not the complete solution, since the OP wanted true/false, not just the regex:
var result = a_string.match(/\b(\w+)\b.*\b\1\b/g) !== null
Or, even shorter, as suggested by O.P.:
var result = /\b(\w+)\b.*\b\1\b/g.test(myStr)

Taking into consideration that spaces are there after each word
var string = "but boys run fast boys are strong";
var strArray= string.split(" ");
var unique = [];
for(var i =0; i< strArray.length; i++)
{
eval(unique[strArray] = new Object());
}

If you know what word you're testing for duplicates you could use regular expressions like this str.match(/boys/g).length > 1 to test if that word shows up more than once (assuming your string is in the str variable).

Hello, here it is:
var temp = "but boys run fast boys are strong";
var count = (temp.match(/boys/g) || []).length;
console.log(count);
I hope I have helped!

You can count the occurrence of each matched word
var str = "but boys run fast boys are strong";
var matches = str.split(/\s/);
var res = matches.map(function(match) {
return str.match(new RegExp(match, "g")).length;
});
var bool = res.some(function(len) {return len > 1}));
console.log(bool);
for (var i = 0; i < matches.length; i++) {
if (res[i] > 1) console.log(matches[i], i);
}

Hello!
Here's an example with the exact answer you need:
var keyword = "boys"
var temp = "but boys run fast boys are strong";
var regex = new RegExp(keyword, "g");
var count = (temp.match(regex) || []).length;
if (count > 0) {
console.log("true, since there are " + count + " repeats of the word '" + keyword + "'");
} else {
console.log("false, not found.");
}
I hope I have helped!

Remove all occurrences except last?

I want to remove all occurrences of substring = . in a string except the last one.
E.G:
1.2.3.4
should become:
123.4

You can use regex with positive look ahead,
"1.2.3.4".replace(/[.](?=.*[.])/g, "");

2-liner:
function removeAllButLast(string, token) {
/* Requires STRING not contain TOKEN */
var parts = string.split(token);
return parts.slice(0,-1).join('') + token + parts.slice(-1)
}
Alternative version without the requirement on the string argument:
function removeAllButLast(string, token) {
var parts = string.split(token);
if (parts[1]===undefined)
return string;
else
return parts.slice(0,-1).join('') + token + parts.slice(-1)
}
Demo:
> removeAllButLast('a.b.c.d', '.')
"abc.d"
The following one-liner is a regular expression that takes advantage of the fact that the * character is greedy, and that replace will leave the string alone if no match is found. It works by matching [longest string including dots][dot] and leaving [rest of string], and if a match is found it strips all '.'s from it:
'a.b.c.d'.replace(/(.*)\./, x => x.replace(/\./g,'')+'.')
(If your string contains newlines, you will have to use [.\n] rather than naked .s)

You can do something like this:
var str = '1.2.3.4';
var last = str.lastIndexOf('.');
var butLast = str.substring(0, last).replace(/\./g, '');
var res = butLast + str.substring(last);
Live example:
http://jsfiddle.net/qwjaW/

You could take a positive lookahead (for keeping the last dot, if any) and replace the first coming dots.
var string = '1.2.3.4';
console.log(string.replace(/\.(?=.*\.)/g, ''));

A replaceAllButLast function is more useful than a removeAllButLast function. When you want to remove just replace with an empty string:
function replaceAllButLast(str, pOld, pNew) {
var parts = str.split(pOld)
if (parts.length === 1) return str
return parts.slice(0, -1).join(pNew) + pOld + parts.slice(-1)
}
var test = 'hello there hello there hello there'
test = replaceAllButLast(test, ' there', '')
console.log(test) // hello hello hello there

Found a much better way of doing this. Here is replaceAllButLast and appendAllButLast as they should be done. The latter does a replace whilst preserving the original match. To remove, just replace with an empty string.
var str = "hello there hello there hello there"
function replaceAllButLast(str, regex, replace) {
var reg = new RegExp(regex, 'g')
return str.replace(reg, function(match, offset, str) {
var follow = str.slice(offset);
var isLast = follow.match(reg).length == 1;
return (isLast) ? match : replace
})
}
function appendAllButLast(str, regex, append) {
var reg = new RegExp(regex, 'g')
return str.replace(reg, function(match, offset, str) {
var follow = str.slice(offset);
var isLast = follow.match(reg).length == 1;
return (isLast) ? match : match + append
})
}
var replaced = replaceAllButLast(str, / there/, ' world')
console.log(replaced)
var appended = appendAllButLast(str, / there/, ' fred')
console.log(appended)
Thanks to #leaf for these masterpieces which he gave here.

You could reverse the string, remove all occurrences of substring except the first, and reverse it again to get what you want.

function formatString() {
var arr = ('1.2.3.4').split('.');
var arrLen = arr.length-1;
var outputString = '.' + arr[arrLen];
for (var i=arr.length-2; i >= 0; i--) {
outputString = arr[i]+outputString;
}
alert(outputString);
}
See it in action here: http://jsbin.com/izebay

var s='1.2.3.4';
s=s.split('.');
s.splice(s.length-1,0,'.');
s.join('');
123.4