Develop Reference

Regular Expressions: Positive and Negative Lookahead Solution

I was doing Freecodecamp RegEx challange, which's about:
Use lookaheads in the pwRegex to match passwords that are greater than 5 characters long, do not begin with numbers, and have two consecutive digits
So far the solution I found passed all the tests was:
/^[a-z](?=\w{5,})(?=.*\d{2}\.)/i
However, when I tried
/^[a-z](?=\w{5,})(?=\D*\d{2,}\D*)/i
I failed the test trying to match astr1on11aut (but passed with astr11on1aut). So can someone help me explaining how ?=\D*\d{2,}\D* failed this test?

So can someone help me explaining how (?=\D*\d{2,}\D*) failed this
test?
Using \D matches any char except a digit, so from the start of the string you can not pass single digit to get to the 2 digits.
Explanation
astr1on11aut astr11on1aut
^ ^^
Can not pass 1 before reaching 11 Can match 2 digits first
Note
the expression could be shortened to (?=\D*\d{2}) as it does not matter if there are 2, 3 or more because 2 is the minimum and the \D* after is is optional.
the first expression ^[a-z](?=\w{5,})(?=.*\d{2}\.) can not match the example data because it expects to match a . literally after 2 digits.

Failed for the \d{2,}
Then let's have a look at the regexp match process
/^[a-z](?=\w{5,})
means start with [a-z], the length is at least 6, no problem.
(?=\D*\d{2,}\D*)
means the first letter should be followed by these parts:
[0 or more no-digit][2 or more digits][0 or more no-digits]
and lets have a look at the test case
astr1on11aut
// ^[a-z] match "a"
// length matchs
// [0 or more no-digit]: "str"
// d{2,} failed: only 1 digit
the ?= position lookahead means exact followed by.
The first regular expression is wrong, it should be
/^[a-z](?=\w{5,})(?=.*\d{2}.*)/i

When it comes to lookahead usage in regex, rememeber to anchor them all at the start of the pattern. See Lookarounds (Usually) Want to be Anchored. This will help you avoid a lot of issues when dealing with password regexps.
Now, let's see what requirements you have:
"are greater than 5 characters long" => (?=.{6}) (it requires any 6 chars other than lien break chars immediately to the right of the current location)
"do not begin with numbers" => (?!\d) (no digit allowed immediately to the right)
have two consecutive digits => (?=.*\d{2}) (any two digit chunk of text is required after any 0 or more chars other than line break chars as many as possible, immediately to the right of the current location).
So, what you may use is
^(?!\d)(?=.{6})(?=.*\d{2})
Note the most expensive part is placed at the end of the pattern so that it could fail quicker if the input is non-matching.
See the regex demo.

Javascript regex to make sure that string matches format x:y

I am trying to parse a string which has two numbers, both can be between 1 and 3 digits, and will have a colon in between. Here are some examples:
"1:1"
"1:12"
"12:1"
"123:12"
Also, the given string may also be invalid, and I need to detect if it is. My attempts so far to make sure the string is valid have looked like this: .match(/[1-9]\:[1-9]/);. But then I noticed that this wont work if a string such as this is inputted: "characters12:4characters". How would I go about validating the string to make sure it is in the format x:y?
Any help would be deeply appreciated.
Edit: numbers which contain 0 at the beginning is valid, but may not be given.

You may use
/^\d{1,3}:\d{1,3}$/
See the regex demo
Details
^ - start of a string
\d{1,3} - one, two or three digits (\d is a shorthand character class that matches any digit (it can also be written as a [0-9] character class) and {1,3} is a limited quantifier that matches1 to 3 consecutive occurrences of the quantified subpattern)
: - a colon
\d{1,3} - one, two or three digits
$ - end of the string.

Regex exact match on number, not digit

I have a scenario where I need to find and replace a number in a large string using javascript. Let's say I have the number 2 and I want to replace it with 3 - it sounds pretty straight forward until I get occurrences like 22, 32, etc.
The string may look like this:
"note[2] 2 2_ someothertext_2 note[32] 2finally_2222 but how about mymomsays2."
I want turn turn it into this:
"note[3] 3 3_ someothertext_3 note[32] 3finally_2222 but how about mymomsays3."
Obviously this means .replace('2','3') is out of the picture so I went to regex. I find it easy to get an exact match when I am dealing with string start to end ie: /^2$/g. But that is not what I have. I tried grouping, digit only, wildcards, etc and I can't get this to match correctly.
Any help on how to exactly match a number (where 0 <= number <= 500 is possible, but no constraints needed in regex for range) would be greatly appreciated.

The task is to find (and replace) "single" digit 2, not embedded in
a number composed of multiple digits.
In regex terms, this can be expressed as:
Match digit 2.
Previous char (if any) can not be a digit.
Next char (if any) can not be a digit.
The regex for the first condition is straightforward - just 2.
In other flavours of regex, e.g. PCRE, to forbid the previous
char you could use negative lookbehind, but unfortunately Javascript
regex does not support it.
So, to circumvent this, we must:
Put a capturing group matching either start of text or something
other than a digit: (^|\D).
Then put regex matching just 2: 2.
The last condition, fortunately, can be expressed as negative lookahead,
because even Javascript regex support it: (?!\d).
So the whole regex is:
(^|\D)2(?!\d)
Having found such a match, you have to replace it with the content
of the first capturing group and 3 (the replacement digit).

You can use negative look-ahead:
(\D|^)2(?!\d)
Replace with: ${1}3
If look behind is supported:
(?<!\d)2(?!\d)
Replace with: 3

See regex in use here
(\D|\b)2(?!\d)
(\D|\b) Capture either a non-digit character or a position that matches a word boundary
(?!\d) Negative lookahead ensuring what follows is not a digit
Alternations:
(^|\D)2(?!\d) # Thanks to #Wiktor in the comments below
(?<!\d)2(?!\d) # At the time of writing works in Chrome 62+
const regex = /(\D|\b)2(?!\d)/g
const str = `note[2] 2 2_ someothertext_2 note[32] 2finally_2222 but how about mymomsays2.`
const subst = "$13"
console.log(str.replace(regex, subst))

Regex to force a specific length and at the same time not starting with a number

I am having trouble building a very complex Regex. These are my constraints:
Length: 8 to 10
Cannot start by a number neither by an underscore
Has at least one capital letter, one number and one character among
the following 3 special characters: _$£
I thought I did it when I got to this:
^([^0-9_])(?=.*[A-Z])(?=.*[0-9])(?=.*[$£_])[A-Za-z0-9$£_]{7,9}$
It correctly fails the 1st statement:
aDf123_ fails because its length is not 8
aDf123_aAAFF fails because its length is more than 10
It correctly fails the 2nd statement:
_aDf1234 fails because it starts with an underscore
1aDf1234 fails because it starts with a number
It correctly fails the 3rd statement:
aaDf1234 fails because it doesn't have any special character
aadf1$34 fails because it doens't have a capital letter
aaDf$£££ fails because it doesn't have a number
And finally, it passes through the validation with a valid string: aaDf$££5
What is the problem? The problem is that the following string fails while it should pass through the validation: Daaa$444
When the capital letter is the first letter my regex doesn't see it. How can I make it see it? I know it is related with the fact that my regex ends with {7,9} instead of {8,10} but I cannot help it because I have to define that it must not start with a number or underscore...

You need to put lookaheads after ^ and replace [^0-9_] with [A-Za-z$£]:
^(?=.*[A-Z])(?=.*[0-9])(?=.*[$£_])[A-Za-z$£][A-Za-z0-9$£_]{7,9}$
See this regex demo.
Else, the first [^0-9_] matches ^, &, etc. that is NOT a digit or _ and the lookaheads only look for the required patterns after the first char. So, any valid string that has a required char only at its beginning will fail.
Alternatively, turn the [^0-9_] into the negative (?![0-9_]) lookahead (and then you will need to replace the {7,9} with {8,10} at the end since the lookahead pattern is non-consuming):
^(?![0-9_])(?=.*[A-Z])(?=.*[0-9])(?=.*[$£_])[A-Za-z0-9$£_]{8,10}$
See this regex demo.

Regular Expression with optional elements in input string in javascript

Can anyone give me the regular expression for currency which have the following formats :
1000 - valid
1,000 - valid
1,000.00 or 1000.00 - valid.
This means, the number May or May Not contain a comma(,) separator every 3 digits.
The number May Or May Not contain a dot (.), and if it carries a dot(.) it should show atleast 1 number after the decimal place.
And lastly it should be numerical characters only. If I need to make my question clear kindly suggest.

/^\d{1,3}(?:(?:,\d{3})*|(?:\d{3})*)(?:\.\d{1,2})?$/
"Between one and three digits, then any number of groups of three digits prefixed by a comma or any number of groups of three digits not prefixed by said comma (disallowing a mix of the two kinds of groups), then an optional group of one or two digits prefixed by a dot."
Note: This regex assumes that you want to validate an entire string against the criteria outlined in your question. If you want to use it to find such numbers in a longer string, you will need to remove the ^ and $ from the beginning and end of the expression.

Something like so should work: (,?\d{3})+(\.\d{2})?. The regex will attempt to match a sequence of 3 digits precedeed by an optional comma, which is then, finally followed by an optional decimal point and 2 digits.
Please refer to this tutorial for more information.
EDIT: As per the comment below, the above regex can fail. I would recommend first using this regular expression: ^[\d][\d.,]+$ to make sure that you only have digits, thousand and decimal seperators. This regular expression will also make sure that the number starts with a digit, not with anything else. You could most likely have one regular expression which does everything, but it will most likely be quite complex.