I have a string that may contain several url links (http or https). I need a script that would remove all those URLs from the string completely and return that same string without them.
I tried so far:
var url = "and I said http://fdsadfs.com/dasfsdadf/afsdasf.html";
var protomatch = /(https?|ftp):\/\//; // NB: not '.*'
var b = url.replace(protomatch, '');
console.log(b);
but this only removes the http part and keeps the link.
How to write the right regex that it would remove everything that follows http and also detect several links in the string?
Thank you so much!
You can use this regex:
var b = url.replace(/(?:https?|ftp):\/\/[\n\S]+/g, '');
//=> and I said
This regex matches and removes any URL that starts with http:// or https:// or ftp:// and matches up to next space character OR end of input. [\n\S]+ will match across multi lines as well.
Did you search for a url parser regex? This question has a few comprehensive answers Getting parts of a URL (Regex)
That said, if you want something much simpler (and maybe not as perfect), you should remember to capture the entire url string and not just the protocol.
Something like
/(https?|ftp):\/\/[\.[a-zA-Z0-9\/\-]+/
should work better. Notice that the added half parses the rest of the URL after the protocol.
Related
I have been trying to make a Reg Exp to match the URL with specific domain name.
So if i want to check if this url is from example.com
what reg exp should be the best?
This reg exp should match following type of URLs:
http://api.example.com/...
http://preview.example.com/...
http://www.example.com/...
http://purhcase.example.com/...
Just simple rule, like http://{something}.example.com/{something} then should pass.
Thank you.
I think this is what you're looking for: (https?:\/\/(.+?\.)?example\.com(\/[A-Za-z0-9\-\._~:\/\?#\[\]#!$&'\(\)\*\+,;\=]*)?).
It breaks down as follows:
https?:\/\/ to match http:// or https:// (you didn't mention https, but it seemed like a good idea).
(.+?\.)? to match anything before the first dot (I made it optional so that, for example, http://example.com/ would be found
example\.com (example.com, of course);
(\/[A-Za-z0-9\-\._~:\/\?#\[\]#!$&'\(\)\*\+,;\=]*)?): a slash followed by every acceptable character in a URL; I made this optional so that http://example.com (without the final slash) would be found.
Example: https://regex101.com/r/kT8lP2/1
Use indexOf javascript API. :)
var url = 'http://api.example.com/api/url';
var testUrl = 'example.com';
if(url.indexOf(testUrl) !== -1) {
console.log('URL passed the test');
} else{
console.log('URL failed the test');
}
EDIT:
Why use indexOf instead of Regular Expression.
You see, what you have here for matching is a simple string (example.com) not a pattern. If you have a fixed string, then no need to introduce semantic complexity by checking for patterns.
Regular expressions are best suited for deciding if patterns are matched.
For example, if your requirement was something like the domain name should start with ex end with le and between start and end, it should contain alphanumeric characters out of which 4 characters must be upper case. This is the usecase where regular expression would prove beneficial.
You have simple problem so it's unnecessary to employ army of 1000 angels to convince someone who loves you. ;)
Use this:
/^[a-zA-Z0-9_.+-]+#(?:(?:[a-zA-Z0-9-]+\.)?[a-zA-Z]+\.)?
(domain|domain2)\.com$/g
To match the specific domain of your choice.
If you want to match only one domain then remove |domain2 from (domain|domain2) portion.
It will help you. https://www.regextester.com/94044
Not sure if this would work for your case, but it would probably be better to rely on the built in URL parser vs. using a regex.
var url = document.createElement('a');
url.href = "http://www.example.com/thing";
You can then call those values using the given to you by the API
url.protocol // (http:)
url.host // (www.example.com)
url.pathname // (/thing)
If that doesn't help you, something like this could work, but is likely too brittle:
var url = "http://www.example.com/thing";
var matches = url.match(/:\/\/(.[^\/]+)(.*)/);
// matches would return something like
// ["://example.com/thing", "example.com", "/thing"]
These posts could also help:
https://stackoverflow.com/a/3213643/4954530
https://stackoverflow.com/a/6168370
Good luck out there!
There are cases where the domain you're looking for could actually be found in the query section but not in the domain section: https://www.google.com/q=www.example.com
This answer would treat that case better.
See this example on regex101.
As you you pointed you only need example.com (write domain then escaped period then com), so use it in regex.
Example
UPDATED
See the answer below
In JavaScript, I want to extract a non-image url from a string e.g.
http://example.com
http://example.com/a.png
http://www.example.ccom/acd.php
http://www.example.com/b.jpg etc.
I would like to extract 1st and 3rd (non-image) URLs and ignore 2nd and 4th (image) URLs.
I tried the following which did not work
(https?:)?\/\/?[^\'"<>]+?^(\.(jpe?g|gif|png))
Which is the modification of the following Image URL Regular Expression (RE) to whom I added ^() (for not) for above snippet
(https?:)?//?[^\'"<>]+?\.(jpg|jpeg|gif|png)
Note: The RE in above examples is case-sensitive, if any clue for making RE case-insensitive
You can use a negative lookahead like these examples It will exclude anything with the string
assuming your urls are newline delimited like your example, something like this should work
(?!.*(jpg|jpeg|gif|png).*).*
EDIT: it looks like my example doesn't work, hopefully it is pointing oyu in the right direction at least
first removing the images:
var tmp = text.replace(/https?:\/\/[\S]+\.(png|jpeg|jpg|gif)/gi, '');
and then matching:
var m = tmp.match(/https?:\/\/[\S]+/gi);
console.log(m);
I am trying to test the pathname of the url, checking if pathname starts with privmsg as well as contains one of the words in the selection. And my quantifier is selecting that at least one word must be found.
New RegExp thanks to one of the answers and I extended it more.
var post = /(^\/privmsg\?).+(post|reply){1}(.*)?/;
My urls will look like
/privmsg?mode=post
/privmsg?mode=reply
/privmsg?mode=reply&p=2 //another way
Though we have other modes that I do not want. I need to just get the constant url beginning with privmsg and having at least post or reply in it. Can someone explain what is wrong with my regex string and if I used the quantifier incorrectly.
Problem now is that it is still coming out false...
You need to allow for arbitrary characters between ? and (post|reply) (i.e. mode=). E.g.:
var post = /^\/privmsg\?.+(post|reply){1}/g;
\/
|match any sequence of|
|1 or more characters |
You miss to include something for mode=.
With your regex you will match strings like /privmsg?post.
So alter your regex to include mode=:
^\/privmsg\?.*(post|reply)$
I'm trying to validate a field to allow relative and absolute urls. I'm using the regex from this post but it is allowing spaces in the url.
var urlRegex = new RegExp(/(\/?[\w-]+)(\/[\w-]+)*\/?|(((http|ftp|https):\/\/)?[\w-]+(\.[\w-]+)+([\w.,#?^=%&:\/~+#-]*[\w#?^=%&\/~+#-])?)/gi);
Example:
// this should work
this/will/work.aspx?say=hello
http://www.example.com/this/will/work.aspx?say=hello
// this shouldn't work but does
and/this will also work/even though it shouldn't
and/this-shouldn't/but it does/also
The code below is what I was originally using to validate just absolute urls and it was working perfectly. If I remember properly, I pulled it from the jquery source. If this could be modified to also accept relative urls that would be perfect, but this is out of my league.
var urlRegex = new RegExp(/^(https?|ftp):\/\/(((([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(%[\da-f]{2})|[!\$&'\(\)\*\+,;=]|:)*#)?(((\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5])\.(\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5])\.(\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5])\.(\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5]))|((([a-z]|\d|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(([a-z]|\d|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])*([a-z]|\d|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])))\.)+(([a-z]|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(([a-z]|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])*([a-z]|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])))\.?)(:\d*)?)(\/((([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(%[\da-f]{2})|[!\$&'\(\)\*\+,;=]|:|#)+(\/(([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(%[\da-f]{2})|[!\$&'\(\)\*\+,;=]|:|#)*)*)?)?(\?((([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(%[\da-f]{2})|[!\$&'\(\)\*\+,;=]|:|#)|[\uE000-\uF8FF]|\/|\?)*)?(\#((([a-z]|\d|-|\.|_|~|[\u00A0-\uD7FF\uF900-\uFDCF\uFDF0-\uFFEF])|(%[\da-f]{2})|[!\$&'\(\)\*\+,;=]|:|#)|\/|\?)*)?$/i);
I think you just need to anchor the pattern so that it has to match the whole string:
var urlRegex = /^(\/?[\w-]+)(\/[\w-]+)*\/?|(((http|ftp|https):\/\/)?[\w-]+(\.[\w-]+)+([\w.,#?^=%&:\/~+#-]*[\w#?^=%&\/~+#-])?)$/gi;
The leading ^ and trailing $ means that the pattern has to match the entire string instead of just some part of it.
edit that said, the pattern has other problems. First, those HTML entities for & (&) need to be just "&". The slashes don't need to be escaped in [] groups, and we don't need the "g" suffix. That leaves us with:
var urlRegex = /^(?:(\/?[\w-]+)(\/[\w-]+)*\/?|(((http|ftp|https):\/\/)?[\w-]+(\.[\w-]+)*([\w.,#?^=%&:/~+#-]*[\w#?^=%&/~+#-])?))$/i;
edit again - oops also need to wrap the whole thing.
I wrote an article about URI validation complete with code snippets for all the various URI components as defined by RFC3986 here:
Regular Expression URI Validation
You may find what you are looking for there. Note however that almost any string represents a valid URI - even an empty string!
I use this js code to match a hostname from a string:
url.match(/:\/\/(www\.)?(.[^/:]+)/);
This works when the url has protocol:// at the beginning. For example:
This works fine:
var url = "http://domain.com/page";
url.match(/:\/\/(www\.)?(.[^/:]+)/);
But this doesn't:
var url = "domain.com/page";
url.match(/:\/\/(www\.)?(.[^/:]+)/);
I have tried:
url.match(/(:\/\/)?(www\.)?(.[^/:]+)/);
And that matches fine the hostname when it doesn't contain protocol://, but when it does contains it it only returns the protocol and not the hostname.
How could I match the domain when it doesn't contains it?
I used this function from Steven Levithan, it parses urls quite decently.
Here's how you use this function
alert(parseUri("www.domain.com/foo").host)
OK before you have a brain meltdown from #xanatos answer here is a simple regex for basic needs. The other answers are more complete and handle more cases than this regex :
(?:(?:(?:\bhttps?|ftp)://)|^)([-A-Z0-9.]+)/
Group 1 will have your host name. URL parsing is a fragile thing to do with regexes. You were on the right track. You had two regexes that worked partially. I simply combined them.
Edit : I was tired yesterday night. Here is the regex for jscript
if (subject.match(/(?:(?:(?:\bhttps?|ftp):\/\/)|^)([\-a-z0-9.]+)\//i)) {
// Successful match
} else {
// Match attempt failed
}
This
var rx = /^(?:(?:ht|f)tp(?:s?)\:\/\/|~\/|\/)?(?:\w+:\w+#)?(?:(?:[-\w]+\.)+(?:com|org|net|gov|mil|biz|info|mobi|name|aero|jobs|museum|travel|[a-z]{2}))(?::[\d]{1,5})?(?:(?:(?:\/(?:[-\w~!$+|.,=]|%[a-f\d]{2})+)+|\/)+|\?|#)?(?:(?:\?(?:[-\w~!$+|.,*:]|%[a-f\d{2}])+=?(?:[-\w~!$+|.,*:=]|%[a-f\d]{2})*)(?:&(?:[-\w~!$+|.,*:]|%[a-f\d{2}])+=?(?:[-\w~!$+|.,*:=]|%[a-f\d]{2})*)*)*(?:#(?:[-\w~!$+|.,*:=]|%[a-f\d]{2})*)?$/;
should be the uber-url parsing regex :-) Taken from here http://flanders.co.nz/2009/11/08/a-good-url-regular-expression-repost/
Test here: http://jsfiddle.net/Qznzx/1/
It shows the uselessness of regexes.
This might be a bit more complex than necessary but it seems to work:
^((?:.+?:\/\/)?(?:.[^/:]+)+)$
A non-capturing group for the protocol. From the start of the string
match any number of characters until a :. There may be zero or one
protocol.
A non-capturing group for the rest of the url. This part must exist.
Group it all up in single group.