autolinker
Version:
Utility to automatically link the URLs, email addresses, phone numbers, hashtags, and mentions (Twitter, Instagram) in a given block of text/HTML
189 lines • 7.56 kB
JavaScript
import { isDigitChar, isAsciiLetterChar, isAlphaNumericOrMarkChar, isUrlSuffixAllowedSpecialChar, isUrlSuffixNotAllowedAsFinalChar, } from '../char-utils';
import { tldRegex } from './known-tlds';
/**
* Regular expression to match an http:// or https:// scheme.
*/
export var httpSchemeRe = /https?:\/\//i;
/**
* Regular expression to match an http:// or https:// scheme as the prefix of
* a string.
*/
export var httpSchemePrefixRe = new RegExp('^' + httpSchemeRe.source, 'i');
/**
* A regular expression used to determine the schemes we should not autolink
*/
export var invalidSchemeRe = /^(javascript|vbscript):/i;
// A regular expression used to determine if the URL is a scheme match (such as
// 'http://google.com', and as opposed to a "TLD match"). This regular
// expression is used to parse out the host along with if the URL has an
// authority component (i.e. '//')
//
// Capturing groups:
// 1. '//' if the URL has an authority component, empty string otherwise
// 2. The host (if one exists). Ex: 'google.com'
//
// See https://www.rfc-editor.org/rfc/rfc3986#appendix-A for terminology
export var schemeUrlRe = /^[A-Za-z][-.+A-Za-z0-9]*:(\/\/)?([^:/]*)/;
// A regular expression used to determine if the URL is a TLD match (such as
// 'google.com', and as opposed to a "scheme match"). This regular
// expression is used to help parse out the TLD (top-level domain) of the host.
//
// See https://www.rfc-editor.org/rfc/rfc3986#appendix-A for terminology
export var tldUrlHostRe = /^(?:\/\/)?([^/#?:]+)/; // optionally prefixed with protocol-relative '//' chars
/**
* Determines if the given character code represents a character that may start
* a scheme (ex: the 'h' in 'http')
*/
export var isSchemeStartChar = isAsciiLetterChar; // Equivalent to checking the RegExp `/[A-Za-z]/`, but aliased for clarity and maintainability
/**
* Determines if the given character is a valid character in a scheme (such as
* 'http' or 'ssh+git'), but only after the start char (which is handled by
* {@link isSchemeStartChar}.
*/
export function isSchemeChar(charCode) {
return (isAsciiLetterChar(charCode) ||
isDigitChar(charCode) ||
charCode === 43 /* Char.Plus */ || // '+'
charCode === 45 /* Char.Dash */ || // '-'
charCode === 46 /* Char.Dot */ // '.'
);
}
/**
* Determines if the character can begin a domain label, which must be an
* alphanumeric character and not an underscore or dash.
*
* A domain label is a segment of a hostname such as subdomain.google.com.
*/
export var isDomainLabelStartChar = isAlphaNumericOrMarkChar; // alias function for clarity
/**
* Determines if the character is part of a domain label (but not a domain label
* start character).
*
* A domain label is a segment of a hostname such as subdomain.google.com.
*/
export function isDomainLabelChar(charCode) {
return charCode === 95 /* Char.Underscore */ || isDomainLabelStartChar(charCode);
}
/**
* Determines if the character is a path character ("pchar") as defined by
* https://tools.ietf.org/html/rfc3986#appendix-A
*
* pchar = unreserved / pct-encoded / sub-delims / ":" / "@"
*
* unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~"
* pct-encoded = "%" HEXDIG HEXDIG
* sub-delims = "!" / "$" / "&" / "'" / "(" / ")"
* / "*" / "+" / "," / ";" / "="
*
* Note that this implementation doesn't follow the spec exactly, but rather
* follows URL path characters found out in the wild (spec might be out of date?)
*/
export function isPathChar(charCode) {
return (isAlphaNumericOrMarkChar(charCode) ||
isUrlSuffixAllowedSpecialChar(charCode) ||
isUrlSuffixNotAllowedAsFinalChar(charCode) // characters in addition to those allowed by isUrlSuffixAllowedSpecialChar()
);
}
/**
* Determines if the character given may begin the "URL Suffix" section of a
* URI (i.e. the path, query, or hash section). These are the '/', '?' and '#'
* characters.
*
* See https://tools.ietf.org/html/rfc3986#appendix-A
*/
export function isUrlSuffixStartChar(charCode) {
return (charCode === 47 /* Char.Slash */ || // '/'
charCode === 63 /* Char.Question */ || // '?'
charCode === 35 /* Char.NumberSign */ // '#'
);
}
/**
* Determines if the top-level domain (TLD) read in the host is a known TLD.
*
* Example: 'com' would be a known TLD (for a host of 'google.com'), but
* 'local' would not (for a domain name of 'my-computer.local').
*/
export function isKnownTld(tld) {
return tldRegex.test(tld.toLowerCase()); // make sure the tld is lowercase for the regex
}
/**
* Determines if the given `url` is a valid scheme-prefixed URL.
*/
export function isValidSchemeUrl(url) {
// If the scheme is 'javascript:' or 'vbscript:', these link
// types can be dangerous. Don't link them.
if (invalidSchemeRe.test(url)) {
return false;
}
var schemeMatch = url.match(schemeUrlRe);
if (!schemeMatch) {
return false;
}
var isAuthorityMatch = !!schemeMatch[1];
var host = schemeMatch[2];
if (isAuthorityMatch) {
// Any match that has an authority ('//' chars) after the scheme is
// valid, such as 'http://anything'
return true;
}
// If there's no authority ('//' chars), check that we have a hostname
// that looks valid.
//
// The host must contain at least one '.' char and have a domain label
// with at least one letter to be considered valid.
//
// Accept:
// - git:domain.com (scheme followed by a host
// Do not accept:
// - git:something ('something' doesn't look like a host)
// - version:1.0 ('1.0' doesn't look like a host)
if (host.indexOf('.') === -1 || !/[A-Za-z]/.test(host)) {
// `letterRe` RegExp checks for a letter anywhere in the host string
return false;
}
return true;
}
/**
* Determines if the given `url` is a match with a valid TLD.
*/
export function isValidTldMatch(url) {
// TLD URL such as 'google.com', we need to confirm that we have a valid
// top-level domain
var tldUrlHostMatch = url.match(tldUrlHostRe);
if (!tldUrlHostMatch) {
// At this point, if the URL didn't match our TLD re, it must be invalid
// (highly unlikely to happen, but just in case)
return false;
}
var host = tldUrlHostMatch[0];
var hostLabels = host.split('.');
if (hostLabels.length < 2) {
// 0 or 1 host label, there's no TLD. Ex: 'localhost'
return false;
}
var tld = hostLabels[hostLabels.length - 1];
if (!isKnownTld(tld)) {
return false;
}
// TODO: Implement these conditions for TLD matcher:
// (
// this.longestDomainLabelLength <= 63 &&
// this.domainNameLength <= 255
// );
return true;
}
// Regular expression to confirm a valid IPv4 address (ex: '192.168.0.1')
// TODO: encode this into the state machine so that we don't need to run this
// regexp separately to confirm the match
var ipV4Re = /^(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)$/;
// Regular expression used to split the IPv4 address itself from any port/path/query/hash
var ipV4PartRe = /[:/?#]/;
/**
* Determines if the given URL is a valid IPv4-prefixed URL.
*/
export function isValidIpV4Address(url) {
// Grab just the IP address
var ipV4Part = url.split(ipV4PartRe, 1)[0]; // only 1 result needed
return ipV4Re.test(ipV4Part);
}
//# sourceMappingURL=uri-utils.js.map