org.apache.brooklyn.util.net

Class Urls

java.lang.Object

org.apache.brooklyn.util.net.Urls

public class Urls
extends java.lang.Object

Constructor Summary

Constructors

Constructor and Description
Urls()

Method Summary

Modifier and Type	Method and Description
static java.lang.String	asDataUrlBase64(MediaType type, byte[] bytes) Creates a "data:..." scheme URL for use with supported parsers, using Base64 encoding.
static java.lang.String	asDataUrlBase64(java.lang.String data) as asDataUrlBase64(String) with plain text
static java.lang.String	decode(java.lang.String text) As encode(String)
static java.lang.String	encode(java.lang.String text) Encodes the string suitable for use in a URL, using default UTF-8 (version of URLEncoder.encode that does not throw checked exception)
static java.lang.String	getBasename(java.lang.String url) return the last segment of the given url before any '?', e.g.
static java.lang.String	getProtocol(java.lang.String url) returns the protocol (e.g.
static boolean	isDirectory(java.lang.String fileUrl)
static boolean	isUrlWithProtocol(java.lang.String x) returns true if the string begins with a non-empty string of letters followed by a colon, i.e.
static boolean	isUrlWithProtocol(java.lang.String x, boolean allowSpacesAfterCharAfterColon, boolean allowMultiline) as isUrlWithProtocol(String) but configurable to be strict (false, false) or allow newline chars (if e.g.
static java.lang.String	mergePaths(java.lang.String... items) returns the items with exactly one "/" between items (whether or not the individual items start or end with /), except where character before the / is a : (url syntax) in which case it will permit multiple (will not remove any).
static <any>	stringToUriFunction()
static <any>	stringToUrlFunction()
static java.io.File	toFile(java.lang.String fileUrl)
static java.net.URI	toUri(java.lang.String uri) creates a URI, preserving null and propagating exceptions *unchecked*
static java.net.URI	toUri(java.net.URL url) creates a URI, preserving null and propagating exceptions *unchecked*
static java.net.URL	toUrl(java.lang.String url) creates a URL, preserving null and propagating exceptions *unchecked*
static java.net.URL	toUrl(java.net.URI uri) creates a URL, preserving null and propagating exceptions *unchecked*

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Urls

public Urls()

Method Detail

stringToUriFunction

public static <any> stringToUriFunction()

stringToUrlFunction

public static <any> stringToUrlFunction()

toUrl

public static final java.net.URL toUrl(@Nullable
                                       java.lang.String url)

creates a URL, preserving null and propagating exceptions *unchecked*

toUrl

public static final java.net.URL toUrl(@Nullable
                                       java.net.URI uri)

creates a URL, preserving null and propagating exceptions *unchecked*

toUri

public static final java.net.URI toUri(@Nullable
                                       java.lang.String uri)

creates a URI, preserving null and propagating exceptions *unchecked*

toUri

public static final java.net.URI toUri(@Nullable
                                       java.net.URL url)

creates a URI, preserving null and propagating exceptions *unchecked*

isUrlWithProtocol

public static boolean isUrlWithProtocol(java.lang.String x)

returns true if the string begins with a non-empty string of letters followed by a colon, i.e. "protocol:" returns true, but "/" returns false, followed by a non-space character. thereafter it is not strict, it can have spaces, but not newlines; see isUrlWithProtocol(String, boolean, boolean)

isUrlWithProtocol

public static boolean isUrlWithProtocol(java.lang.String x,
                                        boolean allowSpacesAfterCharAfterColon,
                                        boolean allowMultiline)

as isUrlWithProtocol(String) but configurable to be strict (false, false) or allow newline chars (if e.g. in an unescaped argument)

mergePaths

public static java.lang.String mergePaths(java.lang.String... items)

returns the items with exactly one "/" between items (whether or not the individual items start or end with /), except where character before the / is a : (url syntax) in which case it will permit multiple (will not remove any). Throws a NullPointerException if any elements of 'items' is null.

encode

public static java.lang.String encode(java.lang.String text)

Encodes the string suitable for use in a URL, using default UTF-8 (version of URLEncoder.encode that does not throw checked exception)

decode

public static java.lang.String decode(java.lang.String text)

As encode(String)

getProtocol

public static java.lang.String getProtocol(java.lang.String url)

returns the protocol (e.g. http) if one appears to be specified, or else null; 'protocol' here should consist of 2 or more _letters_ only followed by a colon (2 required to prevent c:\xxx being treated as a url)

getBasename

public static java.lang.String getBasename(java.lang.String url)

return the last segment of the given url before any '?', e.g. the filename or last directory name in the case of directories (cf unix `basename`)

isDirectory

public static boolean isDirectory(java.lang.String fileUrl)

toFile

public static java.io.File toFile(java.lang.String fileUrl)

asDataUrlBase64

public static java.lang.String asDataUrlBase64(java.lang.String data)

as asDataUrlBase64(String) with plain text

asDataUrlBase64

public static java.lang.String asDataUrlBase64(MediaType type,
                                               byte[] bytes)

Creates a "data:..." scheme URL for use with supported parsers, using Base64 encoding. (But note, by default Java's URL is not one of them, although Brooklyn's ResourceUtils does support it.)

It is not necessary (at least for Brookyn's routines) to base64 encode it, but recommended as that is likely more portable and easier to work with if odd characters are included.

It is worth noting that Base64 uses '+' which can be replaced by ' ' in some URL parsing. But in practice it does not seem to cause issues. An alternative is to use BaseEncoding#base64Url() but it is not clear how widely that is supported (nor what parameter should be given to indicate that type of encoding, as the spec calls for 'base64'!)

null type means no type info will be included in the URL.