io.inversion.utils

Class Path

java.lang.Object

io.inversion.utils.Path

public class Path
extends java.lang.Object

A case insensitive utility abstraction for working with forward slash based paths /like/you/find/in/urls.

When working with Paths, leading and trailing slashes are completely disregarded. Any leading and trailing slashes in the documentation are illustrative only and could be excluded to achieve the same result. Multiple consecutive slashes are treated as a single slash. There will never be an empty string or null path part.

In addition to representing concrete paths, such as a file path or url path, a Path object may contain wildcards, regular expressions, and variable name bindings which are used when comparing a variablized abstract paths with a concrete path.

Paths are primarily used to configure Rules (Engine, Api, Endpoint and Action are all subclasses of Rule) to match against inbound Request urls to determine how the Request will be processed.

Paths can be variablized as follows:

• /animals/* - will match anything that starts with "animals/". "*" matches any number of path segments including zero segments (meaning "animals/" alone will match and so will "animals/dogs/fido"). "*" wildcards are only valid as the last segment in a path.
• /animals/dogs/:dogName - if a path segment starts with a ":" it indicates that the value can be anything but a value is required and should be mapped to the corresponding variable name, in this case "dogName", by whoever is doing the path matching.
• /animals/dogs/{fido|ralph|jackie} - if the path segment is wrapped in "{}" the contents are considered a regular expression Pattern for matching
• /animals/dogs/${fido|ralph|jackie} - a '$' can prefix '{}' as syntatic sugar some may be familiar with.
• /animals/dogs/{dogName:fido|ralph|jackie} - again a regex, this time with a variable binding to "dogName"
• /animals/[{dogs|cats|snakes}]/:animalName/* - if something is wrapped in square brackets "[]" that segment and all subsequent segments are optional. If the segments exists in the path being compared to, they must match the supplied rules, but if the comparison path ends right before the first optional segment, the paths still match.

When used in the context of a Api configuration you may see something like this:

  Engine e = new Engine().withIncludeOn(null, new Path("/apis"));
                         .withApi(new Api().withIncludeOn(null, new Path("bookstore/{storeName:johnsBestBooks|carolsBooksOnMain}"))
                                           .withEndpoint(new Endpoint().withIncludeOn(null, new Path("categories/:category/"))
                                                                       .withAction(new BrowseCategoriesAction().withIncludeOn(null, new Path("[:subcategory]/*")))));
 

See Also:

io.inversion.Rule.withIncludeOn, io.inversion.Rule.withExcludeOn

Constructor Summary

Constructors

Constructor and Description
Path() Creates an empty Path
Path(java.util.List<java.lang.String> parts) Convenience overload of Path(String...).
Path(Path path) Creates a clone of the supplied Path
Path(java.lang.String... part) Constructs a Path based on all of the supplied parts.

Method Summary

Modifier and Type	Method and Description
void	add(java.lang.String parts) Adds part to the end of the Path.
boolean	equals(java.lang.Object o)
Path	extract(java.util.Map params, Path toMatch) Convenience overloading of extract(Map, Path, boolean) with greedy = true.
Path	extract(java.util.Map params, Path matchingConcretePath, boolean greedy) Consumes the matching parts of matchingConcretePath and extracts any named variable to matchingConcretePath if this any of this paths parts contain variable bindings.
java.lang.String	first() Simple way to pull the first element of the path without having to check for size() > 0 first.
java.lang.String	get(int index) Simple way to get element at index without haveint to check for size() > index first.
java.lang.String	getVarName(int index) Extracts a variable name form the path expression if index exists and has a var name.
boolean	isOptional(int index) Square brackets, '[]', indicate that a path path (and by implication, all following parts) are considered optional for path matching.
boolean	isStatic(int index) Checks to see if the value at index is a wildcard, a variable, or is optional.
boolean	isVar(int index) Check to see if the value at index starts with '${', '{', ':' after removing any leading '[' characters.
boolean	isWildcard(int index) Check if the path part at index is equal to '*' without having to check if size() < index first.
java.lang.String	last() Simple way to pull the last element of the path without having to check for size() > 0 first.
boolean	matches(Path concretePath) Checks if this Path is as case insensitive match, including any optional rules, wildcards, and regexes to concretePath.
boolean	matches(java.lang.String toMatch) Convenience overloading of matches(Path).
java.util.List<java.lang.String>	parts() Gets the path parts as a List.
java.lang.String	remove(int index) Simple way to remove the path part at index without having to check for size() < index first.
int	size()
boolean	startsWith(java.util.List<java.lang.String> partsToMatch) Performs a case insensitive string match between this Path and pathsToMatch.
Path	subpath(int fromIndex, int toIndex) Creates a new sub Path.
java.lang.String	toString()

Methods inherited from class java.lang.Object

clone, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

Path

public Path()

Creates an empty Path

Path

public Path(Path path)

Creates a clone of the supplied Path

Parameters:

path - the Path to be cloned

Path

public Path(java.lang.String... part)

Constructs a Path based on all of the supplied parts.

The strings in part may themselves contain "/" characters and will be split into multiple parts correspondingly. Meaning Path p = new Path("part1", "part2/part3", "/part4/", "////part5//part6/", "part7") is valid and would result in a Path with parts "part1", "part2", "part3", "part4", "part5", "part6", "part7".

Parameters:

path - an array of path part strings

Path

public Path(java.util.List<java.lang.String> parts)

Convenience overload of Path(String...).

Parameters:

parts - an list of path part strings

Method Detail

parts

public java.util.List<java.lang.String> parts()

Gets the path parts as a List.

Method signature could easily have been "asList()"

Returns:

a new list with the individual path parts n the originally defined case.

first

public java.lang.String first()

Simple way to pull the first element of the path without having to check for size() > 0 first.

Returns:

the first element in the path if it exists otherwise null

last

public java.lang.String last()

Simple way to pull the last element of the path without having to check for size() > 0 first.

Returns:

the last element in the path if it exists otherwise null

get

public java.lang.String get(int index)

Simple way to get element at index without haveint to check for size() > index first.

Parameters:

index - the index of the path part to retrive

Returns:

the path part at index if it exists otherwise null

add

public void add(java.lang.String parts)

Adds part to the end of the Path.

The parts is exploded via Utils.explode('/', part) first so while the part arg is a single value, it could result in multiple additions.

Parameters:

part -

remove

public java.lang.String remove(int index)

Simple way to remove the path part at index without having to check for size() < index first.

Parameters:

index - the index of the path part to remove

Returns:

the path part previously located at index if it existed otherwise null

startsWith

public boolean startsWith(java.util.List<java.lang.String> partsToMatch)

Performs a case insensitive string match between this Path and pathsToMatch.

Wildcards and regular expressions are not supported in this method, only straight case insensitive string comparison.

Parameters:

partsToMatch -

Returns:

true if each index of partsToMatch is a case insensitive match to this Path at the same index otherwise false.

size

public int size()

Returns:

the number of parts in the Path

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

Returns:

a pretty printed "/" separated path string representation

equals

public boolean equals(java.lang.Object o)

Overrides:

equals in class java.lang.Object

Returns:

true of the objects string representations match

subpath

public Path subpath(int fromIndex,
                    int toIndex)
             throws java.lang.ArrayIndexOutOfBoundsException

Creates a new sub Path.

Parameters:

fromIndex - low endpoint (inclusive) of the subList

toIndex - high endpoint (exclusive) of the subList

Returns:

a subpath from fromIndex (inclusive) to toIndex (exclusive)

Throws:

java.lang.IndexOutOfBoundsException

java.lang.ArrayIndexOutOfBoundsException

isStatic

public boolean isStatic(int index)

Checks to see if the value at index is a wildcard, a variable, or is optional.

Parameters:

index - the path part to check

Returns:

true if the path part at index is a '*' char or starts with '[', '{' or ':'

isWildcard

public boolean isWildcard(int index)

Check if the path part at index is equal to '*' without having to check if size() < index first.

Parameters:

index -

Returns:

true if the path part at index

isVar

public boolean isVar(int index)

Check to see if the value at index starts with '${', '{', ':' after removing any leading '[' characters.

Parameters:

index -

Returns:

true if the value exists and is variableized but not a wildcard, false otherwise.

getVarName

public java.lang.String getVarName(int index)

Extracts a variable name form the path expression if index exists and has a var name.

'varName' would be extracted from getVarName(1) for the following paths.

• /part/:varName/
• /part/{varNam:regex}/
• /part/${varNam:regex}/

Square brackets indicating a path component is optioanl don't impact retrieval of the var name so the following would return the same as there above counterparts:

• /part/[:varName]/
• /[part]/{varNam:regex}]/
• /[part]/[${varNam:regex}]/

Parameters:

index -

Returns:

the variable name binding for the parth part at index if it exists

isOptional

public boolean isOptional(int index)

Square brackets, '[]', indicate that a path path (and by implication, all following parts) are considered optional for path matching.

For example: new Path("first/second/[third]/").matches(new Path("first/second/third"))

Parameters:

index -

Returns:

true if the path part at index exists and starts with '[' and ends with ']'

matches

public boolean matches(java.lang.String toMatch)

Convenience overloading of matches(Path).

Parameters:

toMatch -

Returns:

true if the paths match

matches

public boolean matches(Path concretePath)

Checks if this Path is as case insensitive match, including any optional rules, wildcards, and regexes to concretePath.

As the name implies concretePath is considered to be a literal path not containing optionals, wildcards, and regexes.

As also documented above:

• paths can end with a "/*" character meaning this and all following parts are optional and unconstrained
• wrapping a path part in square brackets, '[]' indicates that it and all following part parts are optional.
• wrapping a path part in '${}' or '${}' indicates that this path part should match via a regular expression such as '${[0-9a-fA-F]{1,8}}' to match a 1 to 8 character alpha numeric part
• you can bind a variable name to a regex by preceding the regex with a name and a colon '/${id:[0-9a-fA-F]{1,8}}/'
• starting a path part with a ':' such as '/:id/' is functionally equivalent to '/${id:.*}'

All non regex comparisons are performed with String.equalsIgnoreCase.

All regexes are compiled with Pattern.CASE_INSENSITIVE.

Parameters:

concretePath -

Returns:

true if this path matches concretePath

extract

public Path extract(java.util.Map params,
                    Path toMatch)

Convenience overloading of extract(Map, Path, boolean) with greedy = true.

See Also:

extract(Map, Path, boolean)

extract

public Path extract(java.util.Map params,
                    Path matchingConcretePath,
                    boolean greedy)

Consumes the matching parts of matchingConcretePath and extracts any named variable to matchingConcretePath if this any of this paths parts contain variable bindings.

If greedy is true, the match will consume through matching optional path parts.

If greedy is false, variable bindings in any optional paths parts will be extracted but the parts will not be removed from matchingConcretePath.

Here is an example:

   Map params = new HashMap();
   Path engineMatch = new Path("apis/myapi/*");
   Path apiMatch = new Path("${version:v1|v2}/:tenant")
   Path endpointMatch = new Path("[${collection:books|categories|orders}]/*");
   Path actionMatch = new Path("[:orderId]/*");

   Path requestPath = new Path("/apis/myapi/v2/bobsBooks/orders/67890");

   engineMatch.extract(requestPath);
   // params is empty, requestPath is now 'v2/bobsBooks/orders/67890'

   apiMatch.extract(requestPath);
   //version=v2 and tenant=bobsBooks have been added to params and requestPath is now 'orders/67890'

   endpointMatch.extract(requestPath);
   //collection=orders has been added to params and requestPath is now '67890'

   actionMatch.extract(requestPath);
   //orderId=67890 has been added to params and requestPath is now empty.
 

The above example is very similar to how an Engine processes paths when selecting an Api, Endpoint, and Actions to run.

Engine will also add the params to the Request Url params as if they had been supplied as name value pairs by the caller on the query string.

Parameters:

params -

matchingConcretePath -

greedy -

Returns:

the same Path object that was passed in but potentially now shorter as matching segments may have been consumed

See Also:

io.inversion.Engine#service(io.inversion.Request, io.inversion.Response)}, io.inversion.Chain#next}