Skip to content

Latest commit

 

History

History
829 lines (640 loc) · 32 KB

core-resources.adoc

File metadata and controls

829 lines (640 loc) · 32 KB

Resources

This chapter covers how Spring handles resources and how you can work with resources in Spring. It includes the following topics:

Introduction

Java’s standard java.net.URL class and standard handlers for various URL prefixes, unfortunately, are not quite adequate enough for all access to low-level resources. For example, there is no standardized URL implementation that may be used to access a resource that needs to be obtained from the classpath or relative to a ServletContext. While it is possible to register new handlers for specialized URL prefixes (similar to existing handlers for prefixes such as http:), this is generally quite complicated, and the URL interface still lacks some desirable functionality, such as a method to check for the existence of the resource being pointed to.

The Resource Interface

Spring’s Resource interface is meant to be a more capable interface for abstracting access to low-level resources. The following listing shows the Resource interface definition:

Java
public interface Resource extends InputStreamSource {

	boolean exists();

	boolean isOpen();

	URL getURL() throws IOException;

	File getFile() throws IOException;

	Resource createRelative(String relativePath) throws IOException;

	String getFilename();

	String getDescription();
}
Kotlin
interface Resource : InputStreamSource {

	fun exists(): Boolean

	val isOpen: Boolean

	val url: URL

	val file: File

	@Throws(IOException::class)
	fun createRelative(relativePath: String): Resource

	val filename: String

	val description: String
}

As the definition of the Resource interface shows, it extends the InputStreamSource interface. The following listing shows the definition of the InputStreamSource interface:

Java
public interface InputStreamSource {

	InputStream getInputStream() throws IOException;
}
Kotlin
interface InputStreamSource {

	val inputStream: InputStream
}

Some of the most important methods from the Resource interface are:

  • getInputStream(): Locates and opens the resource, returning an InputStream for reading from the resource. It is expected that each invocation returns a fresh InputStream. It is the responsibility of the caller to close the stream.

  • exists(): Returns a boolean indicating whether this resource actually exists in physical form.

  • isOpen(): Returns a boolean indicating whether this resource represents a handle with an open stream. If true, the InputStream cannot be read multiple times and must be read once only and then closed to avoid resource leaks. Returns false for all usual resource implementations, with the exception of InputStreamResource.

  • getDescription(): Returns a description for this resource, to be used for error output when working with the resource. This is often the fully qualified file name or the actual URL of the resource.

Other methods let you obtain an actual URL or File object representing the resource (if the underlying implementation is compatible and supports that functionality).

Spring itself uses the Resource abstraction extensively, as an argument type in many method signatures when a resource is needed. Other methods in some Spring APIs (such as the constructors to various ApplicationContext implementations) take a String which in unadorned or simple form is used to create a Resource appropriate to that context implementation or, via special prefixes on the String path, let the caller specify that a specific Resource implementation must be created and used.

While the Resource interface is used a lot with Spring and by Spring, it is actually very useful to use as a general utility class by itself in your own code, for access to resources, even when your code does not know or care about any other parts of Spring. While this couples your code to Spring, it really only couples it to this small set of utility classes, which serve as a more capable replacement for URL and can be considered equivalent to any other library you would use for this purpose.

Note
The Resource abstraction does not replace functionality. It wraps it where possible. For example, a UrlResource wraps a URL and uses the wrapped URL to do its work.

Built-in Resource Implementations

Spring includes the following Resource implementations:

UrlResource

UrlResource wraps a java.net.URL and can be used to access any object that is normally accessible with a URL, such as files, an HTTP target, an FTP target, and others. All URLs have a standardized String representation, such that appropriate standardized prefixes are used to indicate one URL type from another. This includes file: for accessing filesystem paths, http: for accessing resources through the HTTP protocol, ftp: for accessing resources through FTP, and others.

A UrlResource is created by Java code by explicitly using the UrlResource constructor but is often created implicitly when you call an API method that takes a String argument meant to represent a path. For the latter case, a JavaBeans PropertyEditor ultimately decides which type of Resource to create. If the path string contains well-known (to it, that is) prefix (such as classpath:), it creates an appropriate specialized Resource for that prefix. However, if it does not recognize the prefix, it assume the string is a standard URL string and creates a UrlResource.

ClassPathResource

This class represents a resource that should be obtained from the classpath. It uses either the thread context class loader, a given class loader, or a given class for loading resources.

This Resource implementation supports resolution as java.io.File if the class path resource resides in the file system but not for classpath resources that reside in a jar and have not been expanded (by the servlet engine or whatever the environment is) to the filesystem. To address this, the various Resource implementations always support resolution as a java.net.URL.

A ClassPathResource is created by Java code by explicitly using the ClassPathResource constructor but is often created implicitly when you call an API method that takes a String argument meant to represent a path. For the latter case, a JavaBeans PropertyEditor recognizes the special prefix, classpath:, on the string path and creates a ClassPathResource in that case.

FileSystemResource

This is a Resource implementation for java.io.File and java.nio.file.Path handles. It supports resolution as a File and as a URL.

ServletContextResource

This is a Resource implementation for ServletContext resources that interprets relative paths within the relevant web application’s root directory.

It always supports stream access and URL access but allows java.io.File access only when the web application archive is expanded and the resource is physically on the filesystem. Whether or not it is expanded and on the filesystem or accessed directly from the JAR or somewhere else like a database (which is conceivable) is actually dependent on the Servlet container.

InputStreamResource

An InputStreamResource is a Resource implementation for a given InputStream. It should be used only if no specific Resource implementation is applicable. In particular, prefer ByteArrayResource or any of the file-based Resource implementations where possible.

In contrast to other Resource implementations, this is a descriptor for an already-opened resource. Therefore, it returns true from isOpen(). Do not use it if you need to keep the resource descriptor somewhere or if you need to read a stream multiple times.

ByteArrayResource

This is a Resource implementation for a given byte array. It creates a ByteArrayInputStream for the given byte array.

It is useful for loading content from any given byte array without having to resort to a single-use InputStreamResource.

The ResourceLoader

The ResourceLoader interface is meant to be implemented by objects that can return (that is, load) Resource instances. The following listing shows the ResourceLoader interface definition:

Java
public interface ResourceLoader {

	Resource getResource(String location);
}
Kotlin
interface ResourceLoader {

	fun getResource(location: String): Resource
}

All application contexts implement the ResourceLoader interface. Therefore, all application contexts may be used to obtain Resource instances.

When you call getResource() on a specific application context, and the location path specified doesn’t have a specific prefix, you get back a Resource type that is appropriate to that particular application context. For example, assume the following snippet of code was run against a ClassPathXmlApplicationContext instance:

Java
Resource template = ctx.getResource("some/resource/path/myTemplate.txt");
Kotlin
val template = ctx.getResource("some/resource/path/myTemplate.txt")

Against a ClassPathXmlApplicationContext, that code returns a ClassPathResource. If the same method were run against a FileSystemXmlApplicationContext instance, it would return a FileSystemResource. For a WebApplicationContext, it would return a ServletContextResource. It would similarly return appropriate objects for each context.

As a result, you can load resources in a fashion appropriate to the particular application context.

On the other hand, you may also force ClassPathResource to be used, regardless of the application context type, by specifying the special classpath: prefix, as the following example shows:

Java
Resource template = ctx.getResource("classpath:some/resource/path/myTemplate.txt");
Kotlin
val template = ctx.getResource("classpath:some/resource/path/myTemplate.txt")

Similarly, you can force a UrlResource to be used by specifying any of the standard java.net.URL prefixes. The following pair of examples use the file and http prefixes:

Java
Resource template = ctx.getResource("file:///some/resource/path/myTemplate.txt");
Kotlin
val template = ctx.getResource("file:///some/resource/path/myTemplate.txt")
Java
Resource template = ctx.getResource("https://myhost.com/resource/path/myTemplate.txt");
Kotlin
val template = ctx.getResource("https://myhost.com/resource/path/myTemplate.txt")

The following table summarizes the strategy for converting String objects to Resource objects:

Table 1. Resource strings
Prefix Example Explanation

classpath:

classpath:com/myapp/config.xml

Loaded from the classpath.

file:

file:///data/config.xml

Loaded as a URL from the filesystem. See also FileSystemResource Caveats.

http:

https://myserver/logo.png

Loaded as a URL.

(none)

/data/config.xml

Depends on the underlying ApplicationContext.

The ResourceLoaderAware interface

The ResourceLoaderAware interface is a special callback interface which identifies components that expect to be provided with a ResourceLoader reference. The following listing shows the definition of the ResourceLoaderAware interface:

Java
public interface ResourceLoaderAware {

	void setResourceLoader(ResourceLoader resourceLoader);
}
Kotlin
interface ResourceLoaderAware {

	fun setResourceLoader(resourceLoader: ResourceLoader)
}

When a class implements ResourceLoaderAware and is deployed into an application context (as a Spring-managed bean), it is recognized as ResourceLoaderAware by the application context. The application context then invokes setResourceLoader(ResourceLoader), supplying itself as the argument (remember, all application contexts in Spring implement the ResourceLoader interface).

Since an ApplicationContext is a ResourceLoader, the bean could also implement the ApplicationContextAware interface and use the supplied application context directly to load resources. However, in general, it is better to use the specialized ResourceLoader interface if that is all you need. The code would be coupled only to the resource loading interface (which can be considered a utility interface) and not to the whole Spring ApplicationContext interface.

In application components, you may also rely upon autowiring of the ResourceLoader as an alternative to implementing the ResourceLoaderAware interface. The “traditional” constructor and byType autowiring modes (as described in [beans-factory-autowire]) are capable of providing a ResourceLoader for either a constructor argument or a setter method parameter, respectively. For more flexibility (including the ability to autowire fields and multiple parameter methods), consider using the annotation-based autowiring features. In that case, the ResourceLoader is autowired into a field, constructor argument, or method parameter that expects the ResourceLoader type as long as the field, constructor, or method in question carries the @Autowired annotation. For more information, see [beans-autowired-annotation].

Resources as Dependencies

If the bean itself is going to determine and supply the resource path through some sort of dynamic process, it probably makes sense for the bean to use the ResourceLoader interface to load resources. For example, consider the loading of a template of some sort, where the specific resource that is needed depends on the role of the user. If the resources are static, it makes sense to eliminate the use of the ResourceLoader interface completely, have the bean expose the Resource properties it needs, and expect them to be injected into it.

What makes it trivial to then inject these properties is that all application contexts register and use a special JavaBeans PropertyEditor, which can convert String paths to Resource objects. So, if myBean has a template property of type Resource, it can be configured with a simple string for that resource, as the following example shows:

<bean id="myBean" class="...">
	<property name="template" value="some/resource/path/myTemplate.txt"/>
</bean>

Note that the resource path has no prefix. Consequently, because the application context itself is going to be used as the ResourceLoader, the resource itself is loaded through a ClassPathResource, a FileSystemResource, or a ServletContextResource, depending on the exact type of the context.

If you need to force a specific Resource type to be used, you can use a prefix. The following two examples show how to force a ClassPathResource and a UrlResource (the latter being used to access a filesystem file):

<property name="template" value="classpath:some/resource/path/myTemplate.txt">
<property name="template" value="file:///some/resource/path/myTemplate.txt"/>

Application Contexts and Resource Paths

This section covers how to create application contexts with resources, including shortcuts that work with XML, how to use wildcards, and other details.

Constructing Application Contexts

An application context constructor (for a specific application context type) generally takes a string or array of strings as the location paths of the resources, such as XML files that make up the definition of the context.

When such a location path does not have a prefix, the specific Resource type built from that path and used to load the bean definitions depends on and is appropriate to the specific application context. For example, consider the following example, which creates a ClassPathXmlApplicationContext:

Java
ApplicationContext ctx = new ClassPathXmlApplicationContext("conf/appContext.xml");
Kotlin
val ctx = ClassPathXmlApplicationContext("conf/appContext.xml")

The bean definitions are loaded from the classpath, because a ClassPathResource is used. However, consider the following example, which creates a FileSystemXmlApplicationContext:

Java
ApplicationContext ctx =
	new FileSystemXmlApplicationContext("conf/appContext.xml");
Kotlin
val ctx = FileSystemXmlApplicationContext("conf/appContext.xml")

Now the bean definition is loaded from a filesystem location (in this case, relative to the current working directory).

Note that the use of the special classpath prefix or a standard URL prefix on the location path overrides the default type of Resource created to load the definition. Consider the following example:

Java
ApplicationContext ctx =
	new FileSystemXmlApplicationContext("classpath:conf/appContext.xml");
Kotlin
val ctx = FileSystemXmlApplicationContext("classpath:conf/appContext.xml")

Using FileSystemXmlApplicationContext loads the bean definitions from the classpath. However, it is still a FileSystemXmlApplicationContext. If it is subsequently used as a ResourceLoader, any unprefixed paths are still treated as filesystem paths.

Constructing ClassPathXmlApplicationContext Instances — Shortcuts

The ClassPathXmlApplicationContext exposes a number of constructors to enable convenient instantiation. The basic idea is that you can supply merely a string array that contains only the filenames of the XML files themselves (without the leading path information) and also supplies a Class. The ClassPathXmlApplicationContext then derives the path information from the supplied class.

Consider the following directory layout:

com/
  foo/
    services.xml
    daos.xml
    MessengerService.class

The following example shows how a ClassPathXmlApplicationContext instance composed of the beans defined in files named services.xml and daos.xml (which are on the classpath) can be instantiated:

Java
ApplicationContext ctx = new ClassPathXmlApplicationContext(
	new String[] {"services.xml", "daos.xml"}, MessengerService.class);
Kotlin
val ctx = ClassPathXmlApplicationContext(arrayOf("services.xml", "daos.xml"), MessengerService::class.java)

See the {api-spring-framework}/jca/context/SpringContextResourceAdapter.html[ClassPathXmlApplicationContext] javadoc for details on the various constructors.

Wildcards in Application Context Constructor Resource Paths

The resource paths in application context constructor values may be simple paths (as shown earlier), each of which has a one-to-one mapping to a target Resource or, alternately, may contain the special "classpath*:" prefix or internal Ant-style regular expressions (matched by using Spring’s PathMatcher utility). Both of the latter are effectively wildcards.

One use for this mechanism is when you need to do component-style application assembly. All components can 'publish' context definition fragments to a well-known location path, and, when the final application context is created using the same path prefixed with classpath*:, all component fragments are automatically picked up.

Note that this wildcarding is specific to the use of resource paths in application context constructors (or when you use the PathMatcher utility class hierarchy directly) and is resolved at construction time. It has nothing to do with the Resource type itself. You cannot use the classpath*: prefix to construct an actual Resource, as a resource points to just one resource at a time.

Ant-style Patterns

Path locations can contain Ant-style patterns, as the following example shows:

/WEB-INF/*-context.xml
com/mycompany/**/applicationContext.xml
file:C:/some/path/*-context.xml
classpath:com/mycompany/**/applicationContext.xml

When the path location contains an Ant-style pattern, the resolver follows a more complex procedure to try to resolve the wildcard. It produces a Resource for the path up to the last non-wildcard segment and obtains a URL from it. If this URL is not a jar: URL or container-specific variant (such as zip: in WebLogic, wsjar in WebSphere, and so on), a java.io.File is obtained from it and used to resolve the wildcard by traversing the filesystem. In the case of a jar URL, the resolver either gets a java.net.JarURLConnection from it or manually parses the jar URL and then traverses the contents of the jar file to resolve the wildcards.

Implications on Portability

If the specified path is already a file URL (either implicitly because the base ResourceLoader is a filesystem one or explicitly), wildcarding is guaranteed to work in a completely portable fashion.

If the specified path is a classpath location, the resolver must obtain the last non-wildcard path segment URL by making a Classloader.getResource() call. Since this is just a node of the path (not the file at the end), it is actually undefined (in the ClassLoader javadoc) exactly what sort of a URL is returned in this case. In practice, it is always a java.io.File representing the directory (where the classpath resource resolves to a filesystem location) or a jar URL of some sort (where the classpath resource resolves to a jar location). Still, there is a portability concern on this operation.

If a jar URL is obtained for the last non-wildcard segment, the resolver must be able to get a java.net.JarURLConnection from it or manually parse the jar URL, to be able to walk the contents of the jar and resolve the wildcard. This does work in most environments but fails in others, and we strongly recommend that the wildcard resolution of resources coming from jars be thoroughly tested in your specific environment before you rely on it.

The classpath*: Prefix

When constructing an XML-based application context, a location string may use the special classpath*: prefix, as the following example shows:

Java
ApplicationContext ctx =
	new ClassPathXmlApplicationContext("classpath*:conf/appContext.xml");
Kotlin
val ctx = ClassPathXmlApplicationContext("classpath*:conf/appContext.xml")

This special prefix specifies that all classpath resources that match the given name must be obtained (internally, this essentially happens through a call to ClassLoader.getResources(…​)) and then merged to form the final application context definition.

Note
The wildcard classpath relies on the getResources() method of the underlying classloader. As most application servers nowadays supply their own classloader implementation, the behavior might differ, especially when dealing with jar files. A simple test to check if classpath* works is to use the classloader to load a file from within a jar on the classpath: getClass().getClassLoader().getResources("<someFileInsideTheJar>"). Try this test with files that have the same name but are placed inside two different locations. In case an inappropriate result is returned, check the application server documentation for settings that might affect the classloader behavior.

You can also combine the classpath*: prefix with a PathMatcher pattern in the rest of the location path (for example, classpath*:META-INF/*-beans.xml). In this case, the resolution strategy is fairly simple: A ClassLoader.getResources() call is used on the last non-wildcard path segment to get all the matching resources in the class loader hierarchy and then, off each resource, the same PathMatcher resolution strategy described earlier is used for the wildcard subpath.

Other Notes Relating to Wildcards

Note that classpath*:, when combined with Ant-style patterns, only works reliably with at least one root directory before the pattern starts, unless the actual target files reside in the file system. This means that a pattern such as classpath*:*.xml might not retrieve files from the root of jar files but rather only from the root of expanded directories.

Spring’s ability to retrieve classpath entries originates from the JDK’s ClassLoader.getResources() method, which only returns file system locations for an empty string (indicating potential roots to search). Spring evaluates URLClassLoader runtime configuration and the java.class.path manifest in jar files as well, but this is not guaranteed to lead to portable behavior.

Note

The scanning of classpath packages requires the presence of corresponding directory entries in the classpath. When you build JARs with Ant, do not activate the files-only switch of the JAR task. Also, classpath directories may not get exposed based on security policies in some environments — for example, stand-alone applications on JDK 1.7.0_45 and higher (which requires 'Trusted-Library' to be set up in your manifests. See https://stackoverflow.com/questions/19394570/java-jre-7u45-breaks-classloader-getresources).

On JDK 9’s module path (Jigsaw), Spring’s classpath scanning generally works as expected. Putting resources into a dedicated directory is highly recommendable here as well, avoiding the aforementioned portability problems with searching the jar file root level.

Ant-style patterns with classpath: resources are not guaranteed to find matching resources if the root package to search is available in multiple class path locations. Consider the following example of a resource location:

com/mycompany/package1/service-context.xml

Now consider an Ant-style path that someone might use to try to find that file:

classpath:com/mycompany/**/service-context.xml

Such a resource may be in only one location, but when a path such as the preceding example is used to try to resolve it, the resolver works off the (first) URL returned by getResource("com/mycompany");. If this base package node exists in multiple classloader locations, the actual end resource may not be there. Therefore, in such a case you should prefer using classpath*: with the same Ant-style pattern, which searches all class path locations that contain the root package.

FileSystemResource Caveats

A FileSystemResource that is not attached to a FileSystemApplicationContext (that is, when a FileSystemApplicationContext is not the actual ResourceLoader) treats absolute and relative paths as you would expect. Relative paths are relative to the current working directory, while absolute paths are relative to the root of the filesystem.

For backwards compatibility (historical) reasons however, this changes when the FileSystemApplicationContext is the ResourceLoader. The FileSystemApplicationContext forces all attached FileSystemResource instances to treat all location paths as relative, whether they start with a leading slash or not. In practice, this means the following examples are equivalent:

Java
ApplicationContext ctx =
	new FileSystemXmlApplicationContext("conf/context.xml");
Kotlin
val ctx = FileSystemXmlApplicationContext("conf/context.xml")
Java
ApplicationContext ctx =
	new FileSystemXmlApplicationContext("/conf/context.xml");
Kotlin
val ctx = FileSystemXmlApplicationContext("/conf/context.xml")

The following examples are also equivalent (even though it would make sense for them to be different, as one case is relative and the other absolute):

Java
FileSystemXmlApplicationContext ctx = ...;
ctx.getResource("some/resource/path/myTemplate.txt");
Kotlin
val ctx: FileSystemXmlApplicationContext = ...
ctx.getResource("some/resource/path/myTemplate.txt")
Java
FileSystemXmlApplicationContext ctx = ...;
ctx.getResource("/some/resource/path/myTemplate.txt");
Kotlin
val ctx: FileSystemXmlApplicationContext = ...
ctx.getResource("/some/resource/path/myTemplate.txt")

In practice, if you need true absolute filesystem paths, you should avoid using absolute paths with FileSystemResource or FileSystemXmlApplicationContext and force the use of a UrlResource by using the file: URL prefix. The following examples show how to do so:

Java
// actual context type doesn't matter, the Resource will always be UrlResource
ctx.getResource("file:///some/resource/path/myTemplate.txt");
Kotlin
// actual context type doesn't matter, the Resource will always be UrlResource
ctx.getResource("file:///some/resource/path/myTemplate.txt")
Java
// force this FileSystemXmlApplicationContext to load its definition via a UrlResource
ApplicationContext ctx =
	new FileSystemXmlApplicationContext("file:///conf/context.xml");
Kotlin
// force this FileSystemXmlApplicationContext to load its definition via a UrlResource
val ctx = FileSystemXmlApplicationContext("file:///conf/context.xml")