Imports another build file into the current project.
On execution it will select the proper ProjectHelper to parse the imported file, using the same algorithm as the one executed at startup. The selected ProjectHelper instance will then be responsible to actually parse the imported file.
Note as seen above, this task heavily relies on the ProjectHelper implementation and doesn't really perform any work of its own. If you have configured Apache Ant to use a ProjectHelper other than Ant's default, this task may or may not work.
In the common use case where only Ant's default project helper is used, it basically works like
the Entity Includes
as explained in the Ant FAQ, as if the imported file was contained in the importing file,
minus the top <project> tag.
The import task may only be used as a top-level task. This means that it may not
be used in a target.
There are two further functional aspects that pertain to this task and that are not possible with entity includes:
If a target in the main file is also present in at least one of the imported files, the one from the main file takes precedence.
So if I import for example a docsbuild.xml file containing a project
named builddocs
that contains a docs
target, I can define a docs
target in
my main buildfile and that is the one that will be called. This makes it easy to keep the same
target name, so that the overriding target is still called by any other targets—in either
the main or imported buildfile(s)—for which it is a dependency, with a different
implementation. The target from docsbuild.xml is made available by the
name builddocs.docs
. This enables the new implementation to call the old target,
thus enhancing it with tasks called before or after it.
If you use the as attribute of the task, its value will be used to prefix the
overridden target's name instead of the name attribute of the project
tag.
Imported files are treated as they are present in the main buildfile. This makes it easy to understand, but it makes it impossible for them to reference files and resources relative to their path. Because of this, for every imported file, Ant adds a property that contains the path to the imported buildfile. With this path, the imported buildfile can keep resources and be able to reference them relative to its position.
So if I import for example a docsbuild.xml file named builddocs
, I can get
its path as ant.file.builddocs, similarly to the ant.file property of
the main buildfile.
Note that builddocs
is not the filename, but the name attribute present in
the imported project tag.
If the imported file does not have a name attribute,
the ant.file.projectname property will not be set.
Since Ant 1.8.0, the task can also import resources from URLs or classpath resources
(which are URLs, really). If you need to know whether the current build file's source has been
a file or an URL you can consult the property ant.file.type.projectname
(using the same example as above ant.file.type.builddocs) which either have the
value file
or url
.
Suppose your main build file called importing.xml imports a build file imported.xml, located anywhere on the file system, and imported.xml reads a set of properties from imported.properties:
<!-- importing.xml -->
<project name="importing" basedir="." default="...">
<import file="${path_to_imported}/imported.xml"/>
</project>
<!-- imported.xml -->
<project name="imported" basedir="." default="...">
<property file="imported.properties"/>
</project>
This snippet however will resolve imported.properties against the basedir of importing.xml, because the basedir of imported.xml is ignored by Ant. The right way to use imported.properties is:
<!-- imported.xml -->
<project name="imported" basedir="." default="...">
<dirname property="imported.basedir" file="${ant.file.imported}"/>
<property file="${imported.basedir}/imported.properties"/>
</project>
As explained above ant.file.imported stores the path of the build script, that
defines the project called imported
, (in short it stores the path
to imported.xml) and <dirname> takes
its directory. This technique also allows imported.xml to be used as a standalone
file (without being imported in other project).
The above description only works for imported files that actually are imported from files and not from URLs. For files imported from URLs using resources relative to the imported file requires you to use tasks that can work on non-file resources in the first place. To create a relative resource you'd use something like:
<loadproperties>
<url baseUrl="${ant.file.imported}"
relativePath="imported.properties"/>
</loadproperties>
| Attribute | Description | Required |
|---|---|---|
| file | The file to import. If this is a relative file name, the file name will be resolved relative to the importing file. Note: this is unlike most other Ant file attributes, where relative files are resolved relative to basedir. | Yes or a nested resource collection |
| optional | If true, do not stop the build if the file does not exist. |
No; default is false |
| as | Specifies the prefix prepended to the target names. | No; defaults to name attribute of the project tag of the imported
file |
| prefixSeparator | Specifies the separator to be used between the prefix and the target name. | No; defaults to . |
Since Ant 1.8.0
The specified resources will be imported.
<import file="../common-targets.xml"/>
Imports targets from the common-targets.xml file that is in a parent directory.
<import file="${deploy-platform}.xml"/>
Imports the project defined by the property deploy-platform
<import>
<javaresource name="common/targets.xml">
<classpath location="common.jar"/>
</javaresource>
</import>
Imports targets from the targets.xml file that is inside the directory common inside the jar file common.jar.
The short version: Use import if you intend to override a target, otherwise
use include.
When import is used, the imported targets are available by up to two names: their
"normal" name without any prefix and potentially with a prefixed name (the value of
the as attribute or the imported project's name attribute, if any).
When include is used, the included targets are only available in the prefixed form.
When import is used, the imported target's depends attribute remains
unchanged, i.e. it uses "normal" names and allows you to override targets in the dependency
list.
When include is used, the included targets cannot be overridden and
their depends attributes are rewritten so that prefixed names are used. This allows
writers of the included file to control which target is invoked as part of the dependencies.
It is possible to include the same file more than once by using different prefixes;
it is not possible to import the same file more than once.
nested.xml shall be:
<project>
<target name="setUp">
<property name="prop" value="in nested"/>
</target>
<target name="echo" depends="setUp">
<echo>prop has the value ${prop}</echo>
</target>
</project>
When using import like in
<project default="test">
<target name="setUp">
<property name="prop" value="in importing"/>
</target>
<import file="nested.xml" as="nested"/>
<target name="test" depends="nested.echo"/>
</project>
Running the build file will emit:
setUp:
nested.echo:
[echo] prop has the value in importing
test:
When using include like in
<project default="test">
<target name="setUp">
<property name="prop" value="in importing"/>
</target>
<include file="nested.xml" as="nested"/>
<target name="test" depends="nested.echo"/>
</project>
Running the target build file will emit:
nested.setUp:
nested.echo:
[echo] prop has the value in nested
test:
and there won't be any target named echo
on the including build file.