I am rewriting a legacy project using Quartz and Spring Framework. The original configuration is in XML and now I am translating it into Java Config. The xml configuration uses jobDetail to set the job detail property of the trigger bean. However, when I try to use the equivalent method, i.e. the setter: setJobDetails(simpleJobDetail), I got a warning that the setter does not have correct type (expecting JobDetails, but got MethodInvokingJobDetailFactoryBean).
May I know whether it is correct to translate xml bean configuration to Java COnfig by using the equivalent named setter in Java COnfig?
Why in the XML property setting, the trigger bean can set its jobDetail property as the simpleJobDetail bean (which has type MethodInvokingJobDetailFactoryBean) , while the Java Config could not?
XML config:
<!-- For times when you just need to invoke a method on a specific object -->
<bean id="simpleJobDetail"
class="org.springframework.scheduling.quartz.MethodInvokingJobDetailFactoryBean">
<property name="targetObject" ref="reader" />
<property name="targetMethod" value="readData" />
</bean>
<!-- Run the job every 60 seconds with initial delay of 1 second -->
<bean id="trigger"
class="org.springframework.scheduling.quartz.SimpleTriggerFactoryBean">
<property name="jobDetail" ref="simpleJobDetail" />
<property name="repeatInterval" value="600000" />
</bean>
Java Config:
#Bean
public MethodInvokingJobDetailFactoryBean simpleJobDetail() {
MethodInvokingJobDetailFactoryBean simpleJobDetail = new MethodInvokingJobDetailFactoryBean();
simpleJobDetail.setTargetObject(reader());
simpleJobDetail.setTargetMethod("readData");
return simpleJobDetail;
}
#Bean
private Object reader() {
// TODO: 10/13/2016
return null;
}
#Bean
public SimpleTriggerFactoryBean trigger() {
final SimpleTriggerFactoryBean trigger = new SimpleTriggerFactoryBean();
trigger.setJobDetail( simpleJobDetail()); // got warning about wrong type
trigger.setRepeatInterval(60000);
return trigger;
}
Note that simpleJobDetail() returns a factory, not the bean itself. You can rely on autowiring to inject a JobDetail built using this factory.
#Bean
public SimpleTriggerFactoryBean trigger(final JobDetail simpleJobDetail) {
final SimpleTriggerFactoryBean trigger = new SimpleTriggerFactoryBean();
trigger.setJobDetail(simpleJobDetail); // got warning about wrong type
trigger.setRepeatInterval(60000);
return trigger;
}
Hope it helps.
Related
I have a spring based web application and in my application context xml file, I have defined a bean which has all the parameters to connect to database. As part of this bean, for one of the parameters, I have a password key, as shown in the below example and I wanted the value should come from a /vault/password file. This /vault/password is not part of the project/application. This /vault/password will be there in host machine by default.
What is the syntax in applicationContext.xml bean definition, to read a value from a file outside of application context.
<bean class="org.apache.commons.dbcp.BasicDataSource"
destroy-method="close" id="dataSource">
<property name="url" value="jdbc:postgresql://postgres:5432/" />
<property name="username" value="postgres" />
<property name="password" value="/vault/password" />
</bean>
Something like this is probably your best bet:
How to correctly override BasicDataSource for Spring and Hibernate
PROBLEM:
Now I need to provide custom data source based on server environment
(not config), for which I need to calculate driverClassName and url
fields based on some condition.
SOLUTION:
Create a factory (since you need to customize only the creation phase
of the object, you don't need to control the whole lifetime of it).
public class MyDataSourceFactory {
public DataSource createDataSource() {
BasicDataSource target = new BasicDataSource();
if (condition) {
target.setDriverClassName("com.mysql.jdbc.Driver");
target.setUrl("jdbc:mysql://localhost/test?relaxAutoCommit=true");
} else { ... }
return target;
}
}
In your case, your customization would do some I/O to set target.password.
I am working 2 different processes.
I am passing process name in JVM argument. Using that argument either of the processes should call.
my app context XML.
<bean id="propertyPlaceHolderConfigurer" class="org.springframework.beans.factory.config.PropertyPlaceholderConfigurer">
<property name="locations">
<list>
<value>classpath*:properties/${processJVMArg}.properties</value>
</list>
</property>
</bean>
<bean id="splitService" class="com.split.service.SplitService" init-method="process1"><!-- "based on processJVMArg JVM argument should call process1 or process2. " -->
Is there is a way to configure more than one init method, init method should call based on conduction?
Thanks,
Rama
Spring has profiles. Based on set of conditions you can start app with exact profile. Bean beanCreatingBasedOnProfile will be created only in dev profile.
#Bean
#Profile("dev")
public YourClass beanCreatingBasedOnProfile() {
return new YourClass();
}
Also spring has Conditional Beans .
You can construct bean based on property value or anything else. For example:
#ConditionalOnProperty(prefix = "spring.prop", name = "dynamic", matchIfMissing = true)
public YourClass condBean() {
return new YourClass();
}
I have a follofing situation int "super-context.xml":
<bean id="conf" class="ee.Conf"/>
<bean id="service" class="ee.Serivce">
<property name="conf" ref="conf">
</bean>
Now I want to use this "super-context.xml" in various different projects. Say "sub-context.xml" has:
<import resource="super-context.xml"/>
<bean id="subConf1" class="ee.SubConf">
<property name="confloc" value="classpath:ee/customconf1.sss" />
</bean>
<bean id="subConf2" class="ee.SubConf">
<property name="confloc" value="classpath:ee/customconf2.sss" />
</bean>
...
<bean id="subConfn" class="ee.SubConf">
<property name="confloc" value="classpath:ee/customconfn.sss" />
</bean>
ee.Conf is something as follows:
public class Conf ... {
...
public void addSubConf(Resource res) {
//configuration resolving from res
}
...
}
ee.SubConf is something as follows:
public class SubConf ... {
...
#Autowired
ee.Conf superConf;
...
public void setConfloc(Resource res) {
superConf.addSubConf(res);
}
...
}
The problem aries on context load. Beans are initialized in following order (due to ordering in context file): conf, service, subConf1, subConf2, ... subConfn
But service bean actually depends on all the subConf beans (although this can't be deducted from the context definition itself). It loads OK when import in "sub-context.xml" is added after subConf bean definitions.
Reason behind this is implementing modularity. Is it possible to force a bean to load as late as possible ("service" bean in the example) or make beans of certain type load as soon as possible ("subConf" beans in the example), since fixed ordering of beans in "sub-context.xml" partly kills the wished modularity
Or is theree a more pure way to achieve this type of modularity?
I would say that you are approaching the problem in a wrong way. The SubConf shouldn't have a dependency on the Conf to start with. Simply inject the collection of SubConf objects in your Conf object.
public class Conf {
#Autowired
private List<SubConf> subconfs;
}
That way you eliminate the need for the SubConf to call the Conf class and this will remove your circular dependency.
See the Spring reference guide for more information on autowiring.
You can use depends-on
<bean id="beanOne" class="foo.Bar" depends-on="beanTwo" />
How can i inject a properties file containing a Map to be used as additional constructor arg using the field.
With a Map being loaded from a properties file
the bean is currently setup using:
<bean id="graphDbService" class="org.neo4j.kernel.EmbeddedGraphDatabase"
init-method="enableRemoteShell" destroy-method="shutdown">
<constructor-arg index="0" value= "data/neo4j-db"/>
<constructor-arg index="1" value=? />
</bean>
Java Equivalent:
Map<String,String> configuration = EmbeddedGraphDatabase.loadConfigurations( "neo4j_config.props" );
GraphDatabaseService graphDb = new EmbeddedGraphDatabase( "data/neo4j-db", configuration );
Thanks
Something like this:
<bean id="configuration" class="org.neo4j.kernel.EmbeddedGraphDatabase"
factory-method="loadConfigurations">
<constructor-arg value="neo4j_config.props"/>
</bean>
<bean id="graphDbService" class="org.neo4j.kernel.EmbeddedGraphDatabase"
init-method="enableRemoteShell" destroy-method="shutdown">
<constructor-arg index="0" value="data/neo4j-db"/>
<constructor-arg index="1" ref="configuration" />
</bean>
This takes advantage of the ability to create beans using arbitrary static factory methods, in this case using loadConfigurations() as a factory method to create the configuration bean, which is then injected into the proper constructor of EmbeddedGraphDatabase.
Create a bean that loads the properties (and takes the file name as an argument) and inject that instead.
EDIT When using annotations, things like constructor injection become more simple:
#Bean
public Map<String,String> configuration() {
return EmbeddedGraphDatabase.loadConfigurations( "neo4j_config.props" );
}
#Bean
public GraphDatabaseService graphDb() {
return new EmbeddedGraphDatabase( "data/neo4j-db", configuration() );
}
Note that the second bean definition method "simply" calls the first. When this code is executed, Spring will do some magic so you can still override the bean elsewhere (i.e. beans still overwrite each other) and it will make sure that the method body will be executed only once (no matter how often and from where it was called).
If the config is in a different #Configuration class, then you can #Autowired it:
#Autowired
private Map<String,String> configuration;
#Bean
public GraphDatabaseService graphDb() {
return new EmbeddedGraphDatabase( "data/neo4j-db", configuration );
}
I have a spring context xml file with this
<context:property-placeholder location="classpath:cacheConfig.properties"/>
<bean id="cacheManager"
class="org.springframework.cache.ehcache.EhCacheManagerFactoryBean">
<property name="cacheManagerName" value="cacheName"/>
<property name="shared" value="false"/>
<property name="configLocation" value="classpath:cacheConfig.xml"/>
</bean>
the goal is to allow the customer to edit the properties file, like this
cache.maxMemoryElements="2000"
and then in the actual cacheConfig.xml file have this
<cache name="someCacheName"
maxElementsInMemory="${cache.maxMemoryElements}" ... />
so that items we do not want the customer to change are not exposed. Of course the above details are only partially detailed and NOT working. Currently I see this in the log file
Invocation of init method failed; nested exception is net.sf.ehcache.CacheException: Error configuring from input stream. Initial cause was null:149: Could not set attribute "maxElementsInMemory".
Thanks in advance...
Your example uses EhCacheManagerFactoryBean to expose a reference to the CacheManager, with caches defined in the external cacheConfig.xml file. As #ChssPly76 pointed out, Spring's property resolver only works within Spring's own bean definition files.
However, you don't have to define the individual caches in the external file, you can define them right within the Spring bean definition file, using EhCacheFactoryBean:
FactoryBean that creates a named
EHCache Cache instance... If the
specified named cache is not
configured in the cache configuration
descriptor, this FactoryBean will
construct an instance of a Cache with
the provided name and the specified
cache properties and add it to the
CacheManager for later retrieval.
In other words, if you use EhCacheFactoryBean to refer to a named cache that isn't already defined in cacheConfig.xml, then Spring will create and configure a new cache instance and register it with the CacheManager at runtime. That includes specifying things like maxElementsInMemory, and because this would be specified in the Spring bean definition file, you get full support of the property resolver:
<context:property-placeholder location="classpath:cacheConfig.properties"/>
<bean id="myCache" class="org.springframework.cache.ehcache.EhCacheFactoryBean">
<property name="cacheManager" ref="cacheManager"/>
<property name="maxElementsInMemory" value="${cache.maxMemoryElements}"/>
</bean>
<bean id="cacheManager" class="org.springframework.cache.ehcache.EhCacheManagerFactoryBean">
<property name="shared" value="false"/>
<property name="configLocation" value="classpath:cacheConfig.xml"/>
</bean>
This is not how PropertyPlaceholderConfigurer works. It can be used to replace values within context, but not within arbitrary external files. And cacheConfig.xml is an external file - it's just being passed by Spring to EH Cache.
If you are using Maven or Ant, both offer the ability to do filtering of tokens in resource files.
For Maven, you could do something like
<cache name="someCacheName"
maxElementsInMemory="${cache.maxMemoryElements}" ... />
And in a filter file, or in the POM itself, have
cache.maxMemoryElements = 200
Resource Filtering in Maven: The Definitive Guide
With Ant, you do this with FilterSets and the <copy> task.
For anyone who needs to modify the diskstore path which cannot be set as the ehcache javadoc states that the diskstore parameter is ignored, you could create your own implementation of a EhCacheManagerFactoryBean, which allows you to inject the diskstore path; you basically need to intercept the creation of the CacheManager and modify the configuration passed along using your diskstore property, e.g.:
private String diskStorePath;
...getter/setter
public void afterPropertiesSet() throws IOException, CacheException {
if (this.shared) {
// Shared CacheManager singleton at the VM level.
if (this.configLocation != null) {
this.cacheManager = CacheManager.create(this.createConfig());
}
else {
this.cacheManager = CacheManager.create();
}
}
else {
// Independent CacheManager instance (the default).
if (this.configLocation != null) {
this.cacheManager = new CacheManager(this.createConfig());
}
else {
this.cacheManager = new CacheManager();
}
}
if (this.cacheManagerName != null) {
this.cacheManager.setName(this.cacheManagerName);
}
}
private Configuration createConfig() throws CacheException, IOException {
Configuration config = ConfigurationFactory.parseConfiguration(this.configLocation.getInputStream());
DiskStoreConfiguration diskStoreConfiguration = config.getDiskStoreConfiguration();
if (diskStoreConfiguration == null) {
DiskStoreConfiguration diskStoreConfigurationParameter = new DiskStoreConfiguration();
diskStoreConfigurationParameter.setPath(getDiskStorePath());
config.addDiskStore(diskStoreConfigurationParameter);
} else {
diskStoreConfiguration.setPath(getDiskStorePath());
}
return config;
}
The Spring config then looks like this:
<bean id="cacheManager" class="com.yourcompany.package.MyEhCacheManagerFactoryBean" depends-on="placeholderConfig">
<property name="diskStorePath" value="${diskstore.path}"/>
<property name="configLocation" value="classpath:ehcache.xml" />
</bean>