I have a row mapper class which implements RowMapper interface. I need to implement the mapRow method in that. The arguments of it are ResulSet and index. I would like to use this ResultSet object in another bean. How do i get it there ?
Set this object as instance variable. But i would never recommend that for ResultSet. Resultset will be useless once its closed by spring (as spring is managing it).
Better extract the data out of ResultSet store the data as some model bean in the instance variable (but keep in mind, by default the beans are singleton, and for each execution storing the data as instance variable would not make much sense either).
EDIT-- okay to refine a bit more, i am putting an example here
// this is singleton by default, if you want to store data in this bean
// mark it as bean either by annotation or thorugh xml
// so that it can be accessed through spring context
public class MyBeanRowMapper implements RowMapper
{
// you can store something here,
// but with each execution of mapRow, it will be updated
// not recommended to store execution result here
// or inject other bean here // SomeOtherBean someOtherbean;
public Object mapRow(ResultSet rs, int rowNum) throws SQLException {
MyBean myBean = new MyBean();
myBean.setId(rs.getInt("BEAN_ID"));
myBean.setName(rs.getString("NAME"));
// set this bean into gloablly accessible object here
// or to the bean in which you want to access the result
// something like -->// someOtherBean.setMyBean(myBean)
// again be careful if someOtherBean is singleton by default
return myBean;
}
}
Related
I am using spring framework.
I am using objectMapper to desiriale store.json file:
service:
objectMapper.readValue(new File(jsonFilePath), Store.class)
store.json:
{
"type": "Store",
"name": "myStore",
}
Store.class:
#Value("${store.size:1000}")
private Integer sroreSize;
#autowire
private storePersistency storePersistency;
public Store(#JsonProperty("name") String name) {
super(name);
}
I am trying find out how to #autowire beans and #value properties in store.class, beans and properties that exist in applicationContext.
In current example sroreSize and storePersistency still null.
I know that I can inject fields to object mapper and then use #JacksonInject annotation but I have a lot of field to inject - not a good option for me.
Custom desirializer also not a good option for me.
Is there any way not to use custom desirializer or not to inject every bean/property that I need in store.class?
Something that injects all the beans and properties and I simply can use it in Store.class.
So you want some Store fields like storePersistency and sroreSize to be initialized once at application startup (which is when Spring will setup the application context) and then at runtime create multiple different Store objects differing in some fields as name that are initialized by Jackson.
I suggest annotating Store with #Component to get Spring to initialize #Value and #Autowired fields. The #Scope annotation will cause a new independent Store instance to be created each time. Simplified example:
#Component
#Scope(SCOPE_PROTOTYPE)
class Store {
private String name;
#Value("${store.size:1000}")
private Integer sroreSize;
}
Then the key is method readerForUpdating where you can pass an existing instance of Store and Jackson will update that instead of creating a new one as usually:
Store store = context.getBean(Store.class);
objectMapper.readerForUpdating(store).readValue("{\"name\":\"myStore\"}");
Where context is a Spring ApplicationContext reference that I autowired in a test class. You don't need to use the return value of readValue in this case, just inspect the existing store variable and name will be updated.
Java Spring question:
I have a interface MyInterface with one method
void exec (String str);
I have many implementation of MyInterface, say Oneimpl, anotherimpl yetanotherimpl...and so on and can keep adding new implementations.
how do I obtain an instance of a specific implementation using just the name of the implementing class passed as a STRING value , say "someRandomImpl"
The code should be dynamic and can provide a instance of new implementations without code change.
implements ApplicationContextAware
it will autowired ApplicationContext object
use the object like
context.getBean(beanName)
then you get the bean
Setup :
- have a Several Configuration Class containing Bean definitions for my beans
- I will fetched a List of String from a database containing all the bean names and their corresponding Configuration class I want to instantiate dynamically
currently I will do a loop on the list and then call a method passing the beanName and the Configuration Class containing the bean definition:
private Object getBean(String beanName, Class configurationClass) {
Object bean = null;
AbstractApplicationContext context = new AnnotationConfigApplicationContext(
configurationClass);
bean = context.getBean(beanName);
return bean;
}
I would then use the returned object and used Reflections to invoke specific Methods based on a list I fetched from a database.
Question : Is there a proper way to do this ? because for every bean I want to create , I think performance will be affected.
You can use this in spring 4.1
I found the below example in this post - Spring MVC: How to return image in #ResponseBody?
public ResponseEntity<InputStreamResource> downloadUserAvatarImage(#PathVariable Long userId) {
GridFSDBFile gridFsFile = fileService.findUserAccountAvatarById(userId);
return ResponseEntity.ok()
.contentLength(gridFsFile.getLength())
.contentType(MediaType.parseMediaType(gridFsFile.getContentType()))
.body(new InputStreamResource(gridFsFile.getInputStream()));
}
Spring provides bean scope as "Prototype". Means whenever bean is required in application, Spring container will create a fresh/new instance of bean.
Does is follow prototype design pattern also?
Does it create object only once and in subsequent request calls clone() method on created object to create new object?
Also if someone can provide example of prototype in JDK, Spring, Hibernate or any J2EE framework.
No spring does not use cloning to create prototype scoped instances.
Below is the code snippet taken from AbstractBeanFactory.doGetBean() function:
// Create bean instance.
if (mbd.isSingleton()) {
sharedInstance = getSingleton(beanName, new ObjectFactory<Object>() {
#Override
public Object getObject() throws BeansException {
try {
return createBean(beanName, mbd, args);
}
catch (BeansException ex) {
// Explicitly remove instance from singleton cache: It might have been put there
// eagerly by the creation process, to allow for circular reference resolution.
// Also remove any beans that received a temporary reference to the bean.
destroySingleton(beanName);
throw ex;
}
}
});
bean = getObjectForBeanInstance(sharedInstance, name, beanName, mbd);
}
else if (mbd.isPrototype()) {
// It's a prototype -> create a new instance.
Object prototypeInstance = null;
try {
beforePrototypeCreation(beanName);
prototypeInstance = createBean(beanName, mbd, args);
}
finally {
afterPrototypeCreation(beanName);
}
bean = getObjectForBeanInstance(prototypeInstance, name, beanName, mbd);
}
The createBean method call boils down to below code:
BeanUtils.instantiateClass(constructorToUse);
Spring does not use the Prototype Pattern, it uses reflection.
Plus, in order to use clone() it would have to subclass somehow a bean, because clone() is protected, so it does not use clone() either.
Here is a code fragment from
org.springframework.beans.factory.support.SimpleInstantiationStrategy
where you can see the use of java.lang.reflect.Constructor and java.lang.Class reflection method:
public Object instantiate(RootBeanDefinition beanDefinition, String beanName, BeanFactory owner) {
if (beanDefinition.getMethodOverrides().isEmpty()) {
Constructor<?> constructorToUse;
synchronized (beanDefinition.constructorArgumentLock) {
constructorToUse = (Constructor<?>) beanDefinition.resolvedConstructorOrFactoryMethod;
...
constructorToUse = clazz.getDeclaredConstructor((Class[]) null);
...
}
...
}
...
}
So the term prototype is used to suggest that at every call to getBean, you'll get a new instance with the same properties. This is more than a simple call to a constructor however, because you'll get a bean with all dependencies wired and other properties set, so in a sense it is a prototype. Or at least it fits the concept very well.
I have not dug into Spring source code, but I think Beans with prototype scope in Spring are not created using clone() method because it is not mandatory to implement the Cloneable interface for those beans.
Moreover, suppose it is creating them using clone(). It would then be dangerous if someone is expecting deep copy instead of shallow copy.
You can always test it and find the answer.
No. Spring scopes such as prototype or singletone do not follow strictly design patterns. The naming of scopes was used to intuitively suggest behavior container provides.
This way you can have a "singleton" pattern within the container and create another object outside of the container. Similarly "prototype" pattern does not have to implement "clone" functionality.
You may want to look into this link as well: Singleton design pattern vs Singleton beans in Spring container
More elaborate explanations here:
https://springframework.guru/gang-of-four-design-patterns/prototype-pattern/
If I have:
#Autowired private ApplicationContext ctx;
I can get beans and resources by using one of the the getBean methods. However, I can't figure out how to get property values.
Obviously, I can create a new bean which has an #Value property like:
private #Value("${someProp}") String somePropValue;
What method do I call on the ApplicationContext object to get that value without autowiring a bean?
I usually use the #Value, but there is a situation where the SPeL expression needs to be dynamic, so I can't just use an annotation.
In the case where SPeL expression needs to be dynamic, get the property value manually:
somePropValue = ctx.getEnvironment().getProperty("someProp");
If you are stuck on Spring pre 3.1, you can use
somePropValue = ctx.getBeanFactory().resolveEmbeddedValue("${someProp}");
Assuming that the ${someProp} property comes from a PropertyPlaceHolderConfigurer, that makes things difficult. The PropertyPlaceholderConfigurer is a BeanFactoryPostProcessor and as such only available at container startup time. So the properties are not available to a bean at runtime.
A solution would be to create some sort of a value holder bean that you initialize with the property / properties you need.
#Component
public class PropertyHolder{
#Value("${props.foo}") private String foo;
#Value("${props.bar}") private String bar;
// + getter methods
}
Now inject this PropertyHolder wherever you need the properties and access the properties through the getter methods