Seamew Technical Blog

website·Indexed 29 days ago

A collection of technical articles and tutorials covering Java development, Spring Framework (including Spring Boot and AOP), distributed systems like ZooKeeper and Kafka, algorithm design (Dynamic Programming, DFS, and Backtracking), and DevOps tools such as Docker and Linux (CentOS).

Tokens: 41K
Snippets: 170
Records: 226
Agent score: 47%

What's inside Seamew

Find the largest number smaller than n using a specific set of digits
Given a set of available digits (1-9) and a target number n, find the largest number that can be constructed using only those digits (with repetition allowed) such that the resulting number is strictly less than n.
Examples:
- Available digits: {1, 2, 9, 4}, n = 2533 $\rightarrow$ Result: 2499
- Available digits: {1, 2, 5, 4}, n = 2543 $\rightarrow$ Result: 2542
- Available digits: {1, 2, 5, 4}, n = 2541 $\rightarrow$ Result: 2525
- Available digits: {1, 2, 9, 4}, n = 2111 $\rightarrow$ Result: 1999
- Available digits: {5, 9}, n = 5555 $\rightarrow$ Result: 999
https://seamew.top/article/%E7%AE%97%E6%B3%95/%E8%B4%AA%E5%BF%83/%E5%B0%8F%E4%BA%8En%E7%9A%84%E6%9C%80%E5%A4%A7%E6%95%B0%E5%AD%97.html
Understand ZooKeeper ZAB Protocol and Operation Modes
ZooKeeper relies on the ZAB (ZooKeeper Atomic Broadcast) protocol to ensure data synchronization across nodes. It operates in two primary modes:
1. Recovery Mode (Leader Election): Triggered when the current leader node fails. The remaining nodes elect a new leader. Once elected, the cluster transitions to Broadcast Mode to synchronize data with the new leader.
2. Broadcast Mode: The standard operational mode used to synchronize data across all nodes in the cluster.
https://seamew.top/article/%E5%A4%A7%E6%95%B0%E6%8D%AE/zookeeper.html
Understand ThreadLocal for thread-local variable storage
ThreadLocal provides thread-local variables. It ensures that the accessed variable belongs to the current thread, with each thread maintaining its own independently initialized copy of the variable. This effectively provides thread isolation by binding variables to the specific thread, making it useful for passing data in multi-threaded environments without needing explicit synchronization.
https://www.seamew.top/article/JAVA/Threadlocal.html
Understand AOP proxy creation via DefaultAdvisorAutoProxyCreator
The DefaultAdvisorAutoProxyCreator class uses the postProcessBeforeInstantiation method to intercept bean creation. If a suitable TargetSource is found, it retrieves the necessary advices and advisors for the bean and uses a ProxyFactory to create a proxy object. This process effectively 'short-circuits' the standard bean instantiation process, returning the proxy object directly.
```
@Override
public Object postProcessBeforeInstantiation(Class<?> beanClass, String beanName) {
  TargetSource targetSource = getCustomTargetSource(beanClass, beanName);
  if (targetSource != null) {
    // ...
    Object[] specificInterceptors = getAdvicesAndAdvisorsForBean(beanClass, beanName, targetSource);
    // Create proxy via ProxyFactory
    Object proxy = createProxy(beanClass, beanName, specificInterceptors, targetSource);
    return proxy;
  }
  return null;
}
```
https://www.seamew.top/article/spring%E8%BF%9B%E9%98%B6%E5%AD%A6%E4%B9%A0/aop%E5%88%9B%E5%BB%BA%E4%BB%A3%E7%90%86%E5%9F%BA%E6%9C%AC%E6%B5%81%E7%A8%8B.html
Understand Kafka Transactional Guarantees
Kafka transactions provide atomicity across multiple operations, ensuring that a set of actions—specifically consuming messages, producing messages, and committing consumer offsets—either all succeed or all fail. Transactions rely on the Idempotent Producer; enabling transactions automatically enables idempotency.
https://www.seamew.top/article/%E5%A4%A7%E6%95%B0%E6%8D%AE/kafka%E4%BA%8B%E5%8A%A1.html
Understand Spring's Three-Level Cache for Circular Dependencies
Spring uses three internal Map collections to manage Bean states during the creation process to resolve circular dependencies (e.g., Bean A depends on Bean B, and Bean B depends on Bean A).
- First-Level Cache (singletonObjects): Stores fully initialized Beans that are ready for use.
- Second-Level Cache (earlySingletonObjects): Stores 'half-finished' Beans—objects that have been instantiated but have not yet had their properties set or initialization completed.
- Third-Level Cache (singletonFactories): Stores ObjectFactory (lambda expressions). These factories can produce an early reference to a Bean (either a raw half-finished Bean or a proxy Bean). This is critical for handling AOP proxies during circular dependencies.
```
// First-level: Complete Beans
private final Map<String, Object> singletonObjects = new ConcurrentHashMap(256);

// Second-level: Half-finished Beans
private final Map<String, Object> earlySingletonObjects = new ConcurrentHashMap(16);

// Third-level: Bean factories (lambdas)
private final Map<String, ObjectFactory<?>> singletonFactories = new HashMap(16);
```
https://seamew.top/article/spring%E8%BF%9B%E9%98%B6%E5%AD%A6%E4%B9%A0/%E4%B8%89%E7%BA%A7%E7%BC%93%E5%AD%98%E8%A7%A3%E5%86%B3%E5%BE%AA%E7%8E%AF%E4%BE%9D%E8%B5%96.html

Reference ZooKeeper Stat attributes

The Stat object provides metadata about a ZooKeeper node:

Attribute	Description
cZxid	Transaction ID when the node was created
ctime	Time when the node was created
mZxid	Transaction ID of the last update to the node
mtime	Time of the last update to the node
pZxid	Transaction ID of the last change to the node's children list
cversion	Version number of the child nodes
dataVersion	Version number of the node's data
aclVersion	Version number of the node's ACL
ephemeralOwner	SessionID of the creator if ephemeral; 0 if persistent
dataLength	Length of the data content
numChildren	Current number of child nodes

Understand ZooKeeper's ZAB (ZooKeeper Atomic Broadcast) Protocol
ZooKeeper uses the ZAB protocol to ensure data synchronization across nodes. It operates in two primary modes:
1. Recovery Mode (Leader Election): Triggered when the leader node fails. The remaining nodes elect a new leader, which then transitions the cluster into broadcast mode.
2. Broadcast Mode: Used to synchronize data across all nodes once a leader is established.
https://www.seamew.top/article/%E5%A4%A7%E6%95%B0%E6%8D%AE/zookeeper.html

Reference ZooKeeper Stat properties

The Stat object provides metadata about a ZooKeeper node:

Property	Description
cZxid	Transaction ID when the node was created
ctime	Creation time of the node
mZxid	Transaction ID of the last update to the node
mtime	Time of the last update to the node
pZxid	Transaction ID of the last change to the node's children list
cversion	Version number of the child nodes
dataVersion	Version number of the node's data
aclVersion	Version number of the node's ACL
ephemeralOwner	SessionID of the creator if ephemeral; 0 if persistent
dataLength	Length of the data content
numChildren	Current number of child nodes

Understand the relationship between Backtracking and Depth-First Search (DFS)
Backtracking and Depth-First Search (DFS) are complementary techniques. Both are concrete implementations of the recursive thought process, where the algorithm explores as far as possible along each branch before backtracking to the previous node to try different paths.
https://www.seamew.top/article/%E7%AE%97%E6%B3%95/%E5%9B%9E%E6%BA%AF%E7%AE%97%E6%B3%95/%E6%9C%80%E4%BC%98%E8%A7%A3%E5%BF%AB%E9%80%9F%E8%BF%94%E5%9B%9E.html
Understand the role of DefaultAdvisorAutoProxyCreator in Spring AOP
The DefaultAdvisorAutoProxyCreator class is responsible for creating proxy objects for beans that meet specific requirements. It primarily utilizes the postProcessBeforeInstantiation method to intercept bean creation. If a suitable TargetSource is found, it retrieves the necessary advices and advisors for the bean and uses a ProxyFactory to create the proxy object, effectively short-circuiting the standard instantiation process.
```
@Override
public Object postProcessBeforeInstantiation(Class<?> beanClass, String beanName) {
  TargetSource targetSource = getCustomTargetSource(beanClass, beanName);
  if (targetSource != null) {
    if (StringUtils.hasLength(beanName)) {
      this.targetSourcedBeans.add(beanName);
    }
    Object[] specificInterceptors = getAdvicesAndAdvisorsForBean(beanClass, beanName, targetSource);
    // Create proxy via ProxyFactory
    Object proxy = createProxy(beanClass, beanName, specificInterceptors, targetSource);
    this.proxyTypes.put(cacheKey, proxy.getClass());
    return proxy;
  }
  return null;
}
```
https://seamew.top/article/spring%E8%BF%9B%E9%98%B6%E5%AD%A6%E4%B9%A0/aop%E5%88%9B%E5%BB%BA%E4%BB%A3%E7%90%86%E5%9F%BA%E6%9C%AC%E6%B5%81%E7%A8%8B.html
Find the largest number smaller than n using a specific set of digits
Given a set of available digits A (containing numbers from 1-9) and a target number n, find the largest number that can be constructed using digits from A that is strictly less than n. Digits from A can be reused multiple times.
Examples:
- A={1, 2, 9, 4}, n=2533 → Result: 2499
- A={1, 2, 5, 4}, n=2543 → Result: 2542
- A={1, 2, 5, 4}, n=2541 → Result: 2525
- A={1, 2, 9, 4}, n=2111 → Result: 1999
- A={5, 9}, n=5555 → Result: 999
https://www.seamew.top/article/%E7%AE%97%E6%B3%95/%E8%B4%AA%E5%BF%83/%E5%B0%8F%E4%BA%8En%E7%9A%84%E6%9C%80%E5%A4%A7%E6%95%B0%E5%AD%97.html