联系方式

  • QQ:99515681
  • 邮箱:99515681@qq.com
  • 工作时间:8:00-21:00
  • 微信:codinghelp

您当前位置:首页 >> Java编程Java编程

日期:2020-11-20 10:57

Assignment 2 (Individual) 2020 (20%)

Deadlines

You must submit your final work using the stream submission system no later than Monday 16 Nov 2020. Please note that without an extension, submissions beyond this time are not possible.


You are expected to manage your source code, this includes making frequent backups. It is strongly recommended (but not required) to use a private repository for this assignment.

How to submit

1.Upload a zip file consisting of:

a.The Maven project folder (inc. pom.xml)

b.performance-analysis.pdf -measure time and memory consumption

c.coverage.pdf - the pdf version of the coverage report created by Maven

2.upload this file to stream - note: the max upload size is set to 20 MB

3.verify the submission: download the zip file, unzip it into a new folder and inspect content, run Maven from the command line, check the output including generated jar files

Task


Work individually to create the following program in Java.


Create a project assign251_2 using the Maven project layout, and within this project, create a project that implements custom appender and layout objects for Log4j. For this, you will need to create an appender and layout that work with the other Log4j objects (i.e. implementing relevant Log4j abstract classes or interfaces), test them and run profiling tools on them to gauge their correctness and efficiency.


Note there is no main class for this project, it will be run via your tests from sections 3 and 4. You may want to consider a test driven development methodology where your first step is to start with section 3 and work backwards. This will allow you to check that your classes are working correctly as you go.


1.Implement a log4j appender assign251_2.MemAppender  [6 marks]


In this task, you will need to implement a custom log4j appender, which can be used directly with the log4j logger. This MemAppender, unlike normal appenders, stores logs in memory and prints them on demand. There is a limit to how many log events will be kept in memory (this should be configurable), and if the maximum is reached, the oldest logs should be deleted.


Implementation details:

-It enforces the singleton pattern.

-It stores the LoggingEvents in a list. This is supplied by dependency injection (note: if you have already created a default, that is okay).

-It will need a layout. This will need to be able to be supplied when the MemAppender is collected, and via the setLayout() method. If a layout is not supplied, and code calling it is needed, appropriate precondition checks should be used (as some code may not use the appender with the layout, so it is a valid option not to supply one, as long as you don’t use any functionality that requires it).

-There are three ways to get information about the LoggingEvents that it stores:

a.Call the method getCurrentLogs() which will return an unmodifiable list of the LoggingEvents.

b.Call the method getEventStrings() which will return an unmodifiable list of strings (generated using a layout stored in the MemAppender).

c.Call the method printLogs() which will print the logging events to the console using the layout and then clear the logs from its memory.

-It has a property called maxSize, which needs to be configurable. When this size is reached, the oldest logs should be removed to make space for the new ones.

-The number of discarded logs should be tracked, and can be accessed using getDiscardedLogCount(). This should be stored as a long type, as there may be many discarded logs.


Note: Be careful to observe the DRY principle - there are overlapping requirements above.


3 marksCorrect implementation of the singleton pattern and dependency injection options for the list and layout.

2 marksCorrect implementation of the information printing / collection methods, along with sensible precondition checks where appropriate.

1 markCorrect implementation of maxSize and associated features.



2.a layout assign251_2.VelocityLayout  [3 marks]

a.VelocityLayout basically works like PatternLayout, but uses Velocity as the template engine. This layout should work with log4j appenders as well as the MemAppender.

b.Variable to be supported:

i.c (category)

ii.d (date using the default toString() representation)

iii.m (message)

iv.p (priority)

v.t (thread)

vi.n (line separator)

c.This means that the variable syntax is different, e.g. use $m instead of %m

d.VelocityLayout should have options to set its pattern both in the constructor and via a setter. An example string pattern could look like:

“[$p] $c $d: $m”

3.write tests that test your appender and layout in combination with different loggers, levels and appenders  [4 marks]

a.Use JUnit for testing your appender and layout. Aim for good test coverage and precise asserts. [2 marks]

b.Use the tests to show both the appender and layout working with different combinations of built in log4j classes as well as with each other. [2 mark]

c.Tests should be stored in the appropriate locations according to the Maven folder structure.

4.write tests to stress-test your appender/layout by creating a large amount of log statements  [5 marks]

a.Create a separate test class for stress tests.

b.Use these tests to compare the performance between MemAppender using a LinkedList, MemAppender using an ArrayList, ConsoleAppender and FileAppender - measure time and memory consumption (using JConsole or VisualVM or any profiler)

c.Consider how to output your logs in such a way that makes comparisons between the MemAppender and other appenders sensible.

d.Use these scripts to compare the performance between PatternLayout and VelocityLayout

e.Stress tests should test performance before and after maxSize has been reached, and with different maxSize values.

i.parameterised tests may be helpful here.

f.Write a short report summarising your findings (embed screenshots of memory usage charts in this reports taken from VisualVM)

g.The report name should be performance-analysis.pdf

h.Measure your test coverage of the written tests by generating a branch and statement coverage reports using Jacoco or Emma.


Note that the marks for this section will be based around your reporting, the effectiveness of your stress tests in probing into the efficiency of the classes, and the overall integration testing, checking that these classes work in combination with other relevant out-of-the-box classes.



5.write a Maven build script  [2 marks]

a.The Maven script should be used the build the project including compiling, testing, measuring test coverage, dependency analysis.

b.Use the jacoco Maven plugin for measuring test coverage.



Hints


●You can use any development environment you prefer as it is a Maven project.

●Library whitelist: only the following libraries and libraries they depend on can be used: Apache log4j, Apache Velocity, Junit, Guava, Apache Commons Collections, Jacoco and Emma.


Penalties


1.Code that is not self-documenting, or long or complex methods.

2.Violating the Maven standard project layout or Java naming conventions.

3.Use of absolute paths (e.g., libraries should not be referenced using absolute paths like “C:\\Users\\..”, instead use relative references w.r.t. the project root folder)

4.References to local libraries (libraries should be referenced via the Maven repository)

5.Use of libraries not on the whitelist



Bonus Question [2 marks]

You can get 100% for the assignment without this. This will give you additional marks up to the maximum if you lose some elsewhere.


Create an MBean object for each instance of the MemAppender to add JMX monitoring to this object, the properties to be monitored are:

1.the log messages as array

2.the estimated size of the cached logs (total characters)

3.the number of logs that have been discarded



版权所有:编程辅导网 2021 All Rights Reserved 联系方式:QQ:99515681 微信:codinghelp 电子信箱:99515681@qq.com
免责声明:本站部分内容从网络整理而来,只供参考!如有版权问题可联系本站删除。 站长地图

python代写
微信客服:codinghelp