Introduction to PMD source code analyzer

When checking the code , It is often seen that the specification on which a defect is based comes from PMD. The frequency of seeing is higher , Curiosity comes ：PMD What is it ？ Why does it appear so frequently ？

So I searched the Internet , Find out PMD It means a lot of strange things . Can represent Programming Mistake Detector（ Static code checking tool ）,Program Managment Document（ Project management documents ）,Physical Media Dependent（ Physical media decides ）,Pretty Much Done（ Almost finished ）,Project Meets Deadline（ It's the deadline for the project ） wait , Except that the first one has something to do with our code checking , In addition, we should understand it .

But what we are going to talk about here PMD In fact, it is used by the static code check plug-in tool Source code analyzer （source code analyzer）. It can detect common programming defects in the code without running the project code （ Such as unused variables / Empty capture block , Unnecessary object creation, etc ）, The way is to compare with the rules that have been made in advance （ Rules can be customized and extended , You can use Java To write , You can also use XPath Inquire about ）, Compare the parts of the code that do not conform to the rules .

Besides ,PMD It also includes CPD（copy-paste-detector）, A copy and paste detector . It can be C/C++, C#, Dart, Fortran, Go, Groovy, Java, JavaScript, JSP, Kotlin, Lua, Matlab, Modelica, Objective-C, Perl, PHP, PLSQL, Python, Ruby, Salesforce.com Apex, Scala, Swift, Visualforce and XML Duplicate part found in language file .

PMD Support for multiple languages ：Java, JavaScript, Salesforce.com Of Apex and Visualforce, Modelica, PLSQL, Apache Velocity, XML, XSL, Scala.

PMD How to detect defects

PMD How does it work ？

It USES JavaCC（ An open source parser generator and lexical analyzer generator ） and Antlr（ be based on LL Algorithm implementation of the syntax parser generator ）, Parse the source file into an abstract syntax tree ; Then, according to the pre-determined rules, the corresponding nodes in the tree , Analyze its properties or structure , So as to find out the corresponding violations .

Abstract syntax tree , namely AST-Abstract Syntax Tree, Will pass the code structure through “ Trees ” In the form of , Code of each part （ For example, variable declaration 、if-else sentence 、 Variable assignment 、 class 、 Database query, etc ） Will become each node of the tree .

for instance ：

public class ClassA {	public  static  void MethodB() {     if (false) {     	String varStr = "hello HW PaaS!";     }  }}

After converting the above code into a syntax tree , Its structure is similar to UserClass “ClassA” — Method “MethodB” — BlockStatment — IfElseBlockStatment — VariableDeclaraion “varStr”; It can be seen that , The root node of this tree is the class declaration , A child node is a function in a class , Mobile network type if sentence , Variables and other elements .

How to use PMD

Download binary compression package

We from Released version link Download the latest binary compressed package , Then extract it locally ：

After decompression , Carry out orders bin/run.sh pmd Or bin\pmd.bat. Then the command line shown in the following picture , We can check for a specific code file , The following figure shows PMD Detected in the file 1 A violation of the rules ：“for loop , Can be for-each Circular substitution ”：

Reference material

1. https://github.com/pmd/pmd