antlrjavaparser / antlr-java-parser Goto Github PK

The transient modifier is being lost when parsing files.

Example:

private transient String myfield;

Becomes:

private String myfield;

my some java file has unicode variable
that euc-kr encoding and korea language likely "한글변수"

ex
String 한글변수 = "some string";

error
token recognition error at: '한'
token recognition error at: '글'
token recognition error at: '변'
token recognition error at: '수'

Which version of Java does it support? Is it complete? How does it compare to https://github.com/javaparser/javaparser? Are you looking for help for this project?

Thanks!

Node members beginLine, beginColumn, endLine, endColumn seems to be never set.
Using version 1.0.12 from Maven Central

Not available LICENSE file in source directory structure
Please. Added license and copyright notice.
the fedora pakaging guideline is very strictly precise about this problem
https://fedoraproject.org/wiki/Packaging:LicensingGuidelines?rd=Packaging/LicensingGuidelines#License_Text
thanks
regards

I cloned this repo and flowwed instructions from Readme.md
The step "mvn release:prepare" don't succeed.

Failed tests:
TestAstPosition.ifStatementTest:39 Method body index start is incorrect. expected:<80> but was:<75>

I don't understand, because I don't change the source files.

Current version (1.0.15) cannot parse shift operators. Calling JavaParser.parse() will parse the '<<' as '<=' and '>>' as '<'. Besides, it seems that '>>>' cannot be parsed either.

Sample input file:

public class Debug{
    public Debug(){
        1 << 6; //1 <= 6
        1 >> 6; //1 < 6
    }
}

Example:

public class TrySomeChineseCharacters {
private String test1 = "已删除";
private String test2 = "已发布";
}

Parser fails on following code:
return stepBuilders.get( "step" )
.transactionManager( transactionManager )
.<MyEntity, MyEntity>chunk( 20 )
.reader( reader() )
.processor( loggingProcessor() )
.writer( writer() )
.build();

The problem seems to be with the .< part.

Cheers

I created simple test with parsing java files. I've cloned netbeans sources and use this code:

import com.github.antlrjavaparser.JavaParser;
import com.github.antlrjavaparser.api.CompilationUnit;
import java.io.File;

public final class TestAntlrParser {

    public static void main(String[] args) throws Exception {
        parseDirectory(new File("/data/home/hlavki/Download/main-silver"));
    }


    public static void parseDirectory(File path) {
        if (path.isDirectory()) {
            File[] files = path.listFiles();
            for (File file : files) {
                if (!file.getAbsolutePath().contains("/.hg/") && file.getName().endsWith(".java")) {
                    try {
                        CompilationUnit cu = JavaParser.parse(file);
                        if (!cu.getTypes().isEmpty()) {
                            System.out.println(cu.getPackage().getName() + "." + cu.getTypes().get(0).getName());
                        }
                    } catch (Throwable e) {
                        System.out.println("ERROR: " + e.getMessage());
                    }
                } else if (file.isDirectory()) {
                    parseDirectory(file);
                }
            }
        }
    }
}

And created 3 heap dumps while parsing minute after minute. You can see increasing number of ATNConfig instances.

The statements in the default block of a switch statement disappear. it manifests itself when running with DumpVisitor. I think this is due to the else if statement in SwitchBlockStatementGroupContextAdapter:

else if (context.switchLabel().DEFAULT() != null) {
// Explicitly setting these null for readability
switchEntryStmt.setLabel(null);
switchEntryStmt.setStmts(null);
}