StringIndexOutOfBoundsException

[havrikov]

When parsing the attached file a StringIndexOutOfBoundsException is thrown.
flexmark_sioobe.zip
Kotlin code to reproduce assuming the string x contains the file contents:
Parser.builder().extensions(listOf(AbbreviationExtension.create())).build().parse(x)

Should the input be invalid Markdown, an IllegalArgumentException is more appropriate.

[vsch]

@havrikov, thank you for the heads up.

I will take a look at it. I am preparing a new release so it should get fixed shortly.

[vsch]

@havrikov, there is no invalid markdown. That's what makes it so complex to work with.

I tried it on the pre-release build I am working with, in Java because I cannot use Kotlin for the project (you had to mention Kotlin to rub it in) I cannot even use Java 8, to be 7 compatible.

However, it works.

Can you please try it in your setup. Here is the equivalent Java as a ready console app. I want to know for sure it is a bug I already fixed.

If you get an exception, a stack trace printout would help.

import com.vladsch.flexmark.ast.Node;
import com.vladsch.flexmark.ext.abbreviation.AbbreviationExtension;
import com.vladsch.flexmark.html.HtmlRenderer;
import com.vladsch.flexmark.parser.Parser;

import java.util.Arrays;

public class Issue198 {
    public static void main(String[] args) {
        String input = "[![ReportCard][ReportCard-Image]][ReportCard-URL] \n" +
                "[![Build][Build-Status-Image]][Build-Status-URL] \n" +
                "\n" +
                "[ReportCard-URL]: https://goreportcard.com/report/jonaswouters/goclerk\n" +
                "[ReportCard-Image]: https://goreportcard.com/badge/jonaswouters/goclerk\n" +
                "[Build-Status-URL]: https://travis-ci.org/jonaswouters/goclerk\n" +
                "[Build-Status-Image]: https://travis-ci.org/jonaswouters/goclerk.svg?branch=master\n";

        Parser parser = Parser.builder().extensions(Arrays.asList(AbbreviationExtension.create())).build();

        Node node = parser.parse(input);

        HtmlRenderer renderer = HtmlRenderer.builder().extensions(Arrays.asList(AbbreviationExtension.create())).indentSize(4).build();
        System.out.println(renderer.render(node));
    }
}

[havrikov]

What you wrote works for me, but I reduced my original input and can still reproduce the issue as follows (on version 0.28.32):

import com.vladsch.flexmark.ext.abbreviation.AbbreviationExtension;
import com.vladsch.flexmark.parser.Parser;

import java.util.Collections;

public class Issue198 {

  public static void main(String[] args) {
    Parser parser = Parser.builder().extensions(Collections.singletonList(AbbreviationExtension.create())).build();
    parser.parse("\n*[]:a");
  }
}

[vsch]

@havrikov, sorry about the code I used. I don't know where that came from. I thought I grabbed the unzipped file you sent but it was some other sample.

Yes, I can duplicate this. Abbreviation definitions with a blank abbreviation cause the exception.

[vsch]

@havrikov, new release has this fixed. One caution is that there was an API change in the attribute provider factory and html id generator factory to allow re-use of these between HTML and Docx Renderers. Previously NodeRendererContext was passed, now LinkResolverContext. Access to options is still available. Should not affect code beyond class name change for the parameters.

see VERSION.md for details.