Can't access maven repo through URLConnection or HttpsURLConnection (403)

Purpose, just a POC (for now) to automatically and periodically find some CVE tags in the maven repository.

I can access maven just fine through browser and mvn, but am unable to do the same via Java, what am I missing? I've tried UrlConnection, HttpsURLConnection, with and without GET, Content-type, User-Agent, and Accept, it always returns a 403 for all addresses that I try, the same code works fine on other websites like "cve.mitre.org" or "nvd.nist.gov", but fails for "https://mvnrepository.com/artifact/log4j/apache-log4j-extras/1.2.17".

My URL is been built dynamically, with the start "**https://mvnrepository.com/artifact/**", then adding the group, name, and version are added, turning it into a valid address like "https://mvnrepository.com/artifact/log4j/apache-log4j-extras/1.2.17"

    System.setProperty("https.proxyHost", "xxxx");
    System.setProperty("https.proxyPort", "xxxx");

    String content = null;
    try {
        URL obj = new URL(address);
        HttpsURLConnection con = (HttpsURLConnection) obj.openConnection();
        con.setRequestMethod("GET");
        con.setRequestProperty("Content-Type", "application/json");
        con.setRequestProperty("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36");
        con.setRequestProperty("Accept", "*/*");

        con.connect();
        
        BufferedReader br;
        
        if (con.getResponseCode() < 300) {
            br = new BufferedReader(new InputStreamReader(con.getInputStream(), StandardCharsets.UTF_8));
        } else {
            br = new BufferedReader(new InputStreamReader(con.getErrorStream(), StandardCharsets.UTF_8));
        }            

        final StringBuilder sb = new StringBuilder();
        String line;
        while ((line = br.readLine()) != null) {
            sb.append(line);
        }
        br.close();

Solution

This web use anti-bot security CloudFlare.
How to bypass CloudFlare bot protection?
It depends.... Sometimes it is very difficult task or impossible. That you need to do, is simulate a real user with the browser.
With htmlunit browser you can bypass it in this case only and with a good IP address. (i use my own ip address and did only one request)

You need maven dependency:

<dependency>
    <groupId>net.sourceforge.htmlunit</groupId>
    <artifactId>htmlunit</artifactId>
    <version>2.57.0</version>
</dependency>

Here you have some java example:

import com.gargoylesoftware.htmlunit.WebClient;
import com.gargoylesoftware.htmlunit.html.HtmlAnchor;
import com.gargoylesoftware.htmlunit.html.HtmlPage;
import java.io.IOException;
import java.net.URL;
import java.util.List;

public class Maven {

    public static void main(String[] args) throws IOException {

        try (final WebClient webClient = new WebClient()) {
            webClient.getOptions().setJavaScriptEnabled(false);
            URL target = new URL("https://mvnrepository.com/artifact/log4j/apache-log4j-extras/1.2.17");
            final HtmlPage page = webClient.getPage(target);
            List<HtmlAnchor> elements = page.getByXPath("//a[contains(@class, 'vuln')]");
            elements.forEach(element -> System.out.println(element.getTextContent()));
        }
    }
}

OUTPUT:

CVE-2022-23305
CVE-2022-23302
CVE-2021-4104
CVE-2019-17571
View 1 more ...
4 vulnerabilities

I hope I have been able to help you.