Java null pointer error in fetch() about sparkler HOT 9 CLOSED

To solve the issue, just add a throw in public FetchedData fetch(Resource resource) throws Exception, and also add a try/catch in public FetchedData next() around the fetch() function. Cheers

from sparkler.

This is a good fix. I understood that you want to remove the pain of handling such errors on modules. 👍

from sparkler.

@arelaxend Thanks for reporting this issue and also for a quick solution. I will investigate more before putting it around try and catch.
Just a suggestion - If you know what URLs are causing this issue and you don't want to crawl them as well, place URL regex filter(s). It will save you some time. :)

from sparkler.

Thanks @arelaxend If you could submit a pull request with those fixes, it will be awesome 👍 🥇

from sparkler.

@karanjeets
Instead of

Just a suggestion - If you know what URLs are causing this issue and you don't want to crawl them as well, place URL regex filter(s). It will save you some time. :)

I suggest we do this:

To solve the issue, just add a throw in public FetchedData fetch(Resource resource) throws Exception, and also add a try/catch in public FetchedData next() around the fetch() function.

from sparkler.

Hi, I got:
fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists.

Besides, you just have to update the function fetch() in the plugin fetcher-jbbrowser. I choose to do that to be coherent with what you have done in the fetch() function on the app folder. 👍 And so, it means to change the implementation of the fetch() function by the following code:

public FetchedData fetch(Resource resource) {
  LOG.info("JBrowser FETCHER {}", resource.getUrl());
  FetchedData fetchedData;
      /*
  	* In this plugin we will work on only HTML data
  	* If data is of any other data type like image, pdf etc plugin will return client error
  	* so it can be fetched using default Fetcher
  	*/
  try {
    if (!isWebPage(resource.getUrl())) {
      LOG.debug("{} not a html. Falling back to default fetcher.",
          resource.getUrl());
      //This should be true for all URLS ending with 4 character file extension
      //return new FetchedData("".getBytes(), "application/html", ERROR_CODE) ;
      return super.fetch(resource);
    }
    long start = System.currentTimeMillis();

    LOG.debug("Time taken to create driver- {}",
        (System.currentTimeMillis() - start));

    // This will block for the page load and any
    // associated AJAX requests
    driver.get(resource.getUrl());

    int status = driver.getStatusCode();
    //content-type

    // Returns the page source in its current state, including
    // any DOM updates that occurred after page load
    String html = driver.getPageSource();

    //quitBrowserInstance(driver);

    LOG.debug("Time taken to load {} - {} ", resource.getUrl(),
        (System.currentTimeMillis() - start));

    if (!(status >= 200 && status < 300)) {
      // If not fetched through plugin successfully
      // Falling back to default fetcher
      LOG.info(
          "{} Failed to fetch the page. Falling back to default fetcher.",
          resource.getUrl());
      return super.fetch(resource);
    }
    fetchedData = new FetchedData(html.getBytes(), "application/html",
        status);
    resource.setStatus(ResourceStatus.FETCHED.toString());
    fetchedData.setResource(resource);
    return fetchedData;
  } catch (Exception e) {
    LOG.info(
        "{} Failed to fetch the page. Falling back to default fetcher.",
        resource.getUrl());
    return super.fetch(resource);
  }
}

from sparkler.

Plus, for the selenium error, it happens when you build the app, but the jar file for the plugins are not generated! 👍

from sparkler.

Thanks for the comment

Hi, I got:
fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists.

What command did you execute to get this error message?
I hope you are aware of the way to raise pull request without having write permissions. If not, refer to http://stackoverflow.com/a/14681796/1506477. Basically, you (1) fork this repo (2) push your changes to your fork (3) raise a pull request from your fork to this repo.

Plus, for the selenium error, it happens when you build the app, but the jar file for the plugins are not generated!

Ah! That explains the NullPointerException - the 3rd stack trace.

from sparkler.

Fixed in #61
Waiting for one more person to review before I merge

from sparkler.