add three event callbacks

[eakkew]

1. beforePrintEvent: callback function printEvent in iframe
2. beforePrint: callback function will be triggered before iframe created
3. afterPrint: callback function will be triggered before iframe removed

i use beforePrint to make my canvas an image - in my case, chartjs in bootstrap with div class="col"
then afterPrint for removing images created

function beforePrint() {
  console.log('before')
  for (var id in Chart.instances) {
    $(Chart.instances[id].canvas).addClass('d-none').closest('.chart-container')
      .append($('<img>').attr('src', Chart.instances[id].toBase64Image()).addClass('w-100'));
    $('.chartjs-size-monitor').addClass('d-none');
  }
}

$('#report').printThis({
    pageTitle: $('#report-title').text(),
    beforePrint: beforePrint,
});

[jasonday]

Thanks for the pull request. This is very similar to pull request #54 and you can read my reasoning as to why I did not merge that code.

Here's my reasoning (quoting my comment from #54):

The browser stops all processing/rendering when the print dialog is open and restarts when the print dialog is closed. With this pull request, the complete.printThis is not actually executing because it is tied to an event, it is merely in the "queue" when the browser is again active. Because it is arbitrary without specificity, I have not merged this code. However, I have left the pull request open for informative purposes, because for many users this complete event (arbitrary or not) may be useful.

Because these events are not true events (such as an actual complete event from the browser), I have concerns over actual timing, performance, and behavior of these actions because they depend on the code executing sequentially, which is fairly arbitrary.

@oculus42 - any opinion/feedback here?

[eakkew]

It's quite similar but mine using callback function as a hook instead of event trigger. So, I think this pull will execute the hooks in correct sequence.

ps. i tested in chrome, will test more in firefox and safari

[oculus42]

It has been my experience that Firefox has not always been reliable about the timing of the print events and the state of the page.

I am working on some tests to execute against a variety of different platforms.

[oculus42]

I haven't competed the test suite, and I would like to get to a point where we can identify the behavior of the supported browsers, but I think this would be a good addition.