Nexpose::ReportAdHoc strange behavior with different formats

[thesp0nge]

Hi there, creating an ad hoc report, if you supply 'text' or 'csv' file format Rex library complains about parse errors.

Using 'html' or 'raw-xml' report is generated.

report = Nexpose::ReportAdHoc.new(nsc, 'report-card', 'text')
report.addFilter('scan', scan[:scan_id])
puts report.generate

[mdaines-r7]

I'm still seeing problems with the Adhoc reports after the refactoring. I didn't actually change the code that handles the report downloading.

What I'm observing is that the non-XML reports download as expected, but XML reports (incl. HTML) with multi-part MIME types don't download everything for the report. If a user doesn't care about the graphs, I think it might still work.

We need to revisit handling of the multi-part MIME types, maybe using a different library. It might also make sense to have a flag on the generate() method to indicate that graphics don't need to be downloaded. (I know that many people just want to parse the data.)

[mdaines-r7]

I have a patch now that will reliably provide the text of HTML and XML reports. It still does not return images, but works around a problem where some Nexpose instances were attempting to parse PNG data as XML and failing to return anything.

[mdaines-r7]

I think we've met the needs of this feature.