MISP Analyzer: Unable to filter for unpublished events due to boolean logic flaw

[ayushgupta704]

What happened

Right now, the MISP analyzer doesn’t properly handle filtering for unpublished events. Even if published=False is selected in the configuration, that value never actually makes it into the API request sent to MISP.
This happens because of the current check:
if self.published:
Since False evaluates to falsy in Python, the condition fails and the parameter gets skipped entirely. As a result, the request is sent without any published filter.

Environment

1. OS: Linux

What did you expect to happen

Setting published=False should send {"published": False} to the MISP API. This allows analysts to specifically hunt for fresh, indicators that analysts may want to specifically query.

How to reproduce your issue

1. Uncheck the published box in a MISP analyzer configuration.
2. Run any analysis.
3. Observe that the API call sent to MISP (line 100 of misp.py) contains no published parameter
   because if self.published: evaluates to False and skips the assignment.

I’ve already tested a fix for this locally. I’d be happy to take this up and open a PR please assign it to me.

[mlodic]

not sure this is a real issue and was not intended actually...also a fix would require a MISP installation and interaction with it to hard proof that it works

[gks281263]

This issue is a duplicate of #3429 and has already been resolved by PR #3565.

The fix (commit 644cb8f) replaced the truthiness check:

Before (broken)

if self.published:
params["published"] = self.published

with an isinstance check that correctly handles False:

After (fixed)

if isinstance(self.published, bool):
params["published"] = self.published

The same fix was also applied to the metadata parameter which had the identical bug. Unit tests covering the published=False case were added in test_misp.py (MISPPublishedFalseTestCase).
@mlodic