fix: handle JSON character encoding for hec (Umlauts)

[zszia]

To reproduce:

data = {"field_a": "Üü_Öö_Ää_some_text", "field_b": "some_text_Üü_Öö_Ää"}
event = event_writer.create_event(data, time=time.time(), source="solnlib_demo", sourcetype="solnlib_demo")
event_writer.write_events([event])

When indexing data with special characters, these are implicitly UTF-8 encoded.

Here how the raw data looks like in Splunk:

Searching for field_a="Üü_Öö_Ää_some_text" is not working. The only way is to use star (*):

field_a="*_some_text"

which is very inefficient if the first character is an umlaut.

The other workaround would be to search for
field_a="\u00dc\u00fc_\u00d6\u00f6_\u00c4\u00e4_some_text"
which is not practicable.

We are using the "Splunk Add-on for Microsoft Cloud Services" app, which is using solnlib
https://splunkbase.splunk.com/app/3110/
When our security team is trying to find names with umlauts (example: Müller), then it's not working, which is a big problem.
They practically always have to use *. And as you know the azure data is pretty big.

If you are using syntax highlighting, then the data is displayed identical in both cases, which is very confusing for the users, which expects the search for Umlauts to work.

After the fix the raw data looks like this and the Splunk searches are working correctly:

[ghost]

I passed this issue to the team who is working on "Splunk Add-on for Microsoft Cloud Services", next update will be next week.

[artemrys]

@zszia thanks for the contribution, sorry that it took so long to merge it, I've just added tests to verify the behaviour for the future.

[srv-rr-github-token]

🎉 This PR is included in version 4.8.2 🎉

The release is available on:

• v4.8.2
• GitHub release

Your semantic-release bot 📦🚀