Splunk query for mixed content in CSP reports

gistfile1.txt

A csp report looks like 

{
  "csp-report": {
    "document-uri": "http://example.org/page.html",
    "referrer": "http://evil.example.com/haxor.html",
    "blocked-uri": "http://evil.example.com/image.png",
    "violated-directive": "default-src 'self'",
    "original-policy": "default-src 'self'; report-uri http://example.org/csp-report.cgi"
  }
}

Throw away the outer "csp-report". Setup a splunk input, set type to json. Run query.

"document-uri"=https* "blocked-uri"=http://* | rex field="blocked-uri" "https?://(?<blocked_host>.*)" | top blocked_host

Fix mixed content.

Throw away the outer "csp-report". Setup a splunk input, set type to json.

What does this mean? What are the splunk commands to split the body on csp-report="..." and parse the rest into JSON?

I figured out how to parse the payload into JSON. This queries csp-report POSTs, uses rex to isolate the JSON, uses spath to access some interesting JSON values, and puts it in a table...

index=* http_method="POST" "/csp-report*string-for-the-report-I-want"
| rex field=payload "csp-report=(?<raw_csp_report>.+)"
| spath input=raw_csp_report output=document_uri extra-fields.document-uri.host
| spath input=raw_csp_report output=user_agent client_info.user_agent
| spath input=raw_csp_report output=effective_directive csp-report.effective-directive
| table document_uri, effective_directive, user_agent