rec: a single NSEC3 record covering everything is a special case

[omoerbeek]

Fixes #13542

Short description

Checklist

I have:

• read the CONTRIBUTING.md document
• compiled this code
• tested this code
• included documentation (including possible behaviour changes)
• documented the code
• added or modified regression test(s)
• added or modified unit test(s)

[coveralls]

Pull Request Test Coverage Report for Build 7058058194

• 6 of 6 (100.0%) changed or added relevant lines in 2 files are covered.
• 52 unchanged lines in 11 files lost coverage.
• Overall coverage decreased (-0.02%) to 57.369%

---

Files with Coverage Reduction	New Missed Lines	%
pdns/pollmplexer.cc	1	82.61%
pdns/sstuff.hh	1	61.99%
pdns/query-local-address.cc	2	76.6%
pdns/packethandler.cc	3	72.84%
pdns/stubresolver.cc	3	81.63%
pdns/tsigverifier.cc	3	77.22%
pdns/signingpipe.cc	5	83.06%
pdns/ws-auth.cc	5	80.61%
pdns/recursordist/test-syncres_cc1.cc	7	88.91%
pdns/dnsdist-tcp.cc	8	75.82%

Totals
Change from base Build 7057769616:	-0.02%
Covered Lines:	106092
Relevant Lines:	153791

---

💛 - Coveralls

[rgacogne]

Change looks OK to me. It feels a bit weird to do this test in commonPrefixIsLong, I wonder if it wouldn't make more sense to have it in AggressiveNSECCache::isSmallCoveringNSEC3 instead, before calling commonPrefixIsLong, but this makes no functional difference.

[omoerbeek]

Change looks OK to me. It feels a bit weird to do this test in commonPrefixIsLong, I wonder if it wouldn't make more sense to have it in AggressiveNSECCache::isSmallCoveringNSEC3 instead, before calling commonPrefixIsLong, but this makes no functional difference.

Right, I had the same thought this morning. Will change.

[pspacek]

I confirm it works in my tests case with an empty zone.