pdnsutil {add-record,delete-rrset}: Don't append ZONE if NAME ends with .

[ukleinek]

If a NAME ends with a . it is to be understood as an absolute name and appending the zone is not intuitive then.

Closes: #8595

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 12748862565

Details

• 1 of 1 (100.0%) changed or added relevant line in 1 file are covered.
• 37 unchanged lines in 10 files lost coverage.
• Overall coverage increased (+8.8%) to 73.684%

---

Files with Coverage Reduction	New Missed Lines	%
pdns/dnsdistdist/dnsdist-backend.cc	1	66.73%
pdns/misc.cc	2	64.85%
pdns/recursordist/sortlist.cc	2	72.94%
pdns/packethandler.cc	3	72.48%
pdns/remote_logger.cc	3	53.81%
pdns/iputils.cc	3	54.99%
pdns/recursordist/rec-main.cc	3	62.43%
pdns/recursordist/test-syncres_cc1.cc	5	89.91%
pdns/signingpipe.cc	7	84.18%
pdns/recursordist/ext/luawrapper/include/LuaContext.hpp	8	14.17%

Totals
Change from base Build 12745616875:	8.8%
Covered Lines:	55467
Relevant Lines:	71259

---

💛 - Coveralls

[ukleinek]

Pushed with using isCanonical(). While working on this change I found that this function open-codes boost::ends_with() and "fixed" that accordingly. Not sure this is benefical, so feel free to only take the first commit.

[mind04]

Add and delete record in pdnsutil are already somewhat special for pdns with the relative names. I'm not sure we should make them even more special by adding canonical logic.

If we touch this, I prefer switching to full names here like the rest of pdns and drop all the fiddling with apex canonical and/or relative names.

[ukleinek]

The failing check is clang-tidy which wants me to use { } around the if bodies:
statement should be inside braces (readability-braces-around-statements - Level=Warning)
The style I picked matches the surrounding code. Looking again there is a big mixure, some if statements use braces for one line bodies others don't. There are even mixed ones and inconsistencies if the else starts on a new line or after the closing brace. shrug

[miodvallat]

LGTM

[Habbie]

If we touch this, I prefer switching to full names here like the rest of pdns and drop all the fiddling with apex canonical and/or relative names.

I have to say I like this, except for breaking compatibility. But I agree the old form confuses people (including myself) all the time. And it in fact simplifies the code.

[ukleinek]

I reworked this now that adding a RR for a name that doesn't end in $zone uses the relative name (with a warning) and interprets as absolute otherwise. IMHO this is a tad nicer than erroring out (which was what we discussed on irc). With the warning added today, we can drop the relative interpretation in a later release with less expected fallout (assuming people read the warning and adapt their usage).