Fix latest rchk issues

[MichaelChirico]

Following kalibera/rchk#40, rchk is showing some new uncaught PROTECT() issues.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.02%. Comparing base (f3b166b) to head (09d384c).
⚠️ Report is 5 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #7585   +/-   ##
=======================================
  Coverage   99.02%   99.02%           
=======================================
  Files          87       87           
  Lines       16754    16803   +49     
=======================================
+ Hits        16591    16640   +49     
  Misses        163      163           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

• HEAD=rchk slower P<0.001 for DT[by,verbose=TRUE] improved in #6296
  

Generated via commit 09d384c

Download link for the artifact containing the test results: ↓ atime-results.zip

Task	Duration
R setup and installing dependencies	2 minutes and 49 seconds
Installing different package versions	43 seconds
Running and plotting the test cases	4 minutes and 23 seconds

[MichaelChirico]

ah, I forgot about reprotect too. let me try that later.

…

On Sun, Jan 11, 2026, 9:37 AM Benjamin Schwendinger < ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In src/rbindlist.c <#7585 (comment)> : > @@ -303,7 +303,8 @@ SEXP rbindlist(SEXP l, SEXP usenamesArg, SEXP fillArg, SEXP idcolArg, SEXP ignor if (isOrdered(thisCol)) { orderedFactor = true; int thisLen = length(getAttrib(thisCol, R_LevelsSymbol)); - if (thisLen>longestLen) { longestLen=thisLen; longestLevels=getAttrib(thisCol, R_LevelsSymbol); /*for warnings later ...*/longestW=w; longestI=i; } + if (longestLen == -1) longestLevels = PROTECT(R_NilValue); seems we really need to go with the SEXP longestLevels = PROTECT(R_NilValue); nprotect++ — Reply to this email directly, view it on GitHub <#7585 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB2BA5JPX2MWY63O577QNHD4GKC5XAVCNFSM6AAAAACRJUWZ2CVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZTMNBYGE2TMNBVGY> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[MichaelChirico]

Thanks for jamming on this @ben-schwen! I always forget about REPROTECT(), it's clearly the "right" thing to do here. I think we are good to go now.