rchk subsetDT issues

[ben-schwen]

Function subsetDT
  [UP] calling allocating function subsetVectorRaw with argument allocated using Rf_getAttrib(?,S:names) data.table/src/subset.c:334
  [UP] unprotected variable key while calling allocating function Rf_getAttrib(?,S:names) data.table/src/subset.c:346
  [UP] allocating function chin may destroy its unprotected argument (key <arg 1>), which is later used. data.table/src/subset.c:347
  [UP] calling allocating function chin with a fresh pointer (key <arg 1>) data.table/src/subset.c:347
  [UP] unprotected variable key while calling allocating function Rf_allocVector data.table/src/subset.c:355
  [UP] unprotected variable key while calling allocating function Rf_setAttrib(?,S:sorted,V) data.table/src/subset.c:355

[MichaelChirico]

Strange that our rchk action has totally different results than shown on CRAN 😅

Library name (usually package name): data_table
Initialization function: R_init_data_table
Functions: 110
Functions: 1
Checked call to R_registerRoutines: 1
ERROR: too many states (abstraction error?) in function strptime_internal
ERROR: too many states (abstraction error?) in function bcEval_loop
ERROR: too many states (abstraction error?) in function StringValue
ERROR: too many states (abstraction error?) in function RunGenCollect
ERROR: too many states (abstraction error?) in function rbindlist
ERROR: too many states (abstraction error?) in function freadMain
ERROR: too many states (abstraction error?) in function fcaseR
ERROR: too many states (abstraction error?) in function rbindlist

Function cbindlist
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] unsupported form of unprotect, unprotecting all variables, results will be incomplete /rchk/packages/build/loowXoIa/data.table/src/mergelist.c:85

Function fcaseR
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] unsupported form of unprotect, unprotecting all variables, results will be incomplete /rchk/packages/build/loowXoIa/data.table/src/fifelse.c:414

Function forder
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] unsupported form of unprotect, unprotecting all variables, results will be incomplete /rchk/packages/build/loowXoIa/data.table/src/forder.c:867

Function rbindlist
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] protect stack is too deep, unprotecting all variables, results will be incomplete
  [UP] unsupported form of unprotect, unprotecting all variables, results will be incomplete /rchk/packages/build/loowXoIa/data.table/src/rbindlist.c:565
Analyzed 387 functions, traversed 6466838 states.

Rchk version: baa186024c3a38e5db74ac4a72f8e2c1e04417ef
R version: 87292/R Under development (unstable) (2024-11-06 r87292)
LLVM version: 14.0.0

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.97%. Comparing base (3acb570) to head (18b3147).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #7520   +/-   ##
=======================================
  Coverage   98.97%   98.97%           
=======================================
  Files          87       87           
  Lines       16739    16739           
=======================================
  Hits        16567    16567           
  Misses        172      172           

☔ 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.

[ben-schwen]

Strange that our rchk action has totally different results than shown on CRAN 😅

I just noticed that too :/

[MichaelChirico]

I do see the docker image is using a rather old version of R, maybe there's a change. CRAN's version is very recent:

Package built using 89227/R 4.6.0; x86_64-pc-linux-gnu; 2025-12-26 07:45:04 UTC; unix  

[github-actions]

No obvious timing issues in HEAD=subset_rchk

Generated via commit 18b3147

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

Task	Duration
R setup and installing dependencies	2 minutes and 51 seconds
Installing different package versions	45 seconds
Running and plotting the test cases	4 minutes and 21 seconds

[MichaelChirico]

Requested a new rchk image be pushed: kalibera/rchk#40

[MichaelChirico]

CRAN also shows one in _selfrefok, is that excluded intentionally?

Function _selfrefok
  [UP] unprotected variable v while calling allocating function Rf_getAttrib(?,S:names) data.table/src/assign.c:73

[ben-schwen]

CRAN also shows one in _selfrefok, is that excluded intentionally?

Function _selfrefok
  [UP] unprotected variable v while calling allocating function Rf_getAttrib(?,S:names) data.table/src/assign.c:73

Yes, I wanted to dig deeper if we really need the PROTECT in _selfrefok since we use it extensively now