feat : added condition to check the RHS of :=

[venom1204]

Closes #5829

changes made-

• Added a check to handle cases where a function (closure) is mistakenly used on the RHS of :=.
• Specifically checks: is.function(jval)
• Placed inside the if (length(newnames) > 0) block — so it only triggers when creating a new column.
• Prevents malformed data.table objects and confusing errors.

At present, this only covers the function case. Other exotic RHS types (e.g., environments, external pointers) are not yet handled.

Is there anything else that should be checked or added to this condition?
i closed the previous pr and opened this, the whole summary is provided above.

hi @MichaelChirico @tdhock can you please review this when you have time

thanks for your time.

[github-actions]

• HEAD=issue_5829 slower P<0.001 for memrecycle regression fixed in #5463
• HEAD=issue_5829 slower P<0.001 for setDT improved in #5427
  

Generated via commit c8cec80

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	20 seconds
Running and plotting the test cases	2 minutes and 34 seconds

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.51%. Comparing base (2f0d12f) to head (c8cec80).

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #7161   +/-   ##
=======================================
  Coverage   98.50%   98.51%           
=======================================
  Files          81       81           
  Lines       15032    15036    +4     
=======================================
+ Hits        14808    14812    +4     
  Misses        224      224           

☔ 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.

[tdhock]

Why did you open a new PR? https://github.com/Rdatatable/data.table/pull/6742/files has the same changes, right? Can you please close one or the other?

[venom1204]

i closed the previous one ,i thought to make a branch in this repo so ...

[MichaelChirico]

What's the difference between this assignment and the one above?

data.table/R/data.table.R

Line 1218 in 2f0d12f

newnames=setdiff(lhs, names_x)

[venom1204]

both compute setdiff(lhs, names(x)), but the one above uses the cached names_x. I’ll update the later one to use names_x as well for consistency and to avoid unnecessary recomputation.

[MichaelChirico]

sorry i don't understand, why recalculate newnames at all if it's just the same value as above?

[venom1204]

Got it, thanks. The second newnames assignment was redundant. I will take care of it.

[MichaelChirico]

does this error not happen for the case of overwriting an existing column?

[MichaelChirico]

I don't think this will catch the following case (which you should add as a new regression test):

DT=data.table(a=1:10)
DT[,c('a', 'b'):=.(1:10, mean)]

[MichaelChirico]

Overall I suspect it will be more natural to put the fix here inside C

[venom1204]

Okay, that makes sense . I’ll try to implement it

So far, I’m aware to handle these cases:

• Direct assignment of a function: DT[, b := mean]

• Function inside a list: DT[, b := list(mean)]

• Function assigned to a new column

Are there any other specific cases or edge conditions I should be aware of? Let me know so I can cover them in the implementation.

[MichaelChirico]

Function assigned to a new column

I'm not sure how this differs from the first case.

You might try triggering this error from set() as well as :=.