[xushiwei]

No description provided.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.09%. Comparing base (1afd686) to head (ac38380).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1114      +/-   ##
==========================================
+ Coverage   86.29%   91.09%   +4.79%     
==========================================
  Files          26       26              
  Lines        3459     3480      +21     
==========================================
+ Hits         2985     3170     +185     
+ Misses        474      310     -164     

Flag	Coverage Δ
llgo-tests	91.09% <100.00%> (+4.79%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[fennoai]

This is a focused, well-scoped change: it implements the previously-stubbed nodeInterp.LoadExpr, seeds the INT_MAX built-in macro (c_INT_MAX), threads source-position nodes through the auto-macro-func codegen, and adds debugMacroFunc logging. cl/... builds cleanly.

Verified correct: c_INT_MAX = (1 << (unsafe.Sizeof(c.Int(0))*8 - 1)) - 1 evaluates to 2147483647 (c.Int is a fixed int32 in goplus/lib), a platform-independent untyped constant that boxes into any as int. The testdata lines up: LUA_REGISTRYINDEX = -(INT_MAX/2 + 1000) = -1073742823, matching cl/_testc/lua/out.go, and lua_getfield -> Getfield is consistent with in.h. The pointer->value nodeInterp receiver and the removal of the unused fset field are behavior-preserving (the impl now reaches the fileset via v.ctx). The macrofunc.go node threading is sound (all builder methods take src ...ast.Node; goNode returns a true nil, so there is no typed-nil pitfall).

Worth addressing before merge: LoadExpr now runs for real (previously it was a panic("todo") stub), and both its failure modes are fatal because the auto-macro-func codegen runs in a post-unit with no recover (gogen's default HandleErr panics). See the inline comment on LoadExpr. The other items are minor (memory retention, naming, stale TODO).