[flang-commits] [flang] [flang] Insert NAMELIST variables into symbol table during Pre (PR #179599)
via flang-commits
flang-commits at lists.llvm.org
Wed Feb 4 10:06:31 PST 2026
ivanrodriguez3753 wrote:
> Note that in https://flang.llvm.org/docs/Extensions.html it says "Flang processes the NAMELIST group declarations in a scope after it has resolved all of the names in that scope. This means that names that appear before their local declarations do not resolve to host associated objects and do not elicit errors about improper redeclarations of implicitly typed entities."
Could you provide some more context on why this decision was made? Or is it just the way things are? It seems like an intentional deviation from the standard, given its documented in the section "[Behavior in cases where the standard is ambiguous or indefinite](https://flang.llvm.org/docs/Extensions.html#id8)", which I would think we want to minimize, but I'm aware justifications and exceptions exist for everything. Furthermore, we can get the 'desired' behavior without violating the standard by using `implicit none`, right?
https://github.com/llvm/llvm-project/pull/179599
More information about the flang-commits
mailing list