#1651: Problems with expansions in parameter files ---------------------+------------------------------------------------------ Reporter: hinder | Owner: sbrandt Type: defect | Status: accepted Priority: major | Milestone: Component: Cactus | Version: development version Resolution: | Keywords: ---------------------+------------------------------------------------------
Comment (by rhaas):
I am fairly sure do not understand setonce.patch. I am not sure I understand in which respect setting a variable only once is different from disallowing parameters to change after they are being read. setonce.patch seems much more complex that write_after_read.
If both achieve the same effect (up to the unary-ops-change-their-argument issue), I'd prefer write_after_read.patch. It seems to me as if cyclical references can be detected if a parameter is marked as being "written" right when one begins to modify it and before the RHS of the assignment is actually evaluated.