Robert, I appreciate your detailed response.
Based on the code and your explanation, it seems that gc's type checking goes far more in-depth than necessary for the cases I need to analyze. In that case, it seems to me that my best course of action would be to implement my own type verification logic based on the syntax tree generated from parsing, rather than butt into the more low-level type details handled by the existing type checking logic. The code generation is simply a function of re-writing the syntax tree recursively prior to passing it to gc for conversion to a node DAG and full type-checking. This will require some tweaking of gc's main method, so that I can process a full list of parsed files prior to passing them to the gc's type checking and compilation phase. The challenge then becomes deciding what meta-data to store to allow gc to produce meaningful error messages when it encounters issues with the generated types. I appreciate your help here, I will provide an update once I've completed the work. - Jacob McCollum -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
