On Tue, Jan 29, 2013 at 9:20 AM, Juan Jose Garcia-Ripoll <juanjose.garciarip...@gmail.com> wrote: > On Sat, Jan 26, 2013 at 6:53 PM, Faré <fah...@gmail.com> wrote: >> >> I've been adding support for ASDF to save deferred warnings >> (typically, forward function references) in a file, >> so that if you do an incremental compilation, warnings >> from previously compiled files are still correctly handled >> at the end of the virtual with-compilation-unit. > > I still do not get it completely. Could you elaborate a bit?
I should probably add some manual entry about that. A feature that has been requested at least as far back as 2004 is to have some kind of with-compilation-unit around every system's compilation. For that, I want to properly handle forward reference warnings from files that are *not* being (re)compiled in the current session but have been in a previous one. And so on supported implementations (currently CCL, SBCL), I am saving the deferred warnings about such forward references. Could you help me do that with ECL? The three API functions I need are as follows: (defun reify-deferred-warnings () "return a portable S-expression, portably readable and writeable in any Common Lisp implementation using READ within a WITH-SAFE-IO-SYNTAX, that represents the warnings currently deferred by WITH-COMPILATION-UNIT. One of three functions required for deferred-warnings support in ASDF." (defun unreify-deferred-warnings (reified-deferred-warnings) "given a S-expression created by REIFY-DEFERRED-WARNINGS, reinstantiate the corresponding deferred warnings as to be handled at the end of the current WITH-COMPILATION-UNIT. Handle any warning that has been resolved already, such as an undefined function that has been defined since. One of three functions required for deferred-warnings support in ASDF." (defun reset-deferred-warnings () "Reset the set of deferred warnings to be handled at the end of the current WITH-COMPILATION-UNIT. One of three functions required for deferred-warnings support in ASDF." If you insist, I can loosen the restrictions so the reified form can be implementation-specific and saved with read-eval t instead of a portable SEXP subset saved with read-eval nil. Thank you very much for your prompt response. —♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Memory is like an orgasm... it's better when you dont have to fake it. — Seymour Cray ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnnow-d2d _______________________________________________ Ecls-list mailing list Ecls-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ecls-list