From senatorzergling at gmail.com Sun Aug 8 23:30:02 2010 From: senatorzergling at gmail.com (Theam Yong Chew) Date: Mon, 9 Aug 2010 11:30:02 +1200 Subject: [xcvb-devel] Bug/typo Message-ID: Howdy Fare, I might have spotted a typo below (from xcvb .532), where the format string should probably be (setf *features*...) with an s. Yong --- a/driver-commands.lisp +++ b/driver-commands.lisp @@ -89,7 +89,13 @@ will create the desired content. An atomic rename() will have to be performed af (lisp-invocation-arglist :image-path (if image (fullname-enough-namestring env image) *lisp-image-pathname*) :load (mapcar/ #'fullname-enough-namestring env load) - :eval eval))) + :eval (if (or *target-added-features* *target-suppressed-features*) + (format nil "(progn~ + ~{~#[~;(pushnew ~S *features*)~:;(dolist(x'(~@{~S~^ ~}))(pushnew x *features*))~]~}~ + ~@[(setf *feature*(remove~{~#[~; ~S~:;-if(lambda(x)(member x'(~@{~S~^ ~})))~]~} *features*))~]~ + ~A)" + *target-added-features* *target-suppressed-features* eval) + eval)))) (defun compile-file-directly-shell-token (env name &key cfasl) (quit-form From fahree at gmail.com Mon Aug 9 00:26:41 2010 From: fahree at gmail.com (=?ISO-8859-1?Q?Far=E9?=) Date: Sun, 8 Aug 2010 20:26:41 -0400 Subject: [xcvb-devel] Bug/typo In-Reply-To: References: Message-ID: Dear Theam Yong Chew, thanks a lot for spotting the typo. I'm (happily) surprised that anyone is looking at that code. I thought I was the only user as well as the only developer. Are you using XCVB for anything, or considering using it? Are you interested in developing it? [ Fran?ois-Ren? ?VB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] There is never a better measure of what a person is than what he does when he's absolutely free to choose." ? William M. Bulger On 8 August 2010 19:30, Theam Yong Chew wrote: > Howdy Fare, > > I might have spotted a typo below (from xcvb .532), where > the format string should probably be (setf *features*...) with an s. > > Yong > > > > > --- a/driver-commands.lisp > +++ b/driver-commands.lisp > @@ -89,7 +89,13 @@ will create the desired content. An atomic rename() > will have to be performed af > > ? ? (lisp-invocation-arglist > ? ? ?:image-path (if image (fullname-enough-namestring env image) > *lisp-image-pathname*) > ? ? ?:load (mapcar/ #'fullname-enough-namestring env load) > - ? ? :eval eval))) > + ? ? :eval (if (or *target-added-features* *target-suppressed-features*) > + ? ? ? ? ? ? ? (format nil "(progn~ > + ?~{~#[~;(pushnew ~S *features*)~:;(dolist(x'(~@{~S~^ ~}))(pushnew x > *features*))~]~}~ > + ?~@[(setf *feature*(remove~{~#[~; ~S~:;-if(lambda(x)(member > x'(~@{~S~^ ~})))~]~} *features*))~]~ > + ?~A)" > + ? ? ? ? ? ? ? ? ? ? ? *target-added-features* > *target-suppressed-features* eval) > + ? ? ? ? ? ? ? eval)))) > > ?(defun compile-file-directly-shell-token (env name &key cfasl) > ? (quit-form From senatorzergling at gmail.com Thu Aug 12 06:22:44 2010 From: senatorzergling at gmail.com (Theam Yong Chew) Date: Thu, 12 Aug 2010 18:22:44 +1200 Subject: [xcvb-devel] Bug/typo In-Reply-To: References: Message-ID: On 8/9/10, Far? wrote: > Dear Theam Yong Chew, > > thanks a lot for spotting the typo. I'm (happily) surprised that > anyone is looking at that code. I thought I was the only user as well > as the only developer. I tried xcvb a while back, it was just some tyre-kicking, and not much came of it because I didn't have specific use cases in mind. Since then, I have been subscribed to the source repository's RSS feed (who uses RSS readers for blogs anyway?). I didn't run your code, but was merely examining your interesting format string, which was when I noticed the typo. I still don't understand what ~#[ ... means! > Are you using XCVB for anything, or considering using it? Are you > interested in developing it? I haven't been using it, as I don't have any specific use case. asdf has been good enough for me, even more so with your recent improvements. I have been mulling over efficient cross-compilation (specifically, for something like parenscript), since I'm using a simple lisp file for full compilations. Each edit requires a complete, fresh recompile. This is starting to seem slow to me. I'm not sure if xcvb is useful for such work though... Yong.