March 2011 - asdf-devel - mailman3.common-lisp.net

[asdf-devel] source file encoding
by Anton Vodonosov 16 Apr '12

16 Apr '12

Hello. How to specify source files encoding in ASD2? I.e. what value to pass to the :external-format argument of the compile-file? In the old ASDF I did it by (let ((asdf:*compile-file-external-format* #+:clisp charset:cp1251 #+:sbcl :cp1251)) (asdf:operate 'asdf:load-op :mysystem)) But today the variable asdf:*compile-file-external-format* does not exist. Also, it of course would be better to specify the encoding in the .asd file (the system author knows the files encoding, and should care about this, but not the system user). Could you advise me how to do it? Best regards, - Anton

13 61

[asdf-devel] asdf and xcl
by Azamat S. Kalimoulline 04 Apr '11

04 Apr '11

On xcl it print me "getenv not available on your implementation". But xcl have implementation of getenv: CL-USER(1): (getenv "HOME") "/home/turtle" -- Azamat S. Kalimoulline

2 1

[asdf-devel] moving old to new asdf
by Antony 04 Apr '11

04 Apr '11

Hi I have tried to use the newer version of asdf (been using 1.x for long time). I put in (asdf:initialize-source-registry (list :source-registry (list :tree #P"C:/git/") :inherit-configuration)) (asdf:initialize-output-translations (list :output-translations (list "C:/temp/.fasl/") :inherit-configuration )) I like all of my fasls to go under appropriate subdirs of C:/temp/.fasl/ But, they are all going to c:/Users/Antony/AppData/Roaming/common-lisp/cache... I do not want to create any config files. What code would do that. -Antony

3 2

[asdf-devel] System naming conventions
by David Wang 03 Apr '11

03 Apr '11

Hello, QUESTION: 1. Could someone add a system naming convention section to the ASDF documentation, to help future users with similar problems? 2. If someone knows whether my rules-of-thumb (below) are right or wrong, let me know. EXPLANATION: I've been porting a large code base from Franz/Allegro's defsystem to ASDF 2.012. I was originally writing the .ASD files on Windows and recently tried testing the build with the same .ASD files on the Linux variant, Fedora 14, but had some problems with various missing-components. I realized the problem had something to do with the treatment of case in filenames, and ended up fixing problem by examining the ASDF code and inventing some rules-of-thumb: 1. When using Strings to name a system, make sure the filename of the ASD file, the name of the system defined in that file, and any uses of the system-name in other ASD files all have the same case. i.e.: In the file "SyStEmA.asd": (asdf:defsystem "SyStEmA" ...) In the file "SyStEmB.asd" (asdf:defsystem "SyStEmB" :depends-on ("SyStEmA")) 2. If you ever want to use a Symbol to name a system, make sure the filename of the ASD file is all lowercase. The name of the system defined in that file and any uses of the system-name in other ASD files should either be a symbol or a lowercase string. In the file "systema.asd": (asdf:defsystem #:SYSTEMA ...) In the file "systemb.asd" (asdf:defsystem "systemb" :depends-on ("systema")) I don't know if they are correct for all combinations of OSes, CL-variants, and system names, but from my limited knowledge of ASDF's code I think they'll work well. - Thanks! - David Wang P.S. Thanks for helping develop ASDF. I couldn't imagine a [Lisp] world in which it didn't exist.

2 1

[asdf-devel] (asdf:load-system :iterate) loads :iterate-pg & :iterate-tests
by Frank 02 Apr '11

02 Apr '11

Hello there, I've been using iterate for a short while. I noticed that it always loads :iterate-pg and :iterate-tests on (asdf:load-system :iterate). Do those modules have to be pulled in for just the basic :iterate part? I'm using ccl-1.6 and the quicklisp iterate version. Thanks Frank

2 1

[asdf-devel] ASDF 2.014 released
by Faré 31 Mar '11

31 Mar '11

Dear ASDF users, I just blessed ASDF 2.013.7 as 2.014. As compared to 2.013, it includes a significant quality item that makes it essential for everyone who distributes ASDF without an implementation (e.g. Quicklisp). It also includes better SCL support, cosmetic improvements, and robustification of some corner cases. Tests have been added to prevent the regressions observed in the 2.013 release. * Fixed bug preventing upgrade from 2.000 to 2.008 (test added) * Better support for URL pathnames on SCL (test added) * Robustified defsystem-depends-on * Cosmetic changes: * better more portably readable error messages. * remove forward references, again (test added) * canonicalize system names earlier (slight robustness improvement) As usual, I invite people who distribute ASDF to update the version they ship, be it with their implementations, with their software distribution, etc. Happy Happy Joy Joy [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] "Don't worry about people stealing your ideas. If your ideas are any good, you'll have to ram them down people's throats." — Howard Aiken

2 7

[asdf-devel] ASDF and XCL
by Azamat S. Kalimoulline 31 Mar '11

31 Mar '11

XCL has realization of getenv CL-USER(1): (getenv "HOME") "/home/turtle" But ASDF says that no. -- WBR Turtle//BAZON Group

2 1

[asdf-devel] CONSTANTLY in HANDLER-BIND
by Sam Steingold 29 Mar '11

29 Mar '11

Hi, MODULE-PROVIDE-ASDF used CONSTANTLY in HANDLER-BIND to define a handler. IIUC, this merely delegates error handling further, so it is more or less dead code. Is this intentional? -- Sam Steingold (http://sds.podval.org/) on CentOS release 5.5 (Final) X http://pmw.org.il http://iris.org.il http://www.PetitionOnline.com/tap12009/ http://palestinefacts.org http://jihadwatch.org http://honestreporting.com If you want it done right, you have to do it yourself

2 1

[asdf-devel] Small patch for cmucl and subdirectories
by Raymond Toy 28 Mar '11

28 Mar '11

Currently, cmucl fails the asdf test test-configuration. This is because SUBDIRECTORIES returns the wrong things. The test prints out the following before failing because foo1.asd cannot be found: subdirectories of dir2/: (#P"/Volumes/share2/src/clnet/asdf/tmp/dir2/dir3/foo1.asd" #P"/Volumes/share2/src/clnet/asdf/tmp/dir2/dir4/foo2.asd") subdirectories of dir5/: (#P"/Volumes/share2/src/clnet/asdf/tmp/dir5/dir6/foo4.asd" #P"/Volumes/share2/src/clnet/asdf/tmp/dir5/dir7/foo5.asd") 0: (PROCESS-SOURCE-REGISTRY WRAPPING-SOURCE-REGISTRY :REGISTER #<Closure Over Function "DEFUN* FLATTEN-SOURCE-REGISTRY" {48342C89}> :INHERIT ...) 0: PROCESS-SOURCE-REGISTRY returned NIL Source Registry: ((#P"/Volumes/share2/src/clnet/asdf/tmp/dir1/" #P"/Volumes/share2/src/clnet/asdf/tmp/dir2/" #P"/Volumes/share2/src/clnet/asdf/tmp/dir2/dir3/foo1.asd" #P"/Volumes/share2/src/clnet/asdf/tmp/dir2/dir4/foo2.asd" #P"/Volumes/share2/src/clnet/asdf/tmp/dir5/dir6/" #P"/Volumes/share2/src/clnet/asdf/tmp/dir8/" #P"/Volumes/share2/src/clnet/asdf/tmp/dir9/dira/" #P"/Volumes/share2/src/clnet/asdf/tmp/dir9/dira/dirc/foo8.asd")) ;; Loading #P"/Volumes/share2/src/clnet/asdf/tmp/dir1/foo0.asd". Here is a small patch for SUBDIRECTORIES that makes it return the expected subdirectories. This basically makes cmucl use the same code path as scl. If this doesn't make it into another asdf release before the cmucl snapshot due in a week or so, I'll apply this patch to cmucl's version for now. Ray diff --git a/asdf.lisp b/asdf.lisp index 1d3e0b5..4d078ae 100755 --- a/asdf.lisp +++ b/asdf.lisp @@ -3521,9 +3521,9 @@ with a different configuration, so the configuration would be re-read then." (let* ((directory (ensure-directory-pathname directory)) #-(or cormanlisp genera) (wild (merge-pathnames* - #-(or abcl allegro lispworks scl) + #-(or abcl allegro lispworks scl cmu) *wild-directory* - #+(or abcl allegro lispworks scl) "*.*" + #+(or abcl allegro lispworks scl cmu) "*.*" directory)) (dirs #-(or cormanlisp genera) @@ -3532,7 +3532,7 @@ with a different configuration, so the configuration would be re-read then." #+mcl '(:directories t)))) #+cormanlisp (cl::directory-subdirs directory) #+genera (fs:directory-list directory)) - #+(or abcl allegro genera lispworks scl) + #+(or abcl allegro genera lispworks scl cmu) (dirs (remove-if-not #+abcl #'extensions:probe-directory #+allegro #'excl:probe-directory #+lispworks #'lw:file-directory-p

1 0

[asdf-devel] Patch: Rework output formats.
by Douglas Crosher 26 Mar '11

26 Mar '11

Attached is a patch that reworks the output formats to better fit the stream width. A new macro 'compatfmt is added that strips out commands that are not supported by Genera. Regards Douglas Crosher

2 1