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

[asdf-devel] Please test: merged ECL support, created asdf.asd
by Faré 01 Dec '09

01 Dec '09

Dear ASDF developers, I've committed to *a new repo* a few ASDF hacks that I wrote. I'd like some of you to review and/or test it. I'd like to get it reviewed before I push those changes to the official repo. git clone http://common-lisp.net/project/xcvb/git/asdf.git cd asdf git diff -u -a -b 1f7690b49b73b499ff2bd2ad77ec23c273941814 60165185516e699d0daebd0b653e61b83e72d4e6 The main changes are: * Merged the ECL patch. Had to remove a #+ecl in a slot declaration, because the slot is used afterwards without #+ecl. * Limited support to upgrade from a previous installation (if you have old installations, please try to upgrade!). * Notably created an asdf.asd so you can try to (require :asdf) (pushnew "/path/to/asdf/" asdf:*central-registry*) (asdf:oos 'asdf:load-op :asdf) * Some whitespace normalization (hence use -b in diff above). There are more things to do before we can declare asdf fully upgradable, but that's a first step. When all is done, there will be less pressure to synchronize with CL implementers to upgrade before a new feature can be relied upon. See https://bugs.launchpad.net/asdf/+bug/485687 Known issues (juanjo, help appreciated): $ ecl -eval "(require :asdf)" -load "asdf.lisp" ==> works $ ecl -eval "(require :asdf)" -eval "(asdf:oos 'asdf:load-op :asdf)" ==> fails ;;; Finished compiling /home/fare/cl/asdf/asdf.lisp. An error occurred during initialization: NIL is not of type HASH-TABLE.. Fails with both ECL 9.10.2 from debian and ECL 9.11.1 from a semi-recent clbuild. [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] To make an apple pie from scratch, you must first create the universe. -- Carl Sagan

3 6

Re: [asdf-devel] Bizarre consequences of FIND-SYSTEM in ASDF.
by Robert Goldman 30 Nov '09

30 Nov '09

dherring(a)tentpost.com wrote: > Robert Goldman wrote: >> Daniel Herring wrote: >>> On Sun, 29 Nov 2009, Robert Goldman wrote: .... > > There would be several benefits if FIND-SYSTEM were to simply PROBE-FILE > for a system definition in each of the configured paths (e.g. > path1/system.asd, path2/system.asd) until one is found. Better > efficiency, no pollution from unrelated packages, etc. I still don't get it. I'm looking at asdf.lisp, and it's basically doing what you say: 1. call system-definition-pathname to find the system definition file. 2. if there's a previously-loaded system definition, check to see if it's stale. If so, reload. 3. If there's no previously-loaded system definition, load 4. return the system definition. In stock (upstream) asdf, there's no loading in sysdef-central-registry-search, so I still don't see what's going wrong here. Is this SBCL-specific? As for the fact that people can put code like (asdf:oos 'asdf:load-op <system I need at read time>) in their .asd files, that seems to me to be a feature, not a bug. r

1 0

[asdf-devel] Output Stream for ASDF
by thomas.bartscher＠widder-software.com 30 Nov '09

30 Nov '09

Why are warnings of asdf put into *standard-output*? Wouldn't it be easier to put those into something like *asdf-warnigs*? This way users of asdf could redirect those wherever they want.

3 2

[asdf-devel] duplicate-names for module and file with same name?
by Jianshi Huang 27 Nov '09

27 Nov '09

Hi, Currently in parse-component-form function, duplication of names are checked but modules and files are put to the same category. Is there any particular reason for this restriction? e.g. I can't have both foo.lisp foo/ in the same folder since (:module :foo) and (:file "foo") will conflict in names. Cheers, Jianshi Huang

2 2

[asdf-devel] ASDF::COERCE-NAME string-downcases symbols, but lets strings untouched
by Tobias C. Rittweiler 21 Nov '09

21 Nov '09

(defun coerce-name (name) (typecase name (component (component-name name)) (symbol (string-downcase (symbol-name name))) (string name) (t (sysdef-error "~@<invalid component designator ~A~@:>" name)))) I first thought this was some bad kludge to support modern-mode. But vc-annotate told me that was introduced by Nikodemus. What is the reason that symbols are downcased but strings not? It makes (FIND-SYSTEM :FOO) be different from (FIND-SYMBOL (SYMBOL-NAME :FOO)) which is just weird. -T.

3 3

[asdf-devel] broken texinfo link
by nuntius＠strider2.example.org 20 Nov '09

20 Nov '09

asdf.texinfo still refers to Kent's old site at http://world.std.com/~pitman/Papers/Large-Systems.html It should be updated to http://nhplace.com/kent/Papers/Large-Systems.html - Daniel

2 1

[asdf-devel] update defsystem grammar for method forms
by Robert P. Goldman 20 Nov '09

20 Nov '09

I was trying to write some in-line method definitions in defsystem and found that the system grammar supplied in the manual did not specify the grammar for this. The grammar is actually a little tricky, and doesn't support specifying primary methods, so I think an extension to the docs is warranted. Best, r

1 1

[asdf-devel] Patch for buggy URL in asdf.texinfo
by Robert P. Goldman 20 Nov '09

20 Nov '09

Nuntius provided a patch for an out-of-date pointer to Kent Pitman's paper in asdf.texinfo. Here's a patch for that.

1 1

[asdf-devel] First steps for ECL support
by Juan Jose Garcia-Ripoll 17 Nov '09

17 Nov '09

I have split the change into two, as I normally do in the ECL source tree itself. The first change only pertains asdf.lisp and it adds a slot, FLAGS, to COMPILE-OP. These are additional flags for COMPILE-FILE. It is used by ECL because we want to produce two files per source file: an object file and the FASL itself. These files can be separately used for building larger things (shared libraries, programs...) or for immediate loading. The second set of changes is isolated in a single file, asdf-ecl.lisp, which is more complex. This file contains a number of extensions, which may or may not be adapted to other platforms (I believe it is possible, at least for some things). What it implements: - COMPILED-FILE, a new component for distribution of pre-compiled files. I use it to build pre-packaged versions of RT, the compiler and other things that now can be included as dependencies. - A number of operations for creating shared libraries, programs, etc. These operations use traverse in a funny way to find out the order in which files should be combined. I call them BUNDLEs because the result is a single binary file that combines many lisp sources, compiled. It has an important advantage for ECL, apart from distribution, for it takes less resources to load a single file. - A function, MAKE-BUILD, which uses those operations hiding the details. - A new operation, LOAD-FASL-OP, which uses those "bundles". - A hook into REQUIRE using LOAD-FASL-OP. What is implemented more or less follows this http://ecls.sourceforge.net/new-manual/ch16.html Comments are welcome. Juanjo -- Instituto de Física Fundamental, CSIC c/ Serrano, 113b, Madrid 28006 (Spain) http://juanjose.garciaripoll.googlepages.com

2 1

[asdf-devel] Two questions
by Nick Levine 17 Nov '09

17 Nov '09

1. Problems with the :feature feature. (asdf:defsystem foo :components ((:file "pkg") (:file "code") (:file "foo" :depends-on ((:feature :frob))))) If :frob is not on *features*, load-system gives me: Error: Bad dependency (FEATURE FROB). Dependencies must be (:version <version>), (:feature <feature>), or a name If :frob is a *feature* I get a different error: Error: Error component :FROB not found, required by #<CL-SOURCE-FILE "foo" #x8B7A3E6> Did I misunderstand something? 2. Suppose I would like some code to run at compile-time / load-time to determine where the fasls should live (e.g. hypothetically, in a subdirectory named after a timestamp, or according to what patches I have loaded). How do I express this? (I'm using version 1.363) Thanks. - nick

8 16