- May 14, 2010
-
-
Francois-Rene Rideau authored
:exclude statements only affect the current configuration file or expression. :also-exclude adds to the exclusions rather than replace the defaults. Thanks to szergling.
-
- May 12, 2010
-
-
Francois-Rene Rideau authored
(by rtoym's request)
-
- May 11, 2010
-
-
Francois-Rene Rideau authored
Show a good example: use (require :asdf) instead or (require 'asdf).
-
Francois-Rene Rideau authored
* for configuration directories, only match files in .conf that are not unix hidden files. This notably takes care of backups. (Solves issue found by p_l.) * Before calling the provide hook, CMUCL normalizes its argument to a string (as allowed per CLHS), which means symbols become uppercase strings before they reach ASDF that would downcase symbols. I've decided to always downcase the argument provided to this hook. (Solves issue raised by rtoym.) Note that are hook itself doesn't (PROVIDE ...) anything to be pushed onto *MODULES*. Maybe we should: (a) only try find-system when (equal (string arg) (string-upcase arg)) (b) always string-downcase when calling find-system and (c) string-upcase again and provide. Not done in this commit, though. * Add #+cmu (:tree #p"modules:") to the wrapping-source-registry. (rtoym)
-
- May 07, 2010
-
-
Francois-Rene Rideau authored
* minor stylistic cleanups for CCL * refactor use of *load-truename* for more consistency.
-
- May 06, 2010
-
-
Francois-Rene Rideau authored
Also fix a bug in 1.714 (oops) whereby we fail to correctly find symbol-named components (incorrectly used string instead of coerce-name). Thanks to Alessandro Piras for beta-testing on Windows SBCL.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Cleanup a few methods on ECL (moderately tested with cl-launch).
-
- May 05, 2010
-
- May 04, 2010
-
- May 03, 2010
-
-
Francois-Rene Rideau authored
prints the complete access path to the component, not just its name. Also refactor find-component while we're at it, so we can use that path. Document flatten-tree.
-
- May 02, 2010
-
-
Francois-Rene Rideau authored
Introduce component-operation-time as an interface for POIU to hook on. declare ignorables, to hush warnings on ECL without alarming Allegro. Make run-tests return proper result code, fix make test-all. make test-all => success
-
- May 01, 2010
-
-
Francois-Rene Rideau authored
More documentation, notably for the answer directed towards vendors.
-
Francois-Rene Rideau authored
-
- Apr 30, 2010
-
- Apr 29, 2010
-
- Apr 28, 2010
-
- Apr 25, 2010
-
-
Francois-Rene Rideau authored
Fix lisp-version-string for ECL (1.705 bug - oops). Trivial cleanups and small documentation updates.
-
- Apr 21, 2010
-
-
Francois-Rene Rideau authored
* rename :system-dependencies to :defsystem-depends-on. Beware, controversy. janderson suggests we should be using defstruct-like options in the name field. * reorganize a bit the lisp-version-string, include fasl version for CCL (more useful than a PPC-only 64-bit flag that should already be redundant with architecture). * reinstate the perform-with-restarts method for compile-op removed in 1.661: it might seem redundant in nested calls, but it actually isn't.
-
- Apr 19, 2010
-
-
Francois-Rene Rideau authored
for parsing the defsystem form itself. Also, *asdf-verbose* controls the default verbosity of asdf operations (or should I have reused *verbose-out* ?)
-
Francois-Rene Rideau authored
-
- Apr 16, 2010
-
-
Francois-Rene Rideau authored
do-first operations must appear before module operations. Do it in two passes. Appending N times is evil.
- Apr 15, 2010
-
-
Francois-Rene Rideau authored
* split traverse into multiple functions. * don't append things all over traverse, use a collector function * don't insert and remove prune-op to indicate something needs be done even when the collected list is empty... use a boolean flag in all cases. * fix the age-old broken :force (system1 ... system2) feature... with a cerror! * fix the age-old broken (:feature :foo dependency) feature... with a cerror! * tighten :version syntax to only accept the keyword, not any same-named symbol. * use hash-tables for O(1) access where linear search was previously used, notably in visited-nodes and in components-by-name. * use more uniform #+ features for windows. * modify defclass to make POIU happier (hopefully). * factor probe-asd out of sysdef-source-registry-search and sysdef-central-registry-search * fix safe-file-write-date, which I had previously broken when accepting a null pathname. * have serial-depends-on only name the previous thing to depend on. Passes tests on sbcl, clisp, ecl, lispworks, allegro. Fails on gclcvs: bind stack overflow. Is that a regression? small website updates.
-
Robert P. Goldman authored
When trying to explore the behavior of ASDF2 in connection with rewriting that involves Lisp files, I was testing to verify that ASDF will notice if we lose a .lisp file. It does. This test checks that.
- Apr 14, 2010
-
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
GNUmakefile to clean *.bak.
-
- Apr 13, 2010
-
-
Francois-Rene Rideau authored
to avoid weird non-determinism when building in a different directory than expected by the asd files.
-
Francois-Rene Rideau authored
Fix bug 562183: setf methods must return proper value. (Thanks to janderson) Fix bug 562186: wild patterns need :wild version to make clisp happy. Add URL in manual.
-
- Apr 12, 2010
-
-
Francois-Rene Rideau authored
Add tests for configuration files. Also export enable-asdf-binary-locations-compatibility.
-
- Apr 09, 2010
-