- Jan 09, 2013
-
-
Francois-Rene Rideau authored
New pathname functions to work around more logical-pathname issues. See discussion with rpgoldman on asdf-devel, and see CCL ticket:953. Also, tweak build to reflect recent refactoring of asdf. WIP of define-package work needed for clean upgrade (still broken).
-
- Jan 08, 2013
-
-
Francois-Rene Rideau authored
TODO: Create a define-package that can make it upgrade itself.
-
Francois-Rene Rideau authored
Add its forgotten source file test/lambda.lisp. Oops. Skip one of its tests that mysteriously breaks on ECL_bytecodes.
-
- Jan 07, 2013
-
-
Francois-Rene Rideau authored
Bug identified by avodonosov.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
- Jan 05, 2013
-
-
Francois-Rene Rideau authored
Also, fix make-pathname* to make CLISP happy. Uniformly use "backward-compat*" in comments. Rename operation-forced{,-not}-p to action-forced-p. Shuffle some functions around. Refactor operate into an :around and a regular method.
-
- Jan 03, 2013
-
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Remove debugging code from previous commit.
-
- Jan 02, 2013
-
-
Francois-Rene Rideau authored
-
- Jan 01, 2013
-
-
Francois-Rene Rideau authored
This restores the feature of do-first that was lost while refactoring ASDF to fix the timestamp propagation issue. The old asdf-bundle worked implicitly because of do-first: fasl-op DEPENDS-ON compile-op, and compile-op has a DO-FIRST on the load-op of dependencies, NOT a DEPENDS-ON, so if the compile-op is "done" (which did NOT check the timestamps of dependencies), then the dofirst is never consulted and load-op is never traversed. In our new ASDF, we want to always traverse the load-op of dependencies, so we can propagate their timestamps; however, we still want to only actually load them if they are needed. Therefore, visiting actions takes an additional flag "needed in this image", and each visited action can be planned or not planned yet, and a not planned yet action can be visited a second time to plan it after recursively visiting its dependencies with the "needed" flag set. Tests were fixed now to work now that this distinction exists.
-
- Dec 27, 2012
-
-
Francois-Rene Rideau authored
Be more selective in what warnings get quashed when upgrading ASDF. Skip bundle test on ABCL, ECL-bytecodes. Sleep more in test1.
-
Francois-Rene Rideau authored
Found this bug during ASDF upgrade tests. Add ecl_bytecodes to the upgrade test regimes. Rename it from ecl-bytecodes to work with stupid Bourne shell. Add xcl to test regime.
-
Francois-Rene Rideau authored
-
- Dec 26, 2012
-
-
Francois-Rene Rideau authored
(preparing for a future better test system.)
-
Francois-Rene Rideau authored
-
- Dec 17, 2012
-
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Change declare ignore to ignorable to make ACL happier. Remove slot for if-component-dep-fails. Update some tests and workaround a failure on ecl-bytecodes.
-
- Dec 15, 2012
-
-
Francois-Rene Rideau authored
Remove last traces of previous non-sensical :feature features.
-
Francois-Rene Rideau authored
where it ought to always have been, based on a properly cleaned up prepare-op (previously introduced in 2.26.14 as parent-load-op). This is a checkpoint of a mostly working system after major changes; it requires further cleanup and fixes: strangely, fails some tests on CCL but not SBCL. Introduce find-operation as a companion to find-component, subsuming make-sub-operation. The results of component-depends-on are processed with find-operation and resolve-component-spec. component-self-dependencies is fixed to play well with prepare-op. Also, some internal renamings and signature changes: parent-operation becomes upward-operation parent-load-op becomes prepare-op parent-source-load op becomes prepare-source-op parent-component is distinguished from module module-components becomes component-children module-components-by-name becomes component-children-by-name compute-module-components-by-name becomes compute-children-by-name circular-dependency-components becomes circular-dependency-actions component-load-dependencies becomes component-sibling-dependencies (with a stub) introducing child-component, downward-operation. refactoring visit-dependencies to include the loop in which it's always used. refactoring force and force-not processing to use a function action-override-p; force is achieved with a trivial method on operation-done-p for prepare-op, and force-not as a trivial check in visit-action. Tweak tests along the way. Tweak the documentation. Failed to either rename operation-error or introduce action-error in its superclasses: that breaks sb-grovel.
-
- Dec 09, 2012
-
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
2.26.14: Play nicer with timestamp propagation by interning dependency to parent's outer dependencies as something explicit.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
* Get rid of *stamp*, use lexical scoping everywhere * Better handle non-filesystem actions in compute-action-stamp * Have a simpler mechanism for backward compatibility with operation-done-p * Have a more robust make-sub-operation, and use it where appropriate * Test twiddling.
-
- Dec 08, 2012
-
-
Francois-Rene Rideau authored
This fixes the major issue whereby asdf's operation-done-p was failing to check the timestamps of dependencies, only checking those of the immediate operation. https://bugs.launchpad.net/asdf/+bug/1087609 This rewrite is also a vast simplification: * we rely purely on filesystem timestamps of files * we stop using get-universal-time and therefore eschew clock skews issues between processor and filesystem * we do away with the ugly do-first dependencies, previously made necessary by the way get-universal-time messed with load timestamps. * we simplify the protocol to visit components, with a better internal API that we use twice: once, recursively, in traverse, and once, shallowly, in compute-action-stamp (the new internals for operation-done-p). * make-sub-operation is drastically simplified. For the sake of backward-compatibility, the code style had to be contrived, notably introducing a special variable *stamp*, supporting dual modes for operation-done-p, and keeping the old names of many functions that would better be renamed.
-
- Sep 27, 2012
-
-
Francois-Rene Rideau authored
Fix test on old clisp.
-
- Sep 13, 2012
-
-
Francois-Rene Rideau authored
-
- Aug 15, 2012
-
-
Francois-Rene Rideau authored
-
- Jul 31, 2012
-
-
Jean-Claude Beaudoin authored
-
- Jul 27, 2012
-
-
Francois-Rene Rideau authored
* Refactor load-system to use *load-system-operation*, and change require-system and module-provide-asdf to use that. * Remove broken-fasl-support: first, cormanlisp is supposed to have fixed it; second, cormanlisp is dead; third, one can now (setf *load-system-operation* 'load-source-op) instead. * Reindent default-source-registry * Fix a test in light of some previous package frobbing.
-
- Jun 28, 2012
-
-
Francois-Rene Rideau authored
-
- Jun 12, 2012
-
-
Francois-Rene Rideau authored
Don't upgrade ccl or cmucl on older asdf's due to incompatibilities. Shouldn't matter since they already ship with more recent versions of asdf.
-
- May 15, 2012
-
-
Francois-Rene Rideau authored
following discussion with Nikodemus on asdf-devel. Also, improve the way we test for upgrade. Disable upgrade test on cmucl from asdf 2.00*. Restore -i.bak flag in bump-version: it's actually needed.
-
- Apr 23, 2012
-
-
Francois-Rene Rideau authored
-
- Apr 22, 2012
-
-
Francois-Rene Rideau authored
-
- Apr 21, 2012
-
-
Francois-Rene Rideau authored
Update dates in index.html
-
- Apr 16, 2012
-
-
Francois-Rene Rideau authored
Add a hook to autodetect file encoding. Don't use :unspecific on ecl, it breaks asdf-pathname-test.script Disable warnings testing when compiling asdf with SCL. Bug in SCL?
-
- Apr 14, 2012
-
-
Francois-Rene Rideau authored
Make logical pathname test work on disjoint files, for now. Fix test-upgrade wrt recent package tweak in testing.
-