- Jan 08, 2013
-
-
Francois-Rene Rideau authored
ECL output-files semantics has changed, so unintern the old one rather than conflict. Rename *compile-op-compile-file-function* to just *compile-file-function*. Factor all the lisp build perform's through perform-lisp-FOO, so they can be shared with concatenated targets that are not cl-source-file without fighting default methods and class inheritance.
-
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
The TRAVERSE issue makes it easier for user code to override TRAVERSE: don't include FORCE FORCE-NOT VERBOSE in mandatory keyword arguments for all methods. A method is notably provided by buildapp. The bug in VISIT-DEPENDENCIES was causing IF-FEATURE'ed out components to flag whoever depended on them as requiring to be rebuilt. This notably caused a legacy SB-ROTATE-BYTE to always want to rebuild. The dependency visit is now predicated on VALID-ACTION-P, which requires VISIT-DEPENDENCIES to take an additional PLAN argument.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
- Jan 06, 2013
-
-
Francois-Rene Rideau authored
Also, use .all-systems rather than .system-and-dependencies as suffix for monolithic bundle ops.
-
- Jan 05, 2013
-
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Frob CLOS methods so &allow-other-keys. This makes POIU happy, and probably GBBopen, too.
-
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
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Also count planned actions; and count planned background actions for the sake of POIU.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Also, remove previously commented-out utilities from exported symbols. Reinstate action-already-done-p for the sake of POIU.
-
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
Introduce a PLAN object for dependency traversal; have component visit functions use it instead of the ancestor operation. Introduce an explicit ACTION-STATUS class. Introduce a PLAN-RECORD-DEPENDENCY hook for POIU. Also hush warnings around upgrade-only methods. More uniformly redirect output during testing.
-
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 28, 2012
-
-
Francois-Rene Rideau authored
If/when XCL ever resurrects, it's time to further maintain its support.
-
- 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
-
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
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
Also added allegromodern to the testing regime, after I found the recipe to create mlisp: http://www.franz.com/support/faq/#s3q7
-
- Dec 26, 2012
-
-
Francois-Rene Rideau authored
Fixes lp#1093796 (thanks to stassats for reporting).
-
Francois-Rene Rideau authored
(preparing for a future better test system.)
-
Francois-Rene Rideau authored
Thanks to previously added test for noticing the breakage.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
-
- Dec 21, 2012
-
-
Francois-Rene Rideau authored
into component, not child-component.
-
- Dec 19, 2012
-
-
Francois-Rene Rideau authored
This assumption is notably made by tens of systems that use the old ASDF1 method of specifying the pathname type for components: (defmethod source-file-type ((c foo-file) (s module)) "foo") Sigh. My bad for recommending people use module and not system, here.
-
Francois-Rene Rideau authored
Too many clients have grown to depend on it already.
-
Francois-Rene Rideau authored
-
Francois-Rene Rideau authored
so we can upgrade from ASDF 2.018-2.26.
-