Dear Robert,
I believe the plan to enable style-warnings for those using deprecated functions or missing useful metadata is great, but 1- justifies moving to version 3.2. 2- requires testing with a 3.2.0.x alpha release series that already have a version >= 3.2, so no magic problem occurs due to the change, as we had with the switch to 3.0.0. 3- is probably many months away at the current rate. 3- suggests releasing what we have now as 3.1.5.
What are your blockers to a 3.1.5 release (besides lots of testing and corresponding bug fixes)?
I really believe it would be nice to get the current bug fixes and clasp support out of the door before then. Last release 3.1.4 was in October; a lot of small fixes have happened since then, notably fixing default configuration on many systems, and things like chdir on abcl — or making asdf and uiop themselves comply with the latest metadata requirements (note that the test systems don't fulfill them.)
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org It should be a grammatical if not legal offense to ascribe thoughts, opinions and decisions to "we" without a signed power of attorney.
Faré wrote:
Dear Robert,
I believe the plan to enable style-warnings for those using deprecated functions or missing useful metadata is great, but 1- justifies moving to version 3.2. 2- requires testing with a 3.2.0.x alpha release series that already have a version >= 3.2, so no magic problem occurs due to the change, as we had with the switch to 3.0.0. 3- is probably many months away at the current rate. 3- suggests releasing what we have now as 3.1.5.
What are your blockers to a 3.1.5 release (besides lots of testing and corresponding bug fixes)?
I really believe it would be nice to get the current bug fixes and clasp support out of the door before then. Last release 3.1.4 was in October; a lot of small fixes have happened since then, notably fixing default configuration on many systems, and things like chdir on abcl — or making asdf and uiop themselves comply with the latest metadata requirements (note that the test systems don't fulfill them.)
The only blocker I have right now is testing on Windows. The tests pass for me on all the implementations I have on both Linux and Mac OS X.
As far as implementations, I don't have clasp, mkcl, Genera, Corman, or ECL, and my clisp is woefully old. But other than that I have pretty good coverage.
I'm hoping we can get some Windows tests from Dave Cooper, and if those come through clean, we can release 3.1.5.
Cheers, r
The only blocker I have right now is testing on Windows. The tests pass for me on all the implementations I have on both Linux and Mac OS X.
As far as implementations, I don't have clasp, mkcl, Genera, Corman, or ECL, and my clisp is woefully old. But other than that I have pretty good coverage.
I'm hoping we can get some Windows tests from Dave Cooper, and if those come through clean, we can release 3.1.5.
Running on Linux x64 with these implementations: ccl clisp sbcl ecl ecl_bytecodes cmucl allegro lispworks mkcl abcl allegromodern
It passes all tests, though test-program.script is flaky on mkcl due to mkcl sometimes warning about killing a thread on exit. (JCB: How do I disable this warning?)
Anton, can you run cl-test-grid with the latest ASDF 3.1.4.13 from master?
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Faré's Second Law of Dissent: I am Right, whence it follows that all who disagree with me are either (1) evil, (2) stupid or (3) crazy. (The alternatives are not mutually exclusive.) This universal law is valid for all values of "me", including "you".
Running on Linux x64 with these implementations: ccl clisp sbcl ecl ecl_bytecodes cmucl allegro lispworks mkcl abcl allegromodern
It passes all tests [...]
As for upgrade tests, I have the following failures: * clisp won't upgrade from 3.0.1 anymore, but that's OK because it comes with 3.0.2.26. Actually, trying again, it passes. Dunno why it failed at first. * cmucl won't upgrade at all; this is a known bug with its CLOS implementation since the early ASDF 3 days. Disabling upgrade. * lispworks won't load 3.0.3, which had a bug, so I disabled that, and otherwise it passes the upgrade test.
Did I mention that XCL now can't load ASDF anymore? Not that XCL ever mattered but as an experiment in portability. Goodbye, XCL!
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Imitation is the sincerest of flattery. — Charles Caleb Colton
On 6/8/15 Jun 8 -6:28 PM, Faré wrote:
Running on Linux x64 with these implementations: ccl clisp sbcl ecl ecl_bytecodes cmucl allegro lispworks mkcl abcl allegromodern
It passes all tests [...]
As for upgrade tests, I have the following failures:
- clisp won't upgrade from 3.0.1 anymore, but that's OK because it
comes with 3.0.2.26. Actually, trying again, it passes. Dunno why it failed at first.
- cmucl won't upgrade at all; this is a known bug with its CLOS
implementation since the early ASDF 3 days. Disabling upgrade.
Sounds good. What's the bug, BTW? Is it something we can FMAKUNBOUND our way around, or is it just hopeless?
Thanks for rechecking this.
I just finished my Linux checks and they all passed. I will re-run, however, now that you modified the tests one more time. I'll also run the upgrade tests overnight.
With luck, Dave's Windows tests will come back passing, and we can release tomorrow.
cheers, r
On Mon, Jun 8, 2015 at 9:00 PM, Robert Goldman rpgoldman@sift.net wrote:
On 6/8/15 Jun 8 -6:28 PM, Faré wrote:
As for upgrade tests, I have the following failures:
- cmucl won't upgrade at all; this is a known bug with its CLOS
implementation since the early ASDF 3 days. Disabling upgrade.
Sounds good. What's the bug, BTW? Is it something we can FMAKUNBOUND our way around, or is it just hopeless?
Thanks for rechecking this.
I traced that down to my own cmucl installation; it was an ancient installation with an ancient ASDF, so upgrading from the require'd one failed, which was stopping the upgrade tests. I tried with the latestest cmucl snapshot, but it's broken (from a few days ago), so I tried with the previous snapshot instead, from december, and all tests passed, including all upgrade tests (back to 3.0.1 only). The breakage is in upgrading from ASDF 2.32 or earlier, due to redefinition of the the superclasses of system.
So, everything's fine, after all (except the latest cmucl snapshot — I sent mail to the maintainer about that).
I'm restoring the cmucl upgrade tests.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org The reason truth is stranger than fiction is that fiction has to make sense.
With luck, Dave's Windows tests will come back passing, and we can release tomorrow.
Not quite. The next failure looks like it's specific to cygwin, though. At least in Allegro and CCL, when launched from a cygwin bash shell, the "PWD" environment is "/cygdrive/f/..." but this is not recognized by the Lisp -- e.g.
(probe-file (uiop:getenv "PWD")) --> NIL
Should this be considered a bug with the environment, because a Unix-style PWD is being passed into a program which is not guaranteed to understand Unix-style pathnames?
; Loading F:\cl\tests\asdf\asdf-windows\test\script-support.lisp
; Fast loading F:\cl\tests\asdf\asdf-windows\build\fasls\alisp\asdf.fasl
Configuring ASDF
Enabling debugging
Being a bit verbose
Comparing directories
ASDF-TEST:*TEST-DIRECTORY* and
(:PATHNAME-DIRECTORY-PATHNAME (NTH-VALUE 2 (:LOCATE-SYSTEM :TEST-ASDF))) both evaluate to same path:
#P"F:\cl\tests\asdf\asdf-windows\test\"
Frob packages
Running test-encodings.script with acl-9.0-win-x86
; Loading F:\cl\tests\asdf\asdf-windows\test\test-encodings.script
Test TEST-ENCODING-EXPLICIT-U8: should be UTF-8
; Registering #<SYSTEM "test-encoding-explicit-u8">
; Loading F:\cl\tests\asdf\asdf-windows\test\lambda.lisp
TEST ABORTED: illegal namestring: "/cygdrive/f/cl/tests/asdf/asdf-windows/:/cygdrive/f/cl/tests/asdf/asdf-windows/uiop/:/cygdrive/f/cl/tests/asdf/asdf-windows/ext//:" [6]
Evaluation stack:
->(TPL::ZOOM-COMMAND :FROM-READ-EVAL-PRINT-LOOP NIL :COUNT 69 :ALL T)
(SYS::..RUNTIME-OPERATION "applyn" . #1=(:UNKNOWN-ARGS))
(TPL:DO-COMMAND
"zoom"
:FROM-READ-EVAL-PRINT-LOOP
NIL
:COUNT
69
:ALL
T)
(ASDF/IMAGE:RAW-PRINT-BACKTRACE
:STREAM
#2=#<EXCL:TERMINAL-SIMPLE-STREAM
[initial terminal io] fd 0/1
@
#x200fadea>
:COUNT
69
:CONDITION
#3=#<PARSE-ERROR @ #x205d490a>)
(SYS::..RUNTIME-OPERATION #4="lisp_apply" . #1#)
((:INTERNAL ASDF/IMAGE:PRINT-BACKTRACE 0))
(ASDF/STREAM:CALL-WITH-SAFE-IO-SYNTAX
#<Closure (:INTERNAL ASDF/IMAGE:PRINT-BACKTRACE 0) @ #x205d491a>)
(ASDF/IMAGE:PRINT-BACKTRACE :STREAM #2# :COUNT 69 :CONDITION #3#)
(ASDF/IMAGE:PRINT-CONDITION-BACKTRACE #3# :STREAM #2# :COUNT 69)
(SYS::..RUNTIME-OPERATION #4# . #1#)
[... EXCL::EVAL-AS-PROGN ]
(BLOCK
ASDF-TEST:ACALL
(APPLY
(APPLY
'ASDF-TEST:ASYM
(IF
(CONSP ASDF-TEST::NAME)
ASDF-TEST::NAME
(LIST ASDF-TEST::NAME)))
ASDF-TEST::ARGS))
[... EXCL::%EVAL ]
(ASDF-TEST:ACALL
:PRINT-CONDITION-BACKTRACE
#3#
:COUNT
69
:STREAM
#2#)
(SYS::..RUNTIME-OPERATION
#8="_comp_to_interp"
:PRINT-CONDITION-BACKTRACE
#3#
:COUNT
69
:STREAM
#2#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
[... EXCL::EVAL-AS-PROGN ]
(LET*
. #6=(((#5=#:G245
(CONS
(LOAD-TIME-VALUE EXCL::.IGNORE-ERRORS-1.)
(EXCL::FAST EXCL::*HANDLER-CLUSTERS*)))
(EXCL::*HANDLER-CLUSTERS* #5#))
(DECLARE (DYNAMIC-EXTENT #5# EXCL::*HANDLER-CLUSTERS*))
(PROGN
. #7=((ASDF-TEST:ACALL
:PRINT-CONDITION-BACKTRACE
ASDF-TEST::C
:COUNT
69
:STREAM
*ERROR-OUTPUT*)))))
[... EXCL::EVAL-AS-PROGN ]
(CATCH 'EXCL::IGNORE-ERRORS-1 (LET* . #6#))
[... EXCL::%EVAL ]
(EXCL::EVAL-AS-PROGN
((IGNORE-ERRORS . #7#) (ASDF-TEST:LEAVE-TEST "Script failed" 1)))
[... EXCL::%EVAL ]
((:INTERNAL ASDF-TEST::CALL-WITH-TEST) #3#)
(SYS::..RUNTIME-OPERATION #8# #3#)
(SIGNAL #3#)
(ERROR
PARSE-ERROR
:FORMAT-CONTROL
#9="illegal namestring: ~s [~d]"
:FORMAT-ARGUMENTS
#10=("/cygdrive/f/cl/tests/asdf/asdf-windows/:/cygdrive/f/cl/tests/asdf/asdf-windows/uiop/:/cygdrive/f/cl/tests/asdf/asdf-windows/ext//:"
6))
(EXCL::.PARSE-ERROR #9# . #10#)
((FLET EXCL::PARSE-DOS-NAMESTRING EXCL::FAILURE) 6)
(EXCL::PARSE-DOS-NAMESTRINGu
#11="/cygdrive/f/cl/tests/asdf/asdf-windows/:/cygdrive/f/cl/tests/asdf/asdf-windows/uiop/:/cygdrive/f/cl/tests/asdf/asdf-windows/ext//:"
NIL
#P""
0
NIL
NIL)
(PARSE-NAMESTRING #11#)
(PATHNAME #11#)
(ASDF/PATHNAME:ABSOLUTE-PATHNAME-P #11#)
((FLET
ASDF/SOURCE-REGISTRY:PARSE-SOURCE-REGISTRY-STRING
ASDF/SOURCE-REGISTRY::CHECK)
#11#)
(ASDF/SOURCE-REGISTRY:PARSE-SOURCE-REGISTRY-STRING
#12="/cygdrive/f/cl/tests/asdf/asdf-windows/:/cygdrive/f/cl/tests/asdf/asdf-windows/uiop/:/cygdrive/f/cl/tests/asdf/asdf-windows/ext//:")
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #17=(STRING))
#12#
:INHERIT
#13=(ASDF/SOURCE-REGISTRY:USER-SOURCE-REGISTRY
ASDF/SOURCE-REGISTRY:USER-SOURCE-REGISTRY-DIRECTORY
ASDF/SOURCE-REGISTRY:DEFAULT-USER-SOURCE-REGISTRY
ASDF/SOURCE-REGISTRY:SYSTEM-SOURCE-REGISTRY
ASDF/SOURCE-REGISTRY:SYSTEM-SOURCE-REGISTRY-DIRECTORY
ASDF/SOURCE-REGISTRY:DEFAULT-SYSTEM-SOURCE-REGISTRY)
:REGISTER
#14=#<Closure (:INTERNAL
ASDF/SOURCE-REGISTRY:FLATTEN-SOURCE-REGISTRY
0)
@
#x205d554a>)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
#12#
:INHERIT
#13#
:REGISTER
#14#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #23=(SYMBOL))
ASDF/SOURCE-REGISTRY:ENVIRONMENT-SOURCE-REGISTRY
:INHERIT
#13#
:REGISTER
#14#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
ASDF/SOURCE-REGISTRY:ENVIRONMENT-SOURCE-REGISTRY
:REGISTER
#14#
:INHERIT
#13#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
(ASDF/SOURCE-REGISTRY::INHERIT-SOURCE-REGISTRY
#15=(ASDF/SOURCE-REGISTRY:ENVIRONMENT-SOURCE-REGISTRY . #13#)
:REGISTER
#14#)
(ASDF/SOURCE-REGISTRY::PROCESS-SOURCE-REGISTRY-DIRECTIVE
:INHERIT-CONFIGURATION
:INHERIT
#15#
:REGISTER
#14#)
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #21=(CONS))
#16=(:SOURCE-REGISTRY :INHERIT-CONFIGURATION)
:INHERIT
#15#
:REGISTER
#14#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
#16#
:INHERIT
#15#
:REGISTER
#14#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #17#)
#18=""
:INHERIT
#15#
:REGISTER
#14#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
#18#
:REGISTER
#14#
:INHERIT
#15#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
(EXCL::CACHING-MISS
#24=#<STANDARD-GENERIC-FUNCTION
ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY>
#19=(#18# :REGISTER #14# :INHERIT #15#)
#(0
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
#25=#(32787053 T NIL NIL #<BUILT-IN-CLASS CONS> 0 NIL)
#26=#<Closure (:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
@
#x204bf17a>
NIL
NIL
NIL
NIL
NIL
NIL
#(32124820 T NIL NIL #<BUILT-IN-CLASS SYMBOL> 0 NIL)
#<Closure (:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
@
#x205d5712>
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL))
((:INTERNAL EXCL::UPDATE-TO-CACHING-DFUN 0) . #19#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
(ASDF/SOURCE-REGISTRY::INHERIT-SOURCE-REGISTRY
#20=(#18# . #15#)
:REGISTER
#14#)
(ASDF/SOURCE-REGISTRY::PROCESS-SOURCE-REGISTRY-DIRECTIVE
:INHERIT-CONFIGURATION
:INHERIT
#20#
:REGISTER
#14#)
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #21#)
#22=(:SOURCE-REGISTRY :INHERIT-CONFIGURATION)
:INHERIT
#20#
:REGISTER
#14#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
#22#
:INHERIT
#20#
:REGISTER
#14#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
((METHOD ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY #23#)
ASDF/SOURCE-REGISTRY:WRAPPING-SOURCE-REGISTRY
:INHERIT
#20#
:REGISTER
#14#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
((:INTERNAL (:EFFECTIVE-METHOD 1 T T NIL NIL) 0)
ASDF/SOURCE-REGISTRY:WRAPPING-SOURCE-REGISTRY
:REGISTER
#14#
:INHERIT
#20#)
(SYS::..RUNTIME-OPERATION #4# . #1#)
(EXCL::CACHING-MISS
#24#
#27=(ASDF/SOURCE-REGISTRY:WRAPPING-SOURCE-REGISTRY
:REGISTER
#14#
:INHERIT
#20#)
#(0
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
#25#
#26#
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL
NIL))
((:INTERNAL EXCL::UPDATE-TO-CACHING-DFUN 0) . #27#)
[... ASDF/SOURCE-REGISTRY:PROCESS-SOURCE-REGISTRY ]
(ASDF/SOURCE-REGISTRY::INHERIT-SOURCE-REGISTRY
(ASDF/SOURCE-REGISTRY:WRAPPING-SOURCE-REGISTRY . #20#)
:REGISTER
#14#)
(ASDF/SOURCE-REGISTRY:FLATTEN-SOURCE-REGISTRY #18#)
(ASDF/SOURCE-REGISTRY:COMPUTE-SOURCE-REGISTRY #18#)
(ASDF/FIND-SYSTEM:INITIALIZE-SOURCE-REGISTRY #18#)
[... EXCL::%EVAL ]
(EVAL
(UNLESS
(ASDF/SYSTEM:FIND-SYSTEM :ASDF-ENCODINGS NIL)
(ASDF/FIND-SYSTEM:INITIALIZE-SOURCE-REGISTRY #18#)))
(EXCL::SLOLOAD
#28=#<EXCL:FILE-SIMPLE-STREAM
#29=#P"F:\cl\tests\asdf\asdf-windows\test\test-encodings.script" for input pos 2870
@
#x204d5a32>)
(EXCL::LOAD-FROM-STREAM-OR-BUNDLE #28# #29# NIL T 2 NIL NIL)
(EXCL::LOAD-2
#30="test-encodings.script"
T
#31=(#P""
#P(:TYPE "fasl")
#P(:TYPE "cl")
#P(:TYPE "lisp")
#P(:TYPE "lsp"))
:ERROR
NIL
NIL
NIL
NIL
NIL
NIL
NIL
:DEFAULT
NIL
NIL)
(EXCL::LOAD-1
#30#
T
#31#
:ERROR
NIL
NIL
NIL
NIL
NIL
NIL
NIL
:DEFAULT
NIL
NIL)
(LOAD #30#)
[... EXCL::EVAL-AS-PROGN ]
(LET
((ASDF-TEST::NAME (STRING ASDF-TEST::FILE)))
(FORMAT
T
"Running ~A with ~A~%"
ASDF-TEST::NAME
(ASDF-TEST:ACALL :IMPLEMENTATION-IDENTIFIER))
(LOAD ASDF-TEST::NAME))
[... EXCL::%EVAL ]
((:INTERNAL ASDF-TEST:RUN-TEST-SCRIPT))
(SYS::..RUNTIME-OPERATION #8#)
... more older frames ...
Above backtrace due to this condition:
illegal namestring: "/cygdrive/f/cl/tests/asdf/asdf-windows/:/cygdrive/f/cl/tests/asdf/asdf-windows/uiop/:/cygdrive/f/cl/tests/asdf/asdf-windows/ext//:" [6]
Script failed
<The Lisp process exited with status 1 (0x1)>
On 6/9/15 Jun 9 -11:19 AM, Dave Cooper wrote:
With luck, Dave's Windows tests will come back passing, and we can release tomorrow.
Not quite. The next failure looks like it's specific to cygwin, though. At least in Allegro and CCL, when launched from a cygwin bash shell, the "PWD" environment is "/cygdrive/f/..." but this is not recognized by the Lisp -- e.g.
cygwin is a mess. It's an operating system, but no it's not an operating system.
My best guess is that the Lisps would have to be told that they are running under cygwin in some sense, so that they would know to use the cygwin equivalents to the standard posix commands, instead of using the real operating system, which is Windows.
The only thing I can think of would be if we could find something in the environment that would indicate that we're in cygwin context. I don't know what that would be. Does cygwin set some standard environment variable?
Even then, I'm not sure how the lisp implementation could figure out how to find the right cygwin programs....
Considering that this is asdf choking on the CL_SOURCE_REGISTRY defined by the Makefile, which itself can be overridden using ASDF_DEVEL_SOURCE_REGISTRY, can you export the latter environment variable using Windows-style paths and ; as separator? Note that to avoid having to count backslashes in presence of an indeterminate number of levels of quoting, you may want to keep forward slashes.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Nothing is so permanent as a temporary government program. - Milton Friedman
On Tue, Jun 9, 2015 at 12:29 PM, Robert Goldman rpgoldman@sift.net wrote:
On 6/9/15 Jun 9 -11:19 AM, Dave Cooper wrote:
With luck, Dave's Windows tests will come back passing, and we can release tomorrow.
Not quite. The next failure looks like it's specific to cygwin, though. At least in Allegro and CCL, when launched from a cygwin bash shell, the "PWD" environment is "/cygdrive/f/..." but this is not recognized by the Lisp -- e.g.
cygwin is a mess. It's an operating system, but no it's not an operating system.
My best guess is that the Lisps would have to be told that they are running under cygwin in some sense, so that they would know to use the cygwin equivalents to the standard posix commands, instead of using the real operating system, which is Windows.
The only thing I can think of would be if we could find something in the environment that would indicate that we're in cygwin context. I don't know what that would be. Does cygwin set some standard environment variable?
Even then, I'm not sure how the lisp implementation could figure out how to find the right cygwin programs....
On Tuesday, June 9, 2015, Faré fahree@gmail.com wrote:
Considering that this is asdf choking on the CL_SOURCE_REGISTRY defined by the Makefile, which itself can be overridden using ASDF_DEVEL_SOURCE_REGISTRY, can you export the latter environment variable using Windows-style paths and ; as separator?
Where should I try to do such overriding? In my top level "run" script?
Note that to avoid having to count backslashes in presence of an indeterminate number of levels of quoting, you may want to keep forward slashes.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Nothing is so permanent as a temporary government program. - Milton Friedman