closure-common fails to load correctly after compilation

[Mark Evenson]

When compiling the fasl for closure-common/ystream.abcl an
error in StandardClass.getClassLayout(StandardClass:205) in reading back the parameters.

Interestingly, if the system is restarted, the fasl *can* be loaded without complaint, so there seems to be some artifact of the compilation process "hanging around" in memory which causes this problem.

Unknown whether this affects the actual us of closure-common

This error existed in abcl-1.0.1 as well, so it has nothing to do with the recent work on the MOP.

0: (#<FUNCTION {50F2A460}> #<TYPE-ERROR {79D34CA}> #<FUNCTION {50F2A460}>)
  1: (APPLY #<FUNCTION {50F2A460}> (#<TYPE-ERROR {79D34CA}> #<FUNCTION {50F2A460}>))
  2: (SYSTEM::RUN-HOOK SYSTEM::*INVOKE-DEBUGGER-HOOK* #<TYPE-ERROR {79D34CA}> #<FUNCTION {50F2A460}>)
  3: (INVOKE-DEBUGGER #<TYPE-ERROR {79D34CA}>)
  4: org.armedbear.lisp.Lisp.error(Lisp.java:379)
  5: org.armedbear.lisp.Lisp.type_error(Lisp.java:420)
  6: org.armedbear.lisp.Cons.length(Cons.java:240)
  7: org.armedbear.lisp.Cons.copyToArray(Cons.java:352)
  8: org.armedbear.lisp.LispThread.pprint(LispThread.java:896)
  9: org.armedbear.lisp.LispThread.pprint(LispThread.java:916)
 10: org.armedbear.lisp.LispThread.pprint(LispThread.java:916)
 11: org.armedbear.lisp.LispThread.printBacktrace(LispThread.java:825)
 12: org.armedbear.lisp.LispThread.printBacktrace(LispThread.java:807)
 13: org.armedbear.lisp.StandardClass.getClassLayout(StandardClass.java:206)
 14: org.armedbear.lisp.Layout$6.execute(Layout.java:269)
 15: org.armedbear.lisp.Symbol.execute(Symbol.java:802)
 16: org.armedbear.lisp.LispThread.execute(LispThread.java:649)
 17: org.armedbear.lisp.clos_728.execute(clos.lisp:3470)
 18: org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:185)
 19: org.armedbear.lisp.LispThread.execute(LispThread.java:760)
 20: org.armedbear.lisp.Lisp.funcall(Lisp.java:186)
 21: org.armedbear.lisp.Primitives$pf_apply.execute(Primitives.java:2817)
 22: (SYSTEM:%MAKE-INSTANCES-OBSOLETE #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}>)
 23: (#<FUNCTION {1455BCD7}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MIXIN {7203E3B5}> #<STANDARD-CLASS GRAY-S..
 24: (APPLY #<FUNCTION {1455BCD7}> (#<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MIXIN {7203E3B5}> #<STANDARD-CLASS..
 25: (#<STANDARD-GENERIC-FUNCTION REINITIALIZE-INSTANCE {665C1EAE}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MI..
 26: (APPLY #<STANDARD-GENERIC-FUNCTION REINITIALIZE-INSTANCE {665C1EAE}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STR..
 27: (#<FUNCTION {5FE63CB6}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> RUNES::OCTET-INPUT-STREAM :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MIXIN {7203E3B5..
 28: (APPLY #<FUNCTION {5FE63CB6}> (#<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> RUNES::OCTET-INPUT-STREAM :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MIXIN {7..
 29: (#<STANDARD-GENERIC-FUNCTION ENSURE-CLASS-USING-CLASS {1A3526CF}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> RUNES::OCTET-INPUT-STREAM :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-S..
 30: (APPLY #<STANDARD-GENERIC-FUNCTION ENSURE-CLASS-USING-CLASS {1A3526CF}> #<STANDARD-CLASS RUNES::OCTET-INPUT-STREAM {7F83698F}> RUNES::OCTET-INPUT-STREAM (:DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL..
 31: (SYSTEM:ENSURE-CLASS RUNES::OCTET-INPUT-STREAM :DIRECT-SUPERCLASSES (#<STANDARD-CLASS TRIVIAL-GRAY-STREAMS:TRIVIAL-GRAY-STREAM-MIXIN {7203E3B5}> #<STANDARD-CLASS GRAY-STREAMS:FUNDAMENTAL-BINARY-INPUT-..

[ehuelsmann]

Can't reproduce on current trunk.

Hope mevenson can, otherwise we should close as fixed.

[ehuelsmann]

Assigning to mevenson so it's clear that I'm waiting for his input on the reproduction.

[ehuelsmann]

Works for both mevenson and me.

[Evenson Not Org]

CXML loads but XPath *still* borks the compiler on a hairy fasl that we need to graph to debug.
@armedbear .