1. 21 Jul, 2017 1 commit
    • Daniel Kochmański's avatar
      ecl: create-image: don't call internal function compute-init-name · 0acc3687
      Daniel Kochmański authored
      It is wrong to call this function. If :init-name is not supplied, or
      NIL ECL will compute the init name itself (it is not a new
      mechanism).
      
      Enabling specifying init-name makes it possible to build static and
      shared libraries with known initialization function (important, when
      CL module is linked with bigger C/C++ project outside ASDF).
      
      If :init-name is NIL, it will default to compute-init-name
      internally. Calling explicitly is not guaranteed to work in the
      future (internal mechanism).
      0acc3687
  2. 17 Feb, 2017 3 commits
  3. 06 Oct, 2016 1 commit
    • Elias Pipping's avatar
      Clean up use of *features* · 7e361dc2
      Elias Pipping authored
      In particular:
       - clasp does not define :ecl
       - use :os-unix instead of :unix
       - use :os-windows instead of :mswindows or :windows
       - use :os-macosx instead of :darwin
       - use :lispworks7+ instead of :lispworks7
      7e361dc2
  4. 25 Sep, 2016 2 commits
  5. 16 Sep, 2016 1 commit
  6. 02 Sep, 2016 1 commit
  7. 01 Aug, 2016 1 commit
  8. 25 Jul, 2016 1 commit
    • Robert Goldman's avatar
      Redefined *FATAL-CONDITIONS* per 1605650. · 9fd59ed8
      Robert Goldman authored
      See https://bugs.launchpad.net/asdf/+bug/1605650
      
      According to the spec, SERIOUS-CONDITIONs are "All conditions serious
      enough to require interactive intervention if not handled should inherit
      from the type serious-condition."  This is almost exactly the definition
      of *FATAL-CONDITIONS*, so I substituted SERIOUS-CONDITION for ERROR
      here.
      
      Added exceptions to the set of *FATAL-CONDITIONS* in order to fix test
      condition handling for CCL.  The exceptions permit us to have a class of
      conditions (e.g., SERIOUS-CONDITION) that has a subclass that is not a
      FATAL-CONDITION, but that does not have an obvious more specific
      alternative.
      
      The specific example that caused me to add this: On shutdown, CCL
      signals CCL:PROCESS-RESET, a subclass of SERIOUS-CONDITION. So recent
      test harness modifications to handle SERIOUS-CONDITIONs caused the test
      scripts to mistakenly think there were failures because of calls to
      EXIT-LISP.
      9fd59ed8
  9. 18 Nov, 2015 2 commits
  10. 28 Sep, 2015 1 commit
  11. 26 Sep, 2015 1 commit
  12. 07 May, 2015 1 commit
  13. 12 Mar, 2015 1 commit
  14. 29 Aug, 2014 1 commit
  15. 08 Jul, 2014 1 commit
  16. 20 Jun, 2014 1 commit
  17. 12 May, 2014 1 commit
  18. 27 Mar, 2014 1 commit
  19. 25 Mar, 2014 2 commits
  20. 21 Mar, 2014 1 commit
  21. 17 Mar, 2014 2 commits
  22. 04 Mar, 2014 1 commit
  23. 03 Mar, 2014 1 commit
  24. 02 Mar, 2014 1 commit
  25. 01 Mar, 2014 3 commits
    • Francois-Rene Rideau's avatar
      Define and export uiop:argv0, for the sake of a portable buildapp-like dispatched entry. · 15b373a5
      Francois-Rene Rideau authored
      Fix some issues with SCL:
      * it doesn't like run-program from a modified directory?
      * it somehow pushes :non-base-chars-exist-p even though +non-base-chars-exist-p+ is NIL???
      15b373a5
    • Francois-Rene Rideau's avatar
      580e8d1e
    • Francois-Rene Rideau's avatar
      Fixes and refactoring of asdf/bundle and uiop/image for ECL. · c52a4f83
      Francois-Rene Rideau authored
      Testing the latest cl-launch with ECL showed massive bitrot in the bundle support for ECL.
      * Make better use of ASDF3's class hierarchy in redefining asdf/bundle,
        by removing and adding the mixins: goodbye bundle-compile-op and monolithic-,
        hello link-op and gather-op. Also, reinstate bundle-system as a class
        to hold prologue and epilogue (but do we need these, now with portable ASDF3 image support?)
      * rename binary-op to deliver-asd-op, same for monolithic-
      * Introduce image-op as a superclass of program-op,
        for dumping an executable image with the regular top-level.
      * Redo the way operation flags are or aren't propagated:
        gather-op explicitly does not pass the flags,
        so toplevel options are for the toplevel build only
        -- though they belong to the system, not the operation.
        Therefore get rid of no-ld-flags-op -- if there are flags to pass,
        they should be passed explicitly in gather-op;
        or more likely, they might be slots in the system, or the plan,
        or special variables for the current session.
        OPERATE tries harder to preserve the original-initargs,
        which are not clobbered by asdf/bundle's initialize-instance anymore.
      * Punt for command-line-arguments on LispWorks,
        so the user has a slight chance of setting them.
      c52a4f83
  26. 15 Jan, 2014 1 commit
  27. 20 Dec, 2013 4 commits
  28. 07 Dec, 2013 1 commit
  29. 16 Nov, 2013 1 commit