1. 28 May, 2003 1 commit
  2. 19 May, 2003 1 commit
  3. 13 May, 2003 1 commit
  4. 06 May, 2003 2 commits
  5. 19 Mar, 2003 2 commits
  6. 17 Mar, 2003 1 commit
    • Daniel Barlow's avatar
      Bug fix: When we get to a file that needs recompilation, usually we · 9de67ba5
      Daniel Barlow authored
      set :forced on operation-ancestor, to ensure that subsequent
      components are rebuilt.  However, touching a random file in db-sockets
      is not usually a convincing reason to rebuild all of araneida and all
      of cliki, so when we do this cross-system leap, we remove the parent
      link
      
      Unfortunately, that breaks the what-have-we-done-so-far information,
      which is also kept in the ancestor operation.  So, let's not
      do that any more.  Compilations will be slower than intended, but faster than
      the infinite circular compilation behaviour that users of 1.65 observed
      9de67ba5
  7. 16 Mar, 2003 3 commits
  8. 07 Mar, 2003 1 commit
  9. 04 Mar, 2003 1 commit
  10. 03 Mar, 2003 1 commit
  11. 13 Feb, 2003 1 commit
    • Christophe Rhodes's avatar
      Make messages that asdf prints slightly nicer, by using pretty-printer · a5a00774
      Christophe Rhodes authored
      justification routines.
      
      Sucky thing I: CLISP dislikes calls to FORMAT with pretty-print format
      strings.  Bizarrely, though, it accepts calls to FORMATTER with the
      same format strings.  So use FORMATTER everywhere to keep clisp happy.
      
      Sucky thing II: CMUCL can't do ENOUGH-NAMESTRING, because it has this
      weird ~UNSPECIFIC~ thing in its *DEFAULT-PATHNAME-DEFAULTS*.  So don't
      use ENOUGH-NAMESTRING, even though it's crying out for it.
      a5a00774
  12. 08 Feb, 2003 1 commit
    • Daniel Barlow's avatar
      Make the :serial switch actually do something (useful for CLX systems) · d2063dd1
      Daniel Barlow authored
      Some glue that won't affect too many people but that lets ASDF cleanly
      hook the extensible PROVIDE/REQUIRE mechanism in upcoming SBCL 0.7.13
      
      Fix wild-modules to not object when we call (setf (module-components))
      on a new or about-to-be-reinitialised module.  We need to be able to
      do this to get the weeds out, so it should be legal
      d2063dd1
  13. 04 Feb, 2003 1 commit
    • Daniel Barlow's avatar
      Rationalise the system-definition-finding machinery a bit. · 60e59ce3
      Daniel Barlow authored
      *central-registry* no longer takes functions (that feature was broken
      anyway), but a new list *system-definition-search-functions*
      does.  Its default content is a single function designator for
      sysdef-central-registry-search, which does the *central-registry*
      search
      
      Update documentation to match
      
      Fix test1.script to call the preferred OPERATE instead of OOS
      
      Update documentation to describe COMPONENT-PROPERTY method
      60e59ce3
  14. 28 Jan, 2003 1 commit
  15. 09 Dec, 2002 1 commit
  16. 02 Dec, 2002 1 commit
  17. 23 Nov, 2002 1 commit
  18. 18 Nov, 2002 1 commit
  19. 12 Nov, 2002 1 commit
    • Daniel Barlow's avatar
      More frobbing of TRAVERSE · ae9d0db6
      Daniel Barlow authored
       - now it takes two arguments not three
       - and returns (operation . component) pairs
       - which don't include pruned-ops
      
      Sorted out some load-op methods to use input-files instead of faking their
      own compile-ops just to find out what the input files are
      ae9d0db6
  20. 08 Nov, 2002 1 commit
    • Daniel Barlow's avatar
      · d8bd7cfb
      Daniel Barlow authored
      shouldn't do global proclamations, that's bad karma (reported by Gary Byers)
      
      modules maybe now can have no components (Matthew Danish, SF bug id 625738)
      (added an :initform nil, haven't actually tested)
      
      significant reworking to handle recompilation of dependencies properly
       - use of component properties for last-compiled/last-loaded is gone
       - new internal gf INPUT-FILES (COMPONENT OPERATION)
       - TRAVERSE doesn't actually perform anything, but it now has a defined
         return value: a list of the ops and components that need doing.
         OPERATE loops over answers from TRAVERSE
       - intended to have no user-visible effects, but ICBW!
         May break existing systems!
       - still doesn't do cross-module dependencies properly, sigh.  but we're
         closer
       - default output-files method now returns NIL instead of causing an error
      d8bd7cfb
  21. 20 Sep, 2002 5 commits
  22. 17 Sep, 2002 2 commits
  23. 13 Sep, 2002 3 commits
  24. 12 Sep, 2002 1 commit
  25. 09 Sep, 2002 1 commit
  26. 30 Aug, 2002 2 commits
  27. 28 Aug, 2002 2 commits
    • Daniel Barlow's avatar
    • Daniel Barlow's avatar
      *asdf-revision* is a new special variable. It's a list (1 31) or · 951d5829
      Daniel Barlow authored
      similar which corresponds to the CVS revision of asdf.lisp
      - compile-file errors
      
      The default behaviour for coping with compile-file errors has changed:
      
       (defvar  *compile-file-warnings-behaviour* :warn)
      -(defvar  *compile-file-failure-behaviour* :error)
      +(defvar  *compile-file-failure-behaviour* #+sbcl :error #-sbcl :warn)
      
      Most lisp implementations (all that I know of except for sbcl) stop
      and enter the debugger for catastrophic errors in file compilation.
      SBCL otoh has a much finer warning/style-warning distinction and
      should not ever return failure-p unless for a catastrophic error
      
      Fix printing of system-definition-error in CMUCL (thanks to Bob Rogers)
      
      Moved a lot of messing around with 'last-compiled and 'last-loaded
      properties into :before and :after methods on source-file so that
      "don't reload if already loaded" works for new source file types that
      users create
      
      There is a new operation load-source-op based on Kevin Rosenberg's
      implementation of same.  I'm not altogether sure yet how this should
      interact with 'last-compiled and 'last-loaded: suggestions welcomed
      
      OPERATE now does the operation inside a WITH-COMPILATION-UNIT form, to
      reduce noise from the compiler about forward definitions.  Courtesy of
      Bob Rogers
      951d5829