Skip to content
Snippets Groups Projects
  1. Jul 19, 2003
  2. Jul 17, 2003
  3. Jul 01, 2003
  4. Jun 30, 2003
  5. Jun 26, 2003
  6. Jun 25, 2003
  7. Jun 16, 2003
  8. Jun 14, 2003
    • csr21's avatar
      Clean up compilation under sbcl: a couple of IGNOREs, a couple of · 2f7a173a
      csr21 authored
      Clean up compilation under sbcl: a couple of IGNOREs, a couple of
      s/load/:load-toplevel/, one s/*foo*/+foo+/.  Still many many compilation
      notes, but since the render protocol is inherently slow anyway...
      
      darcs-hash:20030614141607-ed5a3-c868eb7fafc194534644a6d5bc3a64e1f86d9fd1.gz
      2f7a173a
    • gilbert's avatar
      imported as is. · eec787e8
      gilbert authored
      darcs-hash:20030614132733-09c7f-4f7e3be91d8fb8ffd2294d9361488b77aa3a424f.gz
      eec787e8
  9. Jun 07, 2003
    • csr21's avatar
      Fix the bugs that the image test reveals. · 6820b4e3
      csr21 authored
      Fix the bugs that the image test reveals.
      The bug fix isn't great, though; we simply disable the FAST-xxx-PIXARRAY
      routines for sbcl, because they demonstrably don't work.  Patches will
      be welcome from anyone who understands what they are meant to do.
      
      darcs-hash:20030607093043-ed5a3-e3dd58efcab2a2f678d25864adb98425c3d91ca6.gz
      6820b4e3
    • csr21's avatar
      Fix the image test: · 4095a1b9
      csr21 authored
      Fix the image test:
      * don't abuse the "cursor hotspot" fields, which are defined to be
        CARD16 or NULL, for communicating the position of the (sub)image on
        the root window; instead, use the image plist (and adjust as necessary
        if a subimage is copied).
      * rather than writing in very very dark red (or blue) on black,
        explicitly ask for white on black.
      
      darcs-hash:20030607090943-ed5a3-65828f7e8c98f4dcaa3d9bdcdb23569a90a16f67.gz
      4095a1b9
  10. Jun 06, 2003
  11. Jun 05, 2003
    • csr21's avatar
      A different approach to the "SBCL DEFCONSTANT" problem: instead of · 128cfa08
      csr21 authored
      A different approach to the "SBCL DEFCONSTANT" problem: instead of
      sticking icky #+sbcl defconstant-eqx everywhere, concentrate the
      nastiness in an :around method for ASDF:PERFORM on CLX-SOURCE-FILE.
      This, of course, means that the sources do not build on sbcl if you
      compile and load them one by one; I can live with that.
      
      darcs-hash:20030605191822-ed5a3-3b02c39acf535f330f3313522984197a7a351ec8.gz
      128cfa08
    • csr21's avatar
      We might as well use OPEN-DEFAULT-DISPLAY here, too · f1f00213
      csr21 authored
      darcs-hash:20030605160514-ed5a3-da771794b914534f83169958cdf8b4d7862b0835.gz
      f1f00213
  12. Jun 04, 2003
  13. Jun 02, 2003
    • csr21's avatar
      Quick, before I delete my repository again: for SBCL, bind · cd37eaab
      csr21 authored
      Quick, before I delete my repository again: for SBCL, bind
      *DERIVE-FUNCTION-TYPES* around compilation of clx files, since we're not
      going to be changing internals very often (and extremely rarely in a
      type-incompatible way).
      
      darcs-hash:20030602124548-ed5a3-0e1d5c74fec24f401a9752459b5aec8f98f9dbac.gz
      cd37eaab
  14. May 29, 2003
  15. May 15, 2003
  16. May 08, 2003
  17. May 06, 2003
  18. Apr 30, 2003
Loading