1. 20 Jan, 2016 1 commit
  2. 29 Dec, 2015 1 commit
  3. 06 Dec, 2015 1 commit
  4. 17 Oct, 2014 1 commit
  5. 22 Feb, 2014 1 commit
  6. 16 Feb, 2014 1 commit
  7. 04 Oct, 2013 1 commit
    • Add direct foreign-vector view of 1D histograms · b2f6ae1f
      Defined two functions, view-range-as-foreign-array and view-bin-as-foreign-array that proved direct foreign-vector view of 1D histograms (shared data memory with the foreign histogram). This maybe be a precursor to some eventual generic "view" function, so the function symbols are not exported yet as the name may change. Also, the defcstruct should be done as a grovel on the C header. However, the functions do work now.
      Liam M. Healy authored
  8. 07 May, 2013 1 commit
  9. 13 Jan, 2012 1 commit
  10. 30 Oct, 2011 1 commit
  11. 12 Sep, 2011 1 commit
  12. 19 Aug, 2011 1 commit
  13. 26 May, 2011 1 commit
  14. 22 Feb, 2011 1 commit
  15. 19 Feb, 2011 1 commit
  16. 12 Jan, 2011 1 commit
  17. 10 Jan, 2011 1 commit
  18. 14 Nov, 2010 2 commits
  19. 16 Jul, 2010 2 commits
  20. 07 Jul, 2010 1 commit
  21. 02 Jul, 2010 1 commit
  22. 01 Jul, 2010 1 commit
  23. 30 Jun, 2010 1 commit
  24. 27 Jun, 2010 1 commit
  25. 27 Dec, 2009 1 commit
  26. 21 Dec, 2009 2 commits
    • Move copy, copy-making-destination to grid · 3680b307
      Moved the definition of the generic function copy-making-destination
      to copy objects while making a destination object, and the wrapper
      function #'copy for that and copy-to-destination to grid/ and c-array:
      package.
      Liam Healy authored
    • Move copy-to-destination to grid · 06f99392
      The generic function definition for copy-to-destination is now in
      grid/ and the c-array package.  This will make it accessible to
      non-GSLL foreign library interfaces, but the new location is a bit
      deceptive because a new method needn't be a subclass of foreign-array.
      Liam Healy authored
  27. 27 Aug, 2009 1 commit
  28. 05 Jul, 2009 1 commit
  29. 17 Jun, 2009 1 commit
  30. 11 Jun, 2009 1 commit
    • Hand expansion of project-ntuple; start writing ntuple example · 9f1f9900
      In order to avoid or defer the complicated process of modifying
      defmfun definitions to handle two cbstructs, the definition of
      project-ntuple has been written by hand as a defun etc., by expanding
      a simpler defmfun and making the necessary modifications and mergers
      with an editor.  It has not yet been tested.  The test given in GSL
      ntuple/test.c is partially ported to Lisp.
      Liam Healy authored
  31. 10 Jun, 2009 1 commit
  32. 25 May, 2009 1 commit
  33. 27 Apr, 2009 1 commit
  34. 21 Feb, 2009 1 commit
  35. 25 Jan, 2009 1 commit
  36. 12 Jan, 2009 1 commit
    • Redefine #'copy · 419e70d4
      Redefine #'copy so that it has one required argument (source) and one
      optional argument (destination).  For marrays, create a destination
      using make-marray, if one was not given, and then use the memcpy
      functions to fill the values.  For all other mobjects, copy will copy
      objects by calling one of two generic functions, #'copy-to-destination
      and #'copy-making-destination, depending on whether a destination was
      supplied.  The former is defined if a GSL _memcpy function exists.
      The latter will look for a GSL _clone function, then make a CL
      instance with the pointer returned as mpointer.  If there is no clone
      function, it will attempt to make a new object using make-load-form,
      so will fail if that method has not been defined.
      
      Added a check to defmfun that signals an error if a variable is used
      in the C arguments other than top-level that is not used in the CL
      arglist.  This should not happen and is usually an indication of a
      typo.
      Liam Healy authored
  37. 04 Jan, 2009 1 commit
    • Elementwise array operations named "elt" · 4bfa337d
      GSL functions that operate from one corresponding array elements and
      then overwrite the first array, as well as those that operate on each
      element of an array with a scalar, are now named starting with "elt",
      as in "elt+"; formerely they started with "m".  This is to reinforce
      the idea that they are not generally a mathematical solution to
      algebraic problems, just elementwise operations.  The scalar
      operations (addition, multiplication) are methods of elt+ and elt* as
      well. Test results 64 bit:
      SBCL: TOTAL: 1223 assertions passed, 5 failed, 0 execution errors.
      CCL: TOTAL: 1228 assertions passed, 0 failed, 0 execution errors.
      Liam Healy authored