1. 20 Apr, 2013 1 commit
  2. 07 Apr, 2013 1 commit
  3. 06 Apr, 2013 1 commit
  4. 30 Mar, 2013 1 commit
    • Liam M. Healy's avatar
      New macro define-derived-system-of-units · 250eedb5
      Liam M. Healy authored
      The new macro define-derived-system-of-units is like define-system-of-units except that it also takes a base system of units from which to derive the new system of units, and does not default to the current system of units.
      250eedb5
  5. 25 Mar, 2013 1 commit
    • Liam M. Healy's avatar
      Complete changes for system of units specification · bc899205
      Liam M. Healy authored
      - Fix number of arguments in calls to functions that no longer take
        sysunits or addition-units arguments.
      - Change nf-option so that it works dynamically, i.e., the default
        value is retrieved at run time, not macro expansion time.
      - Name argument to parameter-value must be in the keyword package.
      - The system of units can be nil, in which case an unreadable object
        is printed, with the sexp for the dimensions printed.  
      - New variable *basis-physical-dimensions* used to give the dimension
        without units.
      - All nf definitions in a separate file.
      
      System now compiles and loads without error.
      bc899205
  6. 23 Mar, 2013 1 commit
    • Liam M. Healy's avatar
      System of units specification is through an nf-option · 6ca09dfb
      Liam M. Healy authored
      The specification of the system-of-units is through an nf-option, and
      no longer passed as arguments to functions.  Systems of units (with or
      without augmented units) are made via make-sysunits or
      define-system-of-units, set-default-system-of-units that call it.
      
      This does not compile.
      6ca09dfb
  7. 22 Mar, 2013 1 commit
  8. 19 Mar, 2013 1 commit
  9. 18 Mar, 2013 1 commit
    • Liam M. Healy's avatar
      Time interval parsing and unparsing, specific TeX format · 61d4de15
      Liam M. Healy authored
      + New functions to convert timeparse to and from time intervals,
      broken out of the formatting functions and exported:
      timeparse-time-interval and time-interval-timeparse.
      
      + New function dms-angle broken out of write-dms; renamed old
      dms-angle to angle-dms; this is to make the usage consistent with
      above (returned thing named first, argument named second,
      e.g. "dms-angle = dms from angle."
      
      + Removed 'required-component argument to iso8601-string, which was unused.
      
      + iso8601-string T-separator argument default for texstyle is a space.
      
      + Example
           (set-nf-options :style :tex)
           (iso8601-time-interval #_1134821.43_s)
           "13d 3h13m41.430s"
      61d4de15
  10. 05 Mar, 2013 1 commit
  11. 25 Feb, 2013 1 commit
  12. 21 Feb, 2013 1 commit
  13. 19 Feb, 2013 1 commit
  14. 18 Feb, 2013 1 commit
  15. 17 Feb, 2013 1 commit
  16. 16 Feb, 2013 1 commit
    • Liam M. Healy's avatar
      New function check-dimension-or-type · b72088c7
      Liam M. Healy authored
      New function check-dimension-or-type which returns T if the physical dimension is correct, or, if not a physical dimension, if the type matches.  This is used by check-parameter-type to give more specific error messages if the physical dimension does not match.
      b72088c7
  17. 10 Feb, 2013 1 commit
  18. 04 Feb, 2013 1 commit
  19. 02 Feb, 2013 1 commit
    • Liam M. Healy's avatar
      Format 1d grid readably always binds vector-format to :horizontal · 4a6db769
      Liam M. Healy authored
      Previously, vector-format was treated as specified whether style was
      :readable or not.  This resulted in missing parentheses when it was
      :coordinate-unit-vectors, and in an odd-looking (but syntactically
      correct) vertical breakup of components when it was :vertical.  This
      has been changed so that it is locally rebound to :horizontal when
      style is :readable.  Tests have been added to test-grid-formatting to
      cover the three readable cases, which should and do now all look the
      same.
      4a6db769
  20. 31 Jan, 2013 2 commits
  21. 23 Jan, 2013 1 commit
  22. 21 Jan, 2013 1 commit
    • Liam M. Healy's avatar
      Reading org-mode tables as parameters · e19a8e0c
      Liam M. Healy authored
      Define a new function #'make-parameters-from-table that is designed to read an org-mode table that is parsed and passed to CL by org-babel.  There is some documentation and usage examples in the new file input-output/org-mode.lisp; this should be separated out into a documentation file and expanded.  Tests should be designed and added.
      e19a8e0c
  23. 02 Jan, 2013 1 commit
  24. 23 Dec, 2012 1 commit
  25. 03 Dec, 2012 1 commit
    • Liam M. Healy's avatar
      New system math-high-tests, file name change norm-vector-product · a439f4b4
      Liam M. Healy authored
      New system math-high-tests test with one test, linear-algebra, that initially has a single form to test addition of a vector to its reverse.  The plan is that this test will contain tests of dyadic arithmetic functions (addition, etc.).  Changed the name of files higher.lisp to norm-vector-product.lisp, as it is more specific.
      a439f4b4
  26. 24 Nov, 2012 2 commits
  27. 11 Jun, 2012 2 commits
  28. 04 Jun, 2012 1 commit
  29. 29 Apr, 2012 1 commit
  30. 28 Apr, 2012 1 commit
  31. 08 Apr, 2012 1 commit
    • Liam M. Healy's avatar
      Start attempts to make array access more efficient · 6f148a2f
      Liam M. Healy authored
      Extend the compiler macros to do a better job of going directly to fast access functions (like cl:aref) when the type is known due to declarations.  Some tests from Sumant.  Note that it seems that SBCL is not pulling variable declarations in variable-information as expected, so the declarations are not actually doing anything.
      6f148a2f
  32. 11 Mar, 2012 1 commit
    • Liam M. Healy's avatar
      Fix format of floats with exponent markers · dd86ab06
      Liam M. Healy authored
      Change find to find-if, which fixes a problem with a duplicate exponent.
      Add argument specific-exponent-marker to format-float-switch; if T,
      use an exponent marker f, d, etc. dependent on the type.  This
      argument defaults to T, and the calling function nf float does not specify it.
      dd86ab06
  33. 20 Feb, 2012 1 commit
  34. 01 Jan, 2012 1 commit
  35. 30 Dec, 2011 2 commits
    • Liam M. Healy's avatar
      f75c6720
    • Liam M. Healy's avatar
      New function read-indexed-data · e389340e
      Liam M. Healy authored
      New function read-indexed-data will read data line-by-line with index
      parameters and value.  The index parameters may be indices or a set of
      numbers that maps to indices.  Input can be a stream, file, or
      string.  Possible improvements would be: accept make-simple-grid
      keyword arguments, fill in an existing grid if supplied, do not
      require maximum-index-parameters.
      e389340e
  36. 28 Dec, 2011 1 commit
    • Liam M. Healy's avatar
      Pass #'length through to CL · bcecd7fe
      Liam M. Healy authored
      It is necessary to pass the function antik:length through to cl:length
      because the function is shadowed (for physical-quantities 'length).
      bcecd7fe