1. 06 May, 2012 1 commit
    • Raymond Toy's avatar
      Fix up some bugs in the last fix for using STRING declarations instead of · 7cb602dc
      Raymond Toy authored
      (SIMPLE-ARRAY CHARACTER).
      
      src/f2cl1.l:
      o Fix typo in RELAX-ARRAY-DECL.
      
      src/f2cl5.l:
      o Need to handle the new STRING declarations correctly in
        GENERATE-WITH-ARRAY.
      
      src/macros.l:
      o In MULTI-ARRAY-DATA-AUX, handle arrays of type CHARACTER specially.
        Make the declaration STRING instead of (SIMPLE-ARRAY CHARACTER).
      7cb602dc
  2. 26 Nov, 2011 1 commit
  3. 24 Nov, 2011 1 commit
    • Raymond Toy's avatar
      Fix up implementation of CYCLE and EXIT. · 28c496a2
      Raymond Toy authored
      f2cl1.l:
      o Go to f2cl-lib::exit instead of loop because that bypasses the update
        of the loop variable.
      
      macros.l:
      o Have the FDO macro insert the CONTINUE and EXIT labels at the
        appropriate places so that the CYCLE and EXIT code goes to the right
        places.
      28c496a2
  4. 23 Nov, 2011 1 commit
  5. 11 Jun, 2011 1 commit
    • Raymond Toy's avatar
      Handle array slicing a bit better with displaced arrays. · 52bbcd8e
      Raymond Toy authored
      f2cl5.l:
      o If the array we're slicing is one of the arrays listed in a
        WITH-MULTI-DATA-ARRAY, use the actual underlying array with the
        appropriate offset.
      
      macros.l:
      o Update ARRAY-SLICE macro to include an additional offset parameter to
        indicate what the, if any, the offset is for this array, if the array
        itself is a displaced (sliced) array.
      52bbcd8e
  6. 28 Feb, 2011 1 commit
  7. 20 Feb, 2011 1 commit
  8. 28 Dec, 2010 1 commit
  9. 17 May, 2010 1 commit
    • rtoy's avatar
      src/f2cl1.l: · aad4f000
      rtoy authored
      o Need to know the actual type when making a compatible sequence.
      o Convert plain integer type to integer4 types, which is what Fortran
        would do.  We don't want general Lisp integer type.
      
      src/macros.l:
      o Change MAKE-COMPATIBLE-SEQ to be a macro.
      o Need to know the actual array type to create the correct type of
        sequence.
      aad4f000
  10. 23 Feb, 2010 1 commit
    • rtoy's avatar
      Support the Fortran capability of passing an array of one type · 12a40813
      rtoy authored
      to a routine expecting a different type.  Currently only supports REAL
      and COMPLEX arrays (and their double precison versions).
      
      NOTES:
      o Update
      
      f2cl0.l:
      o Export new symbols f2cl-copy-seq and make-compatible-seq.
      
      f2cl1.l:
      o New variable *copy-array-parameter* for keeping track of the option
        for f2cl and f2cl-compile.
      o Update f2cl and f2cl-compile to recognize :copy-array-parameter.
      o Modify massage-arglist and generate-call-to-routine to handle the
        new :copy-array-parameter capability.
      
      f2cl5.l:
      o Fix issue where quoted elements were modified.  They shouldn't be.
      o Fix issue where (array simple-float (*)) would get erroneously
        converted to (array simple-float (f2cl-lib:int-mul)).  We want to
        leave bare * alone.
      
      macros.l:
      o New macro f2cl-copy-seq to generate code to copy a sequence
        appropriately.
      o New function to create a compatible array to support
        :copy-array-parameter.
      12a40813
  11. 08 Jan, 2009 1 commit
    • rtoy's avatar
      f2cl0.l: · 8ec239a8
      rtoy authored
      o Export *STOP-SIGNALS-ERROR-P*
      
      macros.l:
      o Add *STOP-SIGNALS-ERROR-P* to allow user to control whether STOP
        signals a continuable error or not.  Default is to signal the
        error.
      8ec239a8
  12. 07 Jan, 2009 5 commits
  13. 15 Sep, 2008 1 commit
  14. 10 Sep, 2008 1 commit
  15. 22 Aug, 2008 1 commit
  16. 21 Aug, 2008 1 commit
  17. 26 Mar, 2008 1 commit
  18. 08 Mar, 2008 1 commit
  19. 07 Mar, 2008 1 commit
  20. 06 Mar, 2008 2 commits
  21. 26 Feb, 2008 1 commit
  22. 22 Feb, 2008 2 commits
  23. 30 Sep, 2007 1 commit
  24. 28 Sep, 2007 4 commits
    • rtoy's avatar
      First attempt at getting implied-do loops in data statements working · d2fa99eb
      rtoy authored
      with nested loops.
      
      f2cl1.l:
      o PARSE-DATA-IMPLIED-DO handles implied do loops even when the loops
        are nested.
      
      macros.l:
      o Update PROCESS-IMPLIED-DO to handle the new forms returned by
        PARSE-DATA-IMPLIED-DO.
      o Don't create constants out of the initializer since we use POP to
        access them one by one.
      o Minor tweak for list-directed output to allow a slightly longer line
        length. This matches what g77 produces for one simple test case.
      d2fa99eb
    • rtoy's avatar
      o For REWIND and CLOSE$, declare the result as ignorable. · fc146e5e
      rtoy authored
      o For list-directed output, don't print out strings as an array with
        spaces between each element.  Strings should go out as strings.
      fc146e5e
    • rtoy's avatar
      Some cleanup for list-directed output: · 44513100
      rtoy authored
      o Complex numbers should be printed in the form (r, i), not #c(r, i)
      o Arrays should print out only the elements instead of #(...).
      44513100
    • rtoy's avatar
      To support multidimensional arrays in implied do loops better, we need · 639359d0
      rtoy authored
      to pass the entire array bounds, including upper and lower limits so
      that array indexing can work.
      
      f2cl5.l:
      o Find the entire array bounds.
      o Don't use make-declaration to get the array type.  Explicitly look
        through *explicit_vble_decls* to find the type.  (Are there other
        places we need to look?)
      
      macros.l:
      o Pass the entire list of array bounds to fref so we can handle
        multidimensional arrays.
      639359d0
  25. 27 Sep, 2007 2 commits
    • rtoy's avatar
      When we run out of format specs, but there are still items to print, · da0e3d7b
      rtoy authored
      we go back and find the first repeat spec and start there.
      
      If there is no such thing, we just reuse the entire format spec.  Not
      sure if this is right or if it's a bug.  Maybe we should signal an
      error?
      da0e3d7b
    • rtoy's avatar
      Support the L edit descriptor better. · aae9d69f
      rtoy authored
      f2cl5.l:
      o Recognize the L descriptor and convert it to ~wA.
      
      macros.l:
      o Convert T and NIL to :T and :F, respectively.  When coupled with ~A,
        this prints as T and F, as desired.
      aae9d69f
  26. 26 Sep, 2007 2 commits
    • rtoy's avatar
      9dbc2392
    • rtoy's avatar
      Better list-directed output. · 7066ece6
      rtoy authored
      f2cl5.l:
      o For list-directed output (format is *), return :list-directed to
        tell format that we're using list-directed output.  (The previous
        scheme didn't really work well.)
      
      macros.l:
      o Add FLATTEN-LIST function
      o Don't output a newline for repeated items.  We shouldn't do that.
      o Add support for :list-directed output.  We recognize that and then
        just output all the args in a special way.
      7066ece6
  27. 25 Sep, 2007 3 commits
    • rtoy's avatar
      f2cl5.l: · 537891f7
      rtoy authored
      o Slight change in the format used for "*" format.
      o Change the repeatable descriptors to remove the repeat count if the
        count is 1.  This was confusing the execute-format when determining
        when to print out newlines.  This change applied to I, F, E, D, and
        G descriptors.
      
      macros.l:
      o Handle printing of "repeat forever" loops better.  An extra arg to
        EXECUTE-FORMAT tells us to repeat "forever".
      o Output a newline at the end of a repeated specification.
      537891f7
    • rtoy's avatar
      For repeated descriptors, we were printing a new line after each item · 441d0b77
      rtoy authored
      instead of after all items had been printed.  Output new line only
      once, when we're done.
      441d0b77
    • rtoy's avatar
      Oops. We should always process #\: and exit only if there is more · 4135b245
      rtoy authored
      args to process.
      4135b245