* Instead of a special variable, the parsed struct type instance is now
  annotated to let the type system know whether we're dealing with
  deprecated semantics (struct type canonicalizes to :pointer) or not.

  This approach should hopefully be much less error-prone.

* Disabled translations for bare structs.

* Fixed and commented out some bogus tests.

Added test struct-values.translation.mem-aref.1 that creates an array
of two struct-pairs, sets their values, then retrieves both pairs.
This test fails: the returned first pair is actually the second, and
the returned second pair is garbage.  Repeating the same test gives
different garbage.

(with-foreign-object (p '(:struct struct-pair) 2)
      (setf (mem-aref p '(:struct struct-pair) 0) '(1 . 2)
            (mem-aref p '(:struct struct-pair) 1) '(3 . 4))
      (values (mem-aref p '(:struct struct-pair) 0)
              (mem-aref p '(:struct struct-pair) 1)))
(3 . 4)
(213829 . 0)

(with-foreign-object (p '(:struct struct-pair) 2)
      (setf (mem-aref p '(:struct struct-pair) 0) '(1 . 2)
            (mem-aref p '(:struct struct-pair) 1) '(3 . 4))
      (values (mem-aref p '(:struct struct-pair) 0)
              (mem-aref p '(:struct struct-pair) 1)))
(3 . 4)
(425347 . 0)

Add struct-pair+double recursive structure translation test, which
passes when run individually in SBCL.

Define a structure struct-pair-default-translate and test
struct-values-default.translation.mem-ref.1 to test translation from
foreign structure to plist.  In order to make this work, rewrite
mem-ref function to parallel exactly the mem-ref compiler macro,
because conditionals were not correct.

Add :argument-precedence-order to defgenerics
translate-into-foreign-memory and translate-to-foreign, and removed
the 'cons specializer in the tests.

Otherwise the default plist-handling method kicks in.

Add respective struct tests, one of which fails. Also, missing union
tests.

Update STRUCT-VALUES.* tests.

All fail but STRUCT-VALUES.TRANSLATION.1.

With tests STRUCT-WRAPPER.[12]

(initial suggestion and patch courtesy of Stelian Ionescu)

- Update test suite to reflect this.
- Tested on Allegro and Lispworks. Should work for ECL as well.

Three new features:
  - stdcall (defcfun, foreign-funcall(-pointer), defcallback)
  - namespaces, associate foreigns vars and functions to a
    specific library. (CLISP and Lispworks only)
  - close-foreign-library actually works now.

Backwards incompatible changes:
  - define-foreign-library's syntax changed slightly, can't
    load more than one foreign library per define-foreign-library
    form anymore.
  - defcvar's syntax changed.
  - foreign-funcall can't funcall pointers anymore. Use
    foreign-funcall-pointer for that.

- CLISP: special case for :[unsigned-]long-long on darwin/ppc.
- SBCL: missing special case for :unsigned-long-long.
- New regression test: STRUCT.ALIGNMENT.8.

- change "docs" to be the default target in doc/Makefile.
- add Stephen the list of authors in cffi-manual
- rename "Index" to "Comprehensive Index" so that Index.html doesn't
  clash with index.html on OSX.
- add varargs example for defcfun
- rename features (again) foreign-funcall -> no-foreign-funcall,
  long-long -> no-long-long.

- Renamed :cffi/no-foreign-funcall and :cffi/no-long-long to
  cffi-features:foreign-funcall and cffi-features:long-long respectively
  (with the opposite meanings of course).
- Also new features (in the cffi-features package): darwin, unix, windows,
  ppc32 and x86. More could be added. The cffi-sys backends are
  responsible for pushing these features.

- New functions: FOREIGN-SLOT-NAMES and FOREIGN-SLOT-OFFSET.
- Test FOREIGN-SLOT-NAMES: STRUCT.NAMES
- Removed respective TODO item and added a new one: document these.

- foreign-struct-slot-value-form and foreign-struct-slot-set-form
  were not quoting the type in the resulting mem-ref form.
- struct.5, regression test.

- Make get-slot-info (and thus foreign-slot-value) work on typedefs
  to a struct type.
- Regression test: STRUCT.4

- WITH-FOREIGN-OBJECT should eval the type parameter. Fixed this
  as well as the examples and tests that assumed it wasn't evaluated.
- document WITH-FOREIGN-OBJECT in the manual.
- removed the pseudo ability to specialize arguments on
  define-type-translator. That didn't quite work, doh.
- removed dead code from cffi-uffi-compat::foreign-type-size
  specialized on the uffi-array-type.
- uffi-compat: parse (:array <type>) as (:array <type> 1)
- uffi-compat: reflect the fact that CFFI:WITH-FOREIGN-OBJECT now
  evaluates the type parameter.

- Make the foreign-slot-value method on simple-struct-slot apply the
  adequate type translator. Same thing for the setter.
- Test this in STRUCT.STRING.1 and STRUCT.STRING.2

- Normalize the different implementation features (:powerpc, :macos,
  :macosx32) into :ppc32 and :darwin.
- Force Allegro, CLISP, Lispworks, SBCLto return 8 as :double's
  alignment on darwin/ppc32
- Correctly calculate alignments for darwin/ppc32's strange ABI.
- New tests: STRUCT.ALIGNMENT.[567].

Bug report and initial bugfixes courtesy of Marco Gidde.

- fixed setf expanders for mem-ref, mem-aref, and foreign-slot-value.
  (also mem-aref was evaluating the type argument twice)
- likewise fixed cmucl's, openmcl's and sbcl's setf expanders for %mem-ref.
- regression tests: mem-aref.eval-type-x2, mem-ref.nested, mem-aref.nested
  and struct.nested-setf.

- Commented out the test STRUCT.ALIGNMENT.5 because an empty struct is
  not valid standard C and some compilers, namely VC++, won't compile it.
- Add -ansi -pedantic to the compiler flags to catch things like this
  sooner next time.

- Fixed notice-foreign-structure-definition to handle empty
  struct definitions correctly.
- Added new test for this: STRUCT.ALIGNMENT.5

- defcvar now works (more) correctly with aggregate types.
- struct types now carry alignment information which is now
  correctly calculated (for unions too) according to the x86
  ABI (tested on darwin/ppc too, but should study other ABI
  docs at some point).
- also tail padding is now correctly added to structs, when
  necessary. (also according to the x86 ABI).
- 4 new tests for this struct bug (reported by Wilco Greven).
- tweaked tests/bindings.lisp to correctly load the library on
  Lispworks and to make it more portable across win/nix/etc..