]> www.ginac.de Git - ginac.git/log
ginac.git
22 years agosome cleanups
Richard Kreckel [Mon, 25 Jun 2001 00:18:10 +0000 (00:18 +0000)]
some cleanups

22 years ago- added find() (like has(), but returns list of all occurrences)
Christian Bauer [Sat, 23 Jun 2001 16:52:04 +0000 (16:52 +0000)]
- added find() (like has(), but returns list of all occurrences)
- added lst::sort() and lst::unique()
- status_flags::expanded is only taken into account when no expand options
  are specified. This makes it possible to re-expand with other options.
- added expand_options::expand_function_args
- collect(foo, {}) doesn't crash any more
- introduced match_same_type() method which is slightly similar to
  is_equal_same_type() but doesn't check subexpressions. Fixed a number of
  match() bugs with this (e.g. a==b matched a!=b, and matrices with different
  dimensions but the same number of elements could match). Only container
  classes with additional member variables that have to be equal for a match
  need to implement match_same_type().

22 years agoadded a way to add more functions to ginsh in the file ginsh_extensions.h
Christian Bauer [Fri, 22 Jun 2001 22:02:28 +0000 (22:02 +0000)]
added a way to add more functions to ginsh in the file ginsh_extensions.h
(this is used by the xloops-variant of ginsh)

22 years agosome cleanups
Christian Bauer [Fri, 22 Jun 2001 01:03:27 +0000 (01:03 +0000)]
some cleanups

22 years ago* shift stuff into CVS before I start breaking things. :-)
Richard Kreckel [Fri, 22 Jun 2001 00:44:13 +0000 (00:44 +0000)]
* shift stuff into CVS before I start breaking things.   :-)

22 years agoremoved a lot of superfluous const_cast<>()s
Christian Bauer [Thu, 21 Jun 2001 19:32:57 +0000 (19:32 +0000)]
removed a lot of superfluous const_cast<>()s

22 years agosome sections where sums/products are constructed are slightly more efficient
Christian Bauer [Tue, 19 Jun 2001 23:58:40 +0000 (23:58 +0000)]
some sections where sums/products are constructed are slightly more efficient
(add(exvector)/mul(exvector) instead of +=/*=)

22 years ago* Fix the other half of the test.
Richard Kreckel [Tue, 19 Jun 2001 22:58:20 +0000 (22:58 +0000)]
* Fix the other half of the test.

22 years ago* Fix a bitch of a bug where 1<I (and all other relationals) returned true.
Richard Kreckel [Tue, 19 Jun 2001 21:33:04 +0000 (21:33 +0000)]
* Fix a bitch of a bug where 1<I (and all other relationals) returned true.
  This did not happen for the binary relational operators between numerics.
  Now the behaviour is more consistent (throwing exceptions when the difference
  becomes complex).
* Some regression tests were actually wrong in the light of this bug.  :-(

22 years agoadded decomp_rational()
Christian Bauer [Tue, 19 Jun 2001 19:50:14 +0000 (19:50 +0000)]
added decomp_rational()

22 years ago- The default implementations of evalf(), diff(), normal() and expand() use
Christian Bauer [Tue, 19 Jun 2001 19:50:02 +0000 (19:50 +0000)]
- The default implementations of evalf(), diff(), normal() and expand() use
  map() where useful. This has the nice effect of having a more reasonable
  default behaviour for container functions (most of the evalfchildren() etc.
  stuff is gone).
- diff() works with non-commutative products (product rule) and no longer
  bails on indexed objects.
- added decomp_rational()
- added sqrfree_parfrac() which doesn't yet work in the general case and
  is unsupported

22 years agodocumentation update
Christian Bauer [Tue, 19 Jun 2001 19:45:42 +0000 (19:45 +0000)]
documentation update

22 years ago- do something about the mad cast disease.
Richard Kreckel [Mon, 18 Jun 2001 16:03:31 +0000 (16:03 +0000)]
- do something about the mad cast disease.

22 years ago* fix strange indentation thingie.
Richard Kreckel [Sun, 17 Jun 2001 20:33:40 +0000 (20:33 +0000)]
* fix strange indentation thingie.

22 years agofixed comments
Christian Bauer [Sat, 16 Jun 2001 20:23:26 +0000 (20:23 +0000)]
fixed comments

22 years agobumped archive version because of new way of specifying symmetry for indexed
Christian Bauer [Sat, 16 Jun 2001 20:23:21 +0000 (20:23 +0000)]
bumped archive version because of new way of specifying symmetry for indexed
objects

22 years ago* Remove some cruft.
Richard Kreckel [Sat, 16 Jun 2001 20:13:15 +0000 (20:13 +0000)]
* Remove some cruft.

22 years ago* Supplement some (now deprecated) macros by inlined template functions:
Richard Kreckel [Sat, 16 Jun 2001 18:38:00 +0000 (18:38 +0000)]
* Supplement some (now deprecated) macros by inlined template functions:
  - ex_to_foobar(baz)  ->  ex_to<foobar>(baz).
* De-document is_of_type(obj, type) and document is_a<type>(obj).
* De-document ex_to_type(obj) and document ex_to<type>(obj) eliminating the
  need to declare one's one inline converters in the mystring example.

22 years ago* Supplement some (now deprecated) macros by inlined template functions:
Richard Kreckel [Sat, 16 Jun 2001 18:35:50 +0000 (18:35 +0000)]
* Supplement some (now deprecated) macros by inlined template functions:
  - ex_to_foobar(baz)  ->  ex_to<foobar>(baz).

22 years ago* Supplement some (now deprecated) macros by inlined template functions:
Richard Kreckel [Sat, 16 Jun 2001 18:35:45 +0000 (18:35 +0000)]
* Supplement some (now deprecated) macros by inlined template functions:
  - ex_to_foobar(baz)  ->  ex_to<foobar>(baz).
* Bump up minor version.

22 years ago* Supplement some (now deprecated) macros by inlined template functions:
Richard Kreckel [Fri, 15 Jun 2001 19:20:52 +0000 (19:20 +0000)]
* Supplement some (now deprecated) macros by inlined template functions:
  - is_of_type(foo, type) -> is_a<type>(foo)
  - is_ex_of_type(foo, type) -> is_a<type>(foo)
  - is_exaclty_of_type(foo, type) -> is_exaclty_a<type>(foo)
  - is_ex_exaclty_of_type(foo, type) -> is_exaclty_a<type>(foo)

22 years ago- Just because it's fun: make the test more general using a formula for
Richard Kreckel [Tue, 12 Jun 2001 22:52:22 +0000 (22:52 +0000)]
- Just because it's fun: make the test more general using a formula for
  the closed form of the result suggested by David B.

22 years ago- Fix GCC 2.95.x brokenness.
Richard Kreckel [Tue, 12 Jun 2001 22:51:39 +0000 (22:51 +0000)]
- Fix GCC 2.95.x brokenness.

22 years ago- cruft removed.
Richard Kreckel [Tue, 12 Jun 2001 14:55:49 +0000 (14:55 +0000)]
- cruft removed.

22 years agointroduced new class for constructing symmetry tree definitions
Christian Bauer [Mon, 11 Jun 2001 23:49:51 +0000 (23:49 +0000)]
introduced new class for constructing symmetry tree definitions

22 years ago- Instead of just totally symmetric or antisymmetric, complex symmetries
Christian Bauer [Mon, 11 Jun 2001 23:48:43 +0000 (23:48 +0000)]
- Instead of just totally symmetric or antisymmetric, complex symmetries
  can now be defined for indexed objects. Symmetries are described by a
  tree of "symmetry" objects that is constructed with the sy_none(),
  sy_symm(), sy_anti() and sy_cycl() functions. The symmetry of a function
  with respect to its arguments can also be defined (this is currently
  only used for the Beta function).
- color_trace() and dirac_trace() can be applied to a more general class
  of expressions, e.g. using it on a relation will take the trace on both
  sides etc.
- Generalized map() to take a function object instead of a function pointer.
  This allows passing an arbitrary number of additional state to the
  function being called.
- The unarchiving functinos find_bool(), find_unsigned() and find_string()
  can take an additional "index" argument.

22 years agofixed typos
Christian Bauer [Mon, 11 Jun 2001 23:46:44 +0000 (23:46 +0000)]
fixed typos

22 years agoadapted for new method of specifying symmetries
Christian Bauer [Mon, 11 Jun 2001 23:46:39 +0000 (23:46 +0000)]
adapted for new method of specifying symmetries

22 years agodocumentation update
Christian Bauer [Mon, 11 Jun 2001 23:46:15 +0000 (23:46 +0000)]
documentation update

22 years ago- added lst::remove_first() and lst::remove_last()
Christian Bauer [Sat, 9 Jun 2001 19:17:32 +0000 (19:17 +0000)]
- added lst::remove_first() and lst::remove_last()
- added symmetrize_cyclic()
- antisymmetrize() is slightly faster
- ensure_if_modifiable() clears the hash_calculated flag

22 years ago- Updated to reflect new version and split-off of ginaccint. release_0-9-0
Richard Kreckel [Thu, 7 Jun 2001 22:48:08 +0000 (22:48 +0000)]
- Updated to reflect new version and split-off of ginaccint.

22 years ago- don't recomment EGCS any more, it's crappy.
Richard Kreckel [Thu, 7 Jun 2001 22:27:59 +0000 (22:27 +0000)]
- don't recomment EGCS any more, it's crappy.

22 years ago- insert date.
Richard Kreckel [Thu, 7 Jun 2001 22:27:15 +0000 (22:27 +0000)]
- insert date.

22 years ago- some concept index tags.
Richard Kreckel [Thu, 7 Jun 2001 22:26:43 +0000 (22:26 +0000)]
- some concept index tags.

22 years ago- matrix::pow(): omit last big multiplication if it's not needed.
Richard Kreckel [Thu, 7 Jun 2001 22:26:16 +0000 (22:26 +0000)]
- matrix::pow(): omit last big multiplication if it's not needed.

22 years agobuilding DVI/PS in separate directory now works
Christian Bauer [Thu, 7 Jun 2001 22:02:15 +0000 (22:02 +0000)]
building DVI/PS in separate directory now works

22 years ago- maximum number of arguments for lists bumped to 16 (for making 4x4 matrices)
Christian Bauer [Thu, 7 Jun 2001 20:09:21 +0000 (20:09 +0000)]
- maximum number of arguments for lists bumped to 16 (for making 4x4 matrices)
- added non-const operator() for setting matrix elements

22 years agoadded section about matrices
Christian Bauer [Thu, 7 Jun 2001 20:07:54 +0000 (20:07 +0000)]
added section about matrices

22 years agousing () operator for setting matrix elements
Christian Bauer [Thu, 7 Jun 2001 20:07:44 +0000 (20:07 +0000)]
using () operator for setting matrix elements

22 years agodocumentation update
Christian Bauer [Thu, 7 Jun 2001 20:07:32 +0000 (20:07 +0000)]
documentation update

22 years ago- mentioned that powers of matrices are not automatically expanded
Christian Bauer [Wed, 6 Jun 2001 22:17:38 +0000 (22:17 +0000)]
- mentioned that powers of matrices are not automatically expanded
- fixed typos

22 years ago- Removed obsolete (even for Cint!) malloc_alloc template arguments.
Richard Kreckel [Wed, 6 Jun 2001 22:16:02 +0000 (22:16 +0000)]
- Removed obsolete (even for Cint!) malloc_alloc template arguments.

22 years ago- Removed cint subdir for good because the new separate package seems to
Richard Kreckel [Wed, 6 Jun 2001 22:15:21 +0000 (22:15 +0000)]
- Removed cint subdir for good because the new separate package seems to
  work properly now (after some longish argument with makecint).

22 years ago- config.*: updated from FSF.
Richard Kreckel [Wed, 6 Jun 2001 22:15:18 +0000 (22:15 +0000)]
- config.*: updated from FSF.
- Removed cint subdir for good because the new separate package seems to
  work properly now (after some longish argument with makecint).

22 years ago- inserted a couple of missing namepace std:: resolutions.
Richard Kreckel [Tue, 5 Jun 2001 21:07:04 +0000 (21:07 +0000)]
- inserted a couple of missing namepace std:: resolutions.

22 years ago- document: collect's argument must be expanded.
Richard Kreckel [Tue, 5 Jun 2001 20:59:12 +0000 (20:59 +0000)]
- document: collect's argument must be expanded.

22 years ago- Small LaTeX output fixes as suggested by Stefan.
Richard Kreckel [Tue, 5 Jun 2001 15:26:35 +0000 (15:26 +0000)]
- Small LaTeX output fixes as suggested by Stefan.

22 years ago- fixed typos
Christian Bauer [Tue, 5 Jun 2001 15:18:29 +0000 (15:18 +0000)]
- fixed typos
- c*M^n with M being a matrix wasn't correctly evalm()'d for n!=1

22 years ago- ++version_major.
Richard Kreckel [Tue, 5 Jun 2001 13:55:01 +0000 (13:55 +0000)]
- ++version_major.
- added matrix::pow() to handle integer exponents with the least amount of
  multiplications possible and...
- ...added power::evalm() to actually use it.
- classhierarchy.fig: add class wildcard.
- some cleanups.

22 years ago- list delimiters are now { } and matrix delimiters are [ ]
Christian Bauer [Fri, 1 Jun 2001 22:50:33 +0000 (22:50 +0000)]
- list delimiters are now { } and matrix delimiters are [ ]
- added evalm() function

22 years ago- in the output, list delimiters are now { } and matrix delimiters are [ ]
Christian Bauer [Fri, 1 Jun 2001 22:50:00 +0000 (22:50 +0000)]
- in the output, list delimiters are now { } and matrix delimiters are [ ]
- added evalm() method for evaluating sums and products of matrices
- added map() method for applying a function to subexpressions
- power(noncommut, posint) automatically expands the product, removed ncpow()

22 years ago- list delimiters are now { }, matrix delimiters are now [ ]
Christian Bauer [Fri, 1 Jun 2001 22:47:03 +0000 (22:47 +0000)]
- list delimiters are now { }, matrix delimiters are now [ ]
- added an example for evalm()
- removed ncpow()

22 years agoadded test for evalm()
Christian Bauer [Fri, 1 Jun 2001 22:46:02 +0000 (22:46 +0000)]
added test for evalm()

22 years agodocumentation update
Christian Bauer [Fri, 1 Jun 2001 22:45:53 +0000 (22:45 +0000)]
documentation update

22 years ago- added numer_denom()
Christian Bauer [Fri, 1 Jun 2001 18:37:13 +0000 (18:37 +0000)]
- added numer_denom()
- minor cleanups

22 years ago- added numer_denom() to get numerator and denominator in one pass
Christian Bauer [Fri, 1 Jun 2001 18:37:00 +0000 (18:37 +0000)]
- added numer_denom() to get numerator and denominator in one pass
- symmetrize() and antisymmetrize() are available as ex methods
- minor cleanups

22 years ago- mentioned numer_denom()
Christian Bauer [Fri, 1 Jun 2001 18:36:10 +0000 (18:36 +0000)]
- mentioned numer_denom()
- symmetrize() and antisymmetrize() are available as ex methods

22 years agodocumentation update
Christian Bauer [Fri, 1 Jun 2001 18:35:32 +0000 (18:35 +0000)]
documentation update

22 years agodeterminant_algo::laplace sometimes had the sign wrong because of the change
Christian Bauer [Thu, 31 May 2001 17:14:25 +0000 (17:14 +0000)]
determinant_algo::laplace sometimes had the sign wrong because of the change
to permutation_sign()

22 years ago- permutation_sign() and shaker_sort() are no longer inlined
Christian Bauer [Wed, 30 May 2001 22:38:28 +0000 (22:38 +0000)]
- permutation_sign() and shaker_sort() are no longer inlined
- symmetrize() no longer limited to 32-bit integers

22 years ago- dirac_trace() is twice as fast
Christian Bauer [Mon, 28 May 2001 22:16:47 +0000 (22:16 +0000)]
- dirac_trace() is twice as fast
- permutation_sign() uses shaker sort
- shaker_sort() doesn't require less-than comparable iterators any more

22 years agodocumented symmetrize() and antisymmetrize()
Christian Bauer [Mon, 28 May 2001 22:15:05 +0000 (22:15 +0000)]
documented symmetrize() and antisymmetrize()

22 years ago- added symmetrize() and antisymmetrize() functions
Christian Bauer [Sun, 27 May 2001 19:53:13 +0000 (19:53 +0000)]
- added symmetrize() and antisymmetrize() functions
- generalized permutation_sign() template
- moved shaker_sort() template to utils.h

22 years ago- renamed ncpower() to ncpow()
Christian Bauer [Fri, 25 May 2001 22:32:48 +0000 (22:32 +0000)]
- renamed ncpower() to ncpow()
- canonicalize_clifford() uses to_rational() now
- minor cleanups

22 years agomentioned ncpow()
Christian Bauer [Fri, 25 May 2001 22:29:34 +0000 (22:29 +0000)]
mentioned ncpow()

22 years agofixed typos
Christian Bauer [Fri, 25 May 2001 21:12:03 +0000 (21:12 +0000)]
fixed typos

22 years agoadded description of wildcards
Christian Bauer [Fri, 25 May 2001 20:51:49 +0000 (20:51 +0000)]
added description of wildcards

22 years ago- removed debugging code in match()
Christian Bauer [Fri, 25 May 2001 20:51:35 +0000 (20:51 +0000)]
- removed debugging code in match()
- added ex::match(const ex & pattern) which doesn't take a list as second
  argument
- expairseq::subs() substitutes in recombined pairs when the pattern is a
  product or a power; this is necessary for things like subs(2*x^2,x^2==3)
  because the first expression is mul(<x,2>,2) and only substituting in the
  "rest" part of the expair is not enough

22 years agoadded section "Pattern matching and advanced substitutions" which covers
Christian Bauer [Fri, 25 May 2001 20:47:43 +0000 (20:47 +0000)]
added section "Pattern matching and advanced substitutions" which covers
match(), has() and subs()

22 years agoadded wildcard and spinormetric classes to check
Christian Bauer [Fri, 25 May 2001 20:47:15 +0000 (20:47 +0000)]
added wildcard and spinormetric classes to check

22 years ago- fix LaTeX-output bug reported by Stefan, remove obsolete has(matrix,ex).
Richard Kreckel [Thu, 24 May 2001 19:07:47 +0000 (19:07 +0000)]
- fix LaTeX-output bug reported by Stefan, remove obsolete has(matrix,ex).

22 years ago- first implementation of pattern matching
Christian Bauer [Thu, 24 May 2001 00:43:11 +0000 (00:43 +0000)]
- first implementation of pattern matching

22 years ago- Clifford traces of many gammas are a lot faster now (especially with gamma5)
Christian Bauer [Mon, 21 May 2001 22:49:16 +0000 (22:49 +0000)]
- Clifford traces of many gammas are a lot faster now (especially with gamma5)
- new simplifications for color classes:
    d.abc T.b T.c = 5/6 T.a
    f.abc T.b T.c = 3/2 I T.a
- added indexed::has_dummy_index_for() member function

22 years ago- mentioned dummy index renaming
Christian Bauer [Mon, 21 May 2001 22:46:34 +0000 (22:46 +0000)]
- mentioned dummy index renaming
- Clifford and color class description slightly updated

22 years agoupdated for changes in Clifford/color classes
Christian Bauer [Mon, 21 May 2001 22:45:48 +0000 (22:45 +0000)]
updated for changes in Clifford/color classes

22 years agoreplaced "precedence" static member variable by virtual precedence() function
Christian Bauer [Sat, 19 May 2001 21:37:44 +0000 (21:37 +0000)]
replaced "precedence" static member variable by virtual precedence() function

22 years ago- adjust preprocessor Cint-checks to the actual situation.
Richard Kreckel [Sat, 19 May 2001 19:11:45 +0000 (19:11 +0000)]
- adjust preprocessor Cint-checks to the actual situation.

22 years ago- dummy index renamer didn't account for internal dummy indices of objects
Christian Bauer [Sat, 19 May 2001 00:42:07 +0000 (00:42 +0000)]
- dummy index renamer didn't account for internal dummy indices of objects
  in products; a~mu.mu-a~nu.nu gets simplified to 0 now
- made a little more use of STL facilities for exvectors, especially in the
  indexed stuff; append_exvector_to_exvector() and index_set_difference() are
  gone and utils.h defines the functors ex_is_less and ex_is_equal

22 years agoadded one more check
Christian Bauer [Sat, 19 May 2001 00:39:55 +0000 (00:39 +0000)]
added one more check

22 years agodocumentation update
Christian Bauer [Sat, 19 May 2001 00:39:49 +0000 (00:39 +0000)]
documentation update

22 years ago- simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
Christian Bauer [Fri, 18 May 2001 00:53:07 +0000 (00:53 +0000)]
- simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
  simplified to "2*a.i*a.i" (or 2*a.j*a.j, you can't know which one)
- fixed possible crash when calling subs() on expressions with non-commutative
  products
- added canonicalize_clifford()

22 years agocorrected typo
Christian Bauer [Fri, 18 May 2001 00:52:12 +0000 (00:52 +0000)]
corrected typo

22 years agoadded checks for dummy index renaming and canonicalize_clifford()
Christian Bauer [Fri, 18 May 2001 00:52:05 +0000 (00:52 +0000)]
added checks for dummy index renaming and canonicalize_clifford()

22 years agodocumentation update
Christian Bauer [Fri, 18 May 2001 00:51:50 +0000 (00:51 +0000)]
documentation update

22 years ago- Missed to #include "config.h" for HAVE_SSTREAM.
Richard Kreckel [Thu, 17 May 2001 20:35:36 +0000 (20:35 +0000)]
- Missed to #include "config.h" for HAVE_SSTREAM.

22 years agofixed bug in mul::print with print_csrc
Christian Bauer [Wed, 16 May 2001 20:04:51 +0000 (20:04 +0000)]
fixed bug in mul::print with print_csrc

22 years ago- Fix thinko in csgn_eval().
Richard Kreckel [Tue, 15 May 2001 23:26:41 +0000 (23:26 +0000)]
- Fix thinko in csgn_eval().

22 years ago- Indentation Fix.
Richard Kreckel [Tue, 15 May 2001 23:26:21 +0000 (23:26 +0000)]
- Indentation Fix.

22 years ago- added patch contributors, added remark to mailing lists. release_0-8-3
Richard Kreckel [Fri, 11 May 2001 23:15:55 +0000 (23:15 +0000)]
- added patch contributors, added remark to mailing lists.

22 years ago- forgot to close #97136.
Richard Kreckel [Fri, 11 May 2001 20:29:03 +0000 (20:29 +0000)]
- forgot to close #97136.

22 years ago- applied `dch -i'.
Richard Kreckel [Fri, 11 May 2001 19:43:49 +0000 (19:43 +0000)]
- applied `dch -i'.

22 years ago- Renamed our detected libreadline version in order to avoid conflicts
Richard Kreckel [Fri, 11 May 2001 19:43:31 +0000 (19:43 +0000)]
- Renamed our detected libreadline version in order to avoid conflicts
  with Chat Rameys (still broken) macros and future CPP definitions in
  libreadline's headers.  Oh what a pain in the ass!

22 years ago- dirac_trace() takes an optional third argument: the expression to use for
Christian Bauer [Fri, 11 May 2001 18:05:32 +0000 (18:05 +0000)]
- dirac_trace() takes an optional third argument: the expression to use for
  the trace of the unit matrix
- more conventional factor I in gamma5

22 years agodirac_trace() now takes three arguments
Christian Bauer [Fri, 11 May 2001 18:04:35 +0000 (18:04 +0000)]
dirac_trace() now takes three arguments

22 years agoadded missing factor I in gamma5 trace
Christian Bauer [Fri, 11 May 2001 18:04:24 +0000 (18:04 +0000)]
added missing factor I in gamma5 trace

22 years ago- Added inclusion of <sstream> or <strstream> since <ginac/ginac.h> now
Richard Kreckel [Fri, 11 May 2001 14:51:40 +0000 (14:51 +0000)]
- Added inclusion of <sstream> or <strstream> since <ginac/ginac.h> now
  does not do so any more.

22 years ago* Zapped header dependency from <strstream>.
Richard Kreckel [Fri, 11 May 2001 14:16:11 +0000 (14:16 +0000)]
* Zapped header dependency from <strstream>.

22 years ago* comparison operators should return const relationals.
Richard Kreckel [Fri, 11 May 2001 14:15:33 +0000 (14:15 +0000)]
* comparison operators should return const relationals.

22 years ago* node assignment was not returning *this.
Richard Kreckel [Fri, 11 May 2001 14:14:53 +0000 (14:14 +0000)]
* node assignment was not returning *this.

22 years ago- color and clifford classes are quite functional now
Christian Bauer [Thu, 10 May 2001 20:41:24 +0000 (20:41 +0000)]
- color and clifford classes are quite functional now
- new "spinidx" class for dotted/undotted indices
- predefined spinor metric tensor (created by spinor_metric())