1 This file records noteworthy changes.
3 1.3.0 (19 October 2004)
4 * The Clifford classes have been generalized to allow working with Clifford
5 algebras generated by arbitrary symmetric tensors or matrices. Also, a lot
6 of new functions for Clifford algebras have been added, including
7 automorphisms and Moebius transformations. [V. Kisil]
8 * Added some basic symbolic and numeric integration facilities. [C. Dams]
9 * The multiple polylogarithm Li() now evaluates numerically for arbitrary
11 * New functions G(a,y) and G(a,s,y) added (another notation for multiple
14 1.2.4 (12 October 2004)
15 * Added ex::unitcontprim() to compute the unit, content, and primitive parts
16 of a polynomial in one go.
17 * binomial(n, k) evaluates for non-integer arguments n.
18 * Li(2,x) now evaluates for +-I.
20 * Fixed bug in Li(n,x) (if Li(2,x) was calculated with high precision the
21 enlargement of the look-up table caused a segmentation fault).
22 * Fixed another bug in the series expansion of powers, and a bug in
25 1.2.3 (13 August 2004)
26 * Added variants of dirac_trace() and color_trace() that take the trace over
27 more than one representation label by specifying a set or list of labels.
28 * diracgamma::contract_with() uses Chisholm identities in 4 dimensions to
29 produce more compact results.
30 * Fixed a bug in the series expansion of powers.
31 * The --enable-html-doc and --enable-ps-doc configure options are gone.
32 Documentation in HTML, DVI, PostScript, and PDF formats is now built with
33 "make html", "make dvi", "make ps", and "make pdf", respectively. The only
34 documentation that gets built by default are the tutorial in .info format
38 * Added const_preorder_iterator and const_postorder_iterator classes (and
39 associated methods ex::pre/postorder_begin/end()) providing tree traversal
41 * Fixed the LaTeX output of the varidx class.
42 * Fixed bugs in series expansion and complex conjugation.
43 * (p.i*p.i).get_free_indices() returns the correct result () instead of (.i).
44 * Symbolic functions without any eval(), evalf() etc. functions now work
46 * integer_content(), content(), and primpart() now also work for polynomials
47 with rational coefficients (calculating the LCM of coefficients'
49 * Added method matrix::rank().
50 * Added function resultant().
51 * Added integer_content() function to ginsh.
54 * Fixed infinite recursion in atan2_evalf() and improved atan2_eval().
55 * Added automatic evaluations for trigonometric functions with negative
56 arguments (e.g. sin(-2) -> -sin(2)).
57 * Fixed a static initialization order goof-up.
58 * Fixed various bugs in series expansion.
61 * Added a structure<T> template class for the easy creation of user-defined
63 * Added support for (acyclic) visitors, to allow cleaner implementations of
65 * Added a const_iterator class that can be used instead of op()/nops().
66 * Completely revamped the implementation of expression output. It is now
67 possible to add new output formats, to change the behavior of predefined
68 formats at run-time, and to have different output styles for algebraic
70 * Symbols can be made non-commutative.
71 * Added a method ex::conjugate() and a function conjugate() for complex
72 conjugation. Symbols can be declared as real or complex-valued.
73 * Improved the speed of subs(), normal(), to_rational() and to_polynomial()
74 by the use of maps instead of lists. The old forms
75 subs(const lst & ls, const lst & lr, unsigned options)
76 to_rational/to_polynomial(lst & repl)
77 are still available for compatibility, but using the new forms
78 subs(const exmap & m, unsigned options)
79 to_rational/to_polynomial(exmap & repl)
80 is more efficient, especially when the number of replacements is large.
81 * quo(), rem(), prem(), sprem(), decomp_rational(), unit(), content(),
82 primpart() and matrix::charpoly() now take a "const ex &" instead of a
84 * Redundant expressions (two ex pointing to different objects are found to be
85 equal in compare()) are now actively deleted/fused to conserve memory and
86 speed up subsequent comparisons. This behavior can be suppressed on a
87 per-object level with status_flags::not_shareable. Lists and matrices are
88 not shareable by default.
89 * Lots of internal streamlining and optimizations.
90 * Caveats for class implementors:
91 - basic::copy() and basic::destroy() are gone; classes derived from
92 basic can use the defaults for the assignment operator and copy
94 - basic::subs(), basic::normal(), basic::to_rational() and
95 basic::to_polynomial() take 'exmap' objects instead of lists.
96 - basic::subs() now descends into subexpressions (if accessible via
97 nops()/op()/let_op()). If you have a custom implementation of subs()
98 that calls basic::subs() after substituting subexpressions, this needs
99 to be changed to a call to subs_one_level().
100 - lst::thislst() and exprseq::thisexprseq() renamed to thiscontainer().
101 - thiscontainer() and associated constructors now take a std::auto_ptr.
102 - Overloading basic::print() is now deprecated. You should use
103 print_func<>() class options instead.
105 1.1.7 (11 March 2004)
106 * Fixed a bug in canonicalize_clifford().
107 * Series expansion now works predictably. All terms with the exponent of the
108 expansion variable smaller than the given order are calculated exactly. If
109 the series is not terminating, the Order function is (at least) of the given
112 1.1.6 (22 January 2004)
113 * Added a function option "dummy()" which means "no options". This simplifies
114 the implementation of symbolic functions which are not to be further
116 * Removed a bug in the numerical evaluation of Li() that caused the system
117 to hang for certain parameter combinations.
118 * Fixed a bug in the calculation of hash values for indices that could lead
119 to wrong results or bogus error messages from simplify_indexed().
120 * Fixed a bug in the evaluation of harmonic polylogarithms for complex
121 arguments with positive imaginary part.
123 1.1.5 (5 November 2003)
124 * Harmonic polylogarithms now numerically evaluate for arbitrary arguments
125 (parameter must still be positive integers).
126 * The zeta function now can also be given a lst as a parameter in which case
127 it becomes a multiple zeta value. The use of mZeta is deprecated.
128 * The order of parameters for the multiple polylogarithm has been corrected.
129 * Documentation for the nested sums functions zeta, harmonic polylog, multiple
130 polylog, etc. has been added.
132 1.1.4 (17 October 2003)
133 * Lists and matrices can now be initialized from comma-separated lists of
134 expressions, like this:
144 This is both faster and produces much smaller code than the old constructors
145 lst(ex, ex, ...) and matrix(unsigned, unsigned, lst), especially in the case
146 of matrices, and is now the recommended way to create these objects.
147 * The function mZeta now evaluates much faster for arbitrary parameters. The
148 harmonic and multiple polylogarithms evaluate considerably faster and check
149 for convergence. The order of parameters for the harmonic polylogarithm
152 1.1.3 (22 August 2003)
153 * Added new symbolic functions for better integration with nestedsums:
154 (multiple) polylogarithm Li(), Nielsen's generalized polylogarithm S(),
155 harmonic polylogarithm H(), and multiple zeta value mZeta().
156 * New exhashmap<T> template intended as a drop-in replacement for
157 std::map<ex, T, ex_is_less> using GiNaC's hashing algorithms.
159 1.1.2 (11 August 2003)
160 * Fixed a bug in the unarchiving of sums and products: terms were not
161 reordered in a canonical way.
162 * Fixed a bug in normal()/numer_denom(): denominator was not made unit
163 normal if it was a simple number.
164 * Improved the speed of subs() in some cases.
167 * lst (and exprseq) provide iterators for read-only element access. For
168 sequential access this is one order faster than using op().
169 * Implemented relational::subs() (this was done in 1.0.9 but inadvertently
170 omitted from the 1.1 branch).
171 * pole_error and do_taylor are available to library users.
172 * Added on-line help and Tab-completion for print(), iprint(), print_latex()
173 and print_csrc() in ginsh.
176 * Removed deprecated macros is_ex_a, is_ex_exactly_a and friends for good.
177 * The scalar_products mechanism allows the specification of an index dimension.
178 * Removed dirac_gamma6/7().
179 * Added ex::to_polynomial().
180 * subs() accepts an optional "options" argument. The option
181 subs_option::subs_algebraic enables "smart" substitutions in products and
183 * Added stream manipulators "dflt", "latex", "python", "python_repr", "tree",
184 "csrc", "csrc_float", "csrc_double", "csrc_cl_N", "index_dimensions" and
185 "no_index_dimensions" to control the output format. Calling basic::print()
186 directly is now deprecated.
187 * Made the hashing more simple and efficient.
188 * Caveats for class implementors:
189 - basic::subs(): third argument changed from "bool" to "unsigned"
190 - unarchiving constructor and basic::unarchive(): "const" removed from
192 - basic::let_op() should only be implemented if write access to
193 subexpressions is desired
194 - simplify_ncmul() renamed to eval_ncmul()
195 - simplified_ncmul() renamed to hold_ncmul()
196 - nonsimplified_ncmul() renamed to reeval_ncmul()
198 1.0.14 (1 March 2003)
199 * Improved the C-source output: complex numbers are printed correctly (using
200 the STL complex<> template or cln::complex()), rational numbers use cl_RA()
201 in the CLN output, and small integers are printed in a more compact format
202 (e.g. "2.0" instead of "2.0000000e+00").
203 * function_options::set_return_type() and function_options::do_not_evalf_params()
206 1.0.13 (27 January 2003)
207 * Contracting epsilon tensors with Euclidean indices now works.
208 * Improved dummy index symmetrization in sums.
209 * Added dirac_gammaL/R(), which can be used instead of dirac_gamma6/7()
210 but are single objects, to allow for a more compact notation of Dirac
212 * Powers with negative numeric exponents are printed as fractions in the
214 * Added symbolic_matrix() for the convenient creation of matrices filled
216 * Added collect_common_factors() which collects common factors from the
218 * simplify_indexed() converts "gamma~mu*p.mu" to "p\".
220 1.0.12 (30 October 2002)
221 * Fixed a bug in power::expand() that could produce invalid expressions.
222 * The input parser no longer ignores extra data following accepted input.
223 * Improved the CLN C-source output (integers are printed as integers, and
224 floating point numbers include the precision).
225 * Fixed a problem in the LaTeX-output of negative fractions.
226 * Added print_latex() and print_csrc() to ginsh.
227 * The sprem() function is now public.
229 1.0.11 (18 September 2002)
230 * Fixed a possible memory corruption in contractions of indexed objects with
231 delta or metric tensors.
232 * Computing the derivative of a power series object with respect to a symbol
233 that is not the expansion variable now works correctly.
234 * Several bugfixes in code generation.
236 1.0.10 (24 July 2002)
237 * Powers of indexed objects are now parenthesized correctly in LaTeX output.
238 * Input parser handles indices (they have to be specified in the same list
240 * Added some limited support for subspaces in the idx and tensor classes.
241 * Fixed a bug in canonicalize() (antisymmetric canonicalization of an already
242 sorted list containing two or more equal objects failed to return 0).
245 * simplify_indexed() now raises/lowers dummy indices to canonicalize the index
246 variance. This allows some simplifications that weren't possible before,
247 like eps~a.b~c~d*X.a*X~b -> 0 and X.a~a-X~a.a -> 0.
248 * Implemented relational::subs().
249 * Fixed bug in simplify_ncmul() for clifford objects.
251 1.0.8 (31 March 2002)
252 * Improvements in memory usage of the expand() methods.
254 1.0.7 (18 March 2002)
255 * Fixed LaTeX output of indexed and matrix objects.
256 * Fixed matrix::pow(n) for n==0 and added helper functions to create unit
257 matrices "ex unit_matrix(unsigned, unsigned)".
260 * "(x+1).subs(x==x-1)" now returns the correct result "x" instead of "x-1".
262 1.0.5 (27 January 2002)
263 * (l)degree(s), coeff(s, n) and collect(s) were extended to accept expressions
264 of any class (except add/mul/ncmul/numeric) for "s". They should even work
265 if "s" is a "power" object, as long as the exponent is non-integer, but with
266 some limitations. For example, you can "collect(a*2^x+b*2^x, 2^x)" to get
267 "(a+b)*2^x", but "degree(2^(3*x), 2^x)" yields 0 instead of 3).
268 * Fixed a small output bug.
270 1.0.4 (24 January 2002)
271 * Speedup in expand().
272 * Faster Bernoulli numbers (Markus Nullmeier).
273 * Some minor bugfixes and documentation updates.
275 1.0.3 (21 December 2001)
276 * Fixed a bug where quo() would call vector::reserve() with a negative
278 * Fix several bugs in code generation.
280 1.0.2 (19 December 2001)
281 * Input parser recognizes "sqrt()", which is also used in the output.
282 * divide(a,b,q) only modifies q if the division succeeds; also, divide(a,b,a)
284 * Fixed small bug in dummy index renaming which could cause it to not
285 recognize renamable indices in some cases.
286 * power::degree() and power::ldegree() throw an exception when encountering
287 a non-integer exponent.
288 * Add output-support for Python bindings.
290 1.0.1 (22 November 2001)
291 * Function sqrfree() handles a few more cases now.
292 * Class relational has real canonical ordering now.
293 * Handle obscene libreadline version numbers when building ginsh.
295 1.0.0 (6 November 2001)
296 * Some internal reorganization resulting in a general speed-up.
297 * The last 3 evaluated expressions in ginsh are now referred to with the
298 tokens '%', '%%' and '%%%'. The old '"', '""' and '"""' remain for
299 compatibility but may be removed in a future version of GiNaC.
301 0.9.4 (20 September 2001)
302 * Functions have better support for external scripting languages.
303 * Interface cleanups and bugfixes.
304 * Fix silly bug in evalf() that prevented things like 2^Pi being computed.
306 0.9.3 (16 August 2001)
307 * series expansion now much more consistent for small order expansion.
308 * lsolve() accepts algorithmic hint as parameter.
311 * Epsilon tensor is more functional.
312 * simplify_indexed() is better at detecting expressions that vanish for
314 * Several little bugfixes and consistency enhancements.
317 * Ctors of class numeric are not explicit any more. All built-in callers for
318 pseudofunctions are now templated and default to ex arguments which relaxes
319 the need for explicit ctors.
320 * New functions/methods:
322 - remove_first(), remove_last(), sort() and unique() for lists
323 - symmetrize_cyclic()
325 * Instead of just totally symmetric or antisymmetric, complex symmetries
326 can now be defined for indexed objects. Symmetries are described by a
327 tree of "symmetry" objects that is constructed with the sy_none(),
328 sy_symm(), sy_anti() and sy_cycl() functions. The symmetry of a function
329 with respect to its arguments can also be defined (this is currently
330 only used for the Beta function).
331 * Generalized map() to take a function object instead of a function pointer.
332 This allows passing an arbitrary number of additional state to the
333 function being called.
334 * color_trace(), dirac_trace(), diff(), expand(), evalf() and normal() work
335 better with container classes, e.g. using color_trace() on a relation will
336 take the trace on both sides, using diff() on a matrix differentiates every
338 * diff() works properly with non-commutative products and indexed objects.
339 * New option flag "expand_function_args" for expand().
340 * Supplement some (now deprecated) macros by inlined template functions:
341 - is_of_type(foo, type) -> is_a<type>(foo)
342 - is_ex_of_type(foo, type) -> is_a<type>(foo)
343 - is_exactly_of_type(foo, type) -> is_exactly_a<type>(foo)
344 - is_ex_exactly_of_type(foo, type) -> is_exactly_a<type>(foo)
345 - ex_to_foobar(baz) -> ex_to<foobar>(baz)
346 * rem(c, p[x], x) (c: numeric, p[x]: polynomial) erroneously returned p[x]
348 * Small bugfixes in pattern matching.
349 * Updated libtool to version 1.4.
352 * In the output and in ginsh, lists are now delimited by { } braces, and
353 matrices are delimited by single [ ] brackets.
354 * simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
355 simplified to "2*a.i*a.i" (or "2*a.j*a.j").
356 * New functions/methods:
357 - canonicalize_clifford() (helpful when comparing expressions containing
359 - symmetrize() and antisymmetrize()
360 - numer_denom() (return numerator and denominator in one call)
361 - map() (apply function to subexpressions)
362 - evalm() (evaluate sums, products and integer powers of matrices)
363 * Added a new function match() for performing pattern matching. subs() and
364 has() also accept patterns as arguments. A pattern can be any expression,
365 optionally containing wildcard objects. These are constructed with the
366 call "wild(<unsigned>)" and are denoted as "$0", "$1" etc. in the output
368 * Positive integer powers of non-commutative expressions (except matrices)
369 are automatically expanded.
370 * Removed cint subdirectory, ginaccint is a separate package now due to
371 packaging considerations.
372 * Several little bugfixes.
375 * color and clifford classes are functional and documented.
376 * New "spinidx" class for dotted/undotted indices.
377 * Predefined spinor metric tensor (created by spinor_metric()).
378 * Symbols can have a LaTeX name, e.g. symbol s("s", "\\sigma");
379 * LaTeX output of indexed objects is much nicer.
380 * Fixed some build problems (with recent libreadline).
381 * Semantics of arithmetic operators now follows the C++ rules more strictly.
383 0.8.2 (24 April 2001)
384 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() work with
385 non-symbols as the second argument in ginsh.
386 * the argument to collect() can be a list of objects in which case the
387 result is either a recursively collected polynomial, or a polynomial in
388 a distributed form with terms like coeff*x1^e1*...*xn^en, as specified by
389 the second argument to collect().
390 * Several bugfixes (including a nasty memory leak in .normal()).
391 * class matrix: solve() doesn't call algorithms redundantly any more and
392 inverse() falls back to solve() which works in more general cases.
394 0.8.1 (16 April 2001)
395 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() can now
396 be used with constants, functions and indexed expressions as well, so you
397 can use it to collect by powers of Pi or sin(x), or to find the coefficient
400 - it only works with symbols, constants, functions and indexed expressions,
401 trying to find the coefficient of, e.g., "x^2" or "x+y" won't work;
402 - it does not know about dummy index summations; the coefficient of
403 gamma~0 in p.mu*gamma~mu should be p.0 but is returned as 0;
404 - using coeff(), tcoeff(), lcoeff() or collect() on elements of
405 noncommutative products might return wrong or surprising results.
406 * subs() no longer only substitutes symbols and indices but performs a more
407 general "syntactic substitution", i.e. it substitutes whole objects in sub-
408 expressions. You can subs((a+b)^2,a+b==3) and get 9, but subs(a+b+c,a+b==3)
411 - substituting numerics (subs(expr, 2==4)) will not replace then in all
412 occurences; in general, you shouldn't substitute numerics, though.
413 * Added preliminary (re)implementations of color and clifford classes.
414 * simplify_indexed(): contraction of symmetric and antisymmetric tensors
416 * Replaced the various print*() member functions by a single print() that
417 takes a print_context object that determines the output formatting. This
418 should make it easier to add more output types, such as LaTeX output,
419 which is based on work by Stefan Weinzierl.
420 * Added functions to retrieve the properties stored in archive objects
421 outside of unarchive() (for printing or debugging purposes).
422 * Some bugfixes (indexed objects, archive writing).
423 * .collect() on non-polynomials is now algebraically correct.
425 0.8.0 (24 March 2001)
426 * Complete revamp of indexed objects. Instead of multiple classes for
427 indexed things and their indices there is now only one "indexed" class
428 and two types of indices: "idx" for simple indices and "varidx" for
429 indices with variance. There are predefined delta, epsilon and metric
430 tensors, and a function simplify_indexed() that performs canonicalization
431 and dummy index summations. Matrix objects can be indexed for doing simple
433 * Added an option "expand_indexed" to expand() to perform expansion of
434 indexed objects like (a+b).i -> a.i + b.i
435 * Renamed get_indices() to get_free_indices(), which no longer returns
436 dummy indices and checks the consistency of indices in sums.
437 * sqrfree() factorization fixed and improved syntactically.
438 * subs() works on matrices.
439 * Matrices can be constructed from flat list of elements; diagonal matrices
440 can be constructed from list of diagonal elements with diag_matrix().
441 * Fixed memory leak in expand().
442 * Operator% for objects of class ncmul has gone. Use operator* now for that
443 case too, which is much more natural.
445 0.7.3 (28 February 2001)
446 * Several bugfixes and minor performance tunings.
447 * Added a section to the tutorial about adding new algebraic classes to GiNaC.
448 * Closed many in-source documentation gaps.
450 0.7.2 (17 February 2001)
451 * Several bugfixes in power series expansion, one of them critical.
453 0.7.1 (7 February 2001)
454 * Fix problems with Cint that were caused by CLN's overloaded operator new.
455 * Fix compilation errors with GCC3.
456 * normal() handles large sums of fractions better and normalizes the exponent
457 of power expressions.
458 * expand() always expands the exponent and transforms x^(a+b) -> x^a*x^b.
459 * Some bugfixes of series expansion around branch cuts of special functions.
461 0.7.0 (15 December 2000)
462 * Requires CLN 1.1 now. Class numeric doesn't use an indirect pointer to the
463 actual representation any more. This is a speedup.
464 * mul::expand() was reengineered to not allocate excess temporary memory.
465 * Non-integer powers of a symbol are treated as constants by (l)degree() and
466 coeff(). Using these functions on an expression containing such powers used
467 to fail with an internal error message. The side-effect is that collect()
468 can be used on expressions which are not polynomials.
469 * Added a man page for the ginac-config script.
470 * Ctor of numeric from char* honors Digits.
472 0.6.4 (10 August 2000)
473 * Complete revamp of methods in class matrix. Some redundant (and poor)
474 implementations of elimination schemes were thrown out. The code is now
475 highly orthogonal, more flexible and much more efficient.
476 * Some long standing and quite nasty bugs were discovered and fixed in the
477 following functions: add::normal(), heur_gcd(), sr_gcd() and Order_eval().
480 * Derivatives are now assembled in a slightly different manner (i.e. they
481 might 'look' different on first sight). Under certain circumstances this
482 can result in a dramatic speedup because it gives hashing a better chance,
483 especially when computing higher derivatives.
484 * Some series expansions of built-in functions have been reengineered.
485 * The algorithm for computing determinants can be chosen by the user. See
486 ginac/flags.h and ginac/matrix.h.
487 * The Dilogarithm (Li2) now has floating point evaluation, derivative and a
488 proper series expansion.
489 * Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
490 * Some minor bugfixes, one major lsolve()-bugfix and documentation updates.
493 * ginaccint.bin is now launched by a binary program instead of by a scripts.
494 This allows us to write #!-scripts. A small test suite for GiNaC-cint was
496 * Several minor bugfixes.
499 * Cleanup in the interface to Cint. The required version is now Cint 5.14.38.
500 * Several bugfixes in target install.
503 * IMPORTANT: Several interface changes make programs written with GiNaC
504 much clearer but break compatibility with older versions:
505 - f(x).series(x,p[,o]) -> f(x).series(x==p,o)
506 - series(f(x),x,p[,o]) -> series(f(x),x==p,o)
507 - gamma() -> tgamma() (The true Gamma function, there is now also
508 log(tgamma()), called lgamma(), in accord with ISO/IEC 9899:1999.)
509 - EulerGamma -> Euler
510 * #include'ing ginac.h defines the preprocessor symbols GINACLIB_MAJOR_VERSION,
511 GINACLIB_MINOR_VERSION, and GINACLIB_MICRO_VERSION with the respective GiNaC
512 library version numbers.
513 * Expressions can be constructed from strings like this:
514 ex e("2*x+y", lst(x, y));
515 * ex::to_rational() provides a way to extend the domain of functions like
516 gcd() and divide() that only work on polynomials or rational functions (the
517 good old ex::subs() method reverses this process)
518 * Calling diff() on a function that has no derivative defined returns the
519 inert derivative function "Derivative".
520 * Several new timings in the check target. Some of them may be rather rude
521 at your machine, feel free to interrupt them.
523 0.5.4 (15 March 2000)
524 * Some algorithms in class matrix (notably determinant) were replaced by
525 less brain-dead ones and should now have much better performance.
526 * Checks were completely reorganized and split up into three parts:
527 a) exams (small regression tests with predefined input)
528 b) checks (lenghty coherence checks with random input)
529 c) timings (for coherence and crude benchmarking)
530 * Behaviour of .evalf() was changed: it doesn't .evalf() any exponents.
531 * Expanded expressions now remember they are expanded to prevent
532 superfluous expansions.
533 * Small bugfixes and improvements in the series expansion.
535 0.5.3 (23 February 2000)
536 * A more flexible scheme for registering functions was implemented,
537 allowing for remembering, too.
540 0.5.2 (16 February 2000)
541 * Mainly fixes a bug in the packaging of release 0.5.1.
543 0.5.1 (14 February 2000)
544 * Fixes a small number of bugs.
546 0.5.0 (7 February 2000)
547 * Expressions can be written ("archived") to files and read therefrom.
548 * Addition of GiNaC-cint, which lets you write complete programs in
549 an interactive shell-like manner in your favoured programming
552 0.4.1 (13 December 1999)
553 * Series Expansion of Gamma function and some other trigonometric
554 functions at their poles works now.
555 * Many more evaluations of special functions at points where
557 * info_flags::rational doesn't return true for complex extensions
558 any more---use info_flags::crational for the old behaviour.
559 info_flags::integer and -::cinteger work similarly, the same
560 holds for types like info_flags::rational_polynomial.
562 0.4.0 (26 November 1999)
563 * First public release.