1 This file records noteworthy changes.
4 * GCD avoids to produce expanded expressions. [A. Sheplyakov]
5 * Fixed bug in expanding expressions containing dummy indices.
8 * integral() and eval_integ() can be used from ginsh.
9 * Integrals can be series-expanded.
10 * Fixed a library initialization problem.
11 * GiNaC compiles from tarball even if lex/flex is missing.
12 * Fixed bugs in canonicalize_clifford(), clifford_prime() and clifford_to_lst().
13 * clifford_moebius_map(), remove_dirac_ONE() and LaTeX output of Clifford
14 objects now care about representation labels.
16 * Better output for slashed expressions
18 1.3.0 (19 October 2004)
19 * The Clifford classes have been generalized to allow working with Clifford
20 algebras generated by arbitrary symmetric tensors or matrices. Also, a lot
21 of new functions for Clifford algebras have been added, including
22 automorphisms and Moebius transformations. [V. Kisil]
23 * Added some basic symbolic and numeric integration facilities. [C. Dams]
24 * The multiple polylogarithm Li() now evaluates numerically for arbitrary
26 * New functions G(a,y) and G(a,s,y) added (another notation for multiple
29 1.2.4 (12 October 2004)
30 * Added ex::unitcontprim() to compute the unit, content, and primitive parts
31 of a polynomial in one go.
32 * binomial(n, k) evaluates for non-integer arguments n.
33 * Li(2,x) now evaluates for +-I.
35 * Fixed bug in Li(n,x) (if Li(2,x) was calculated with high precision the
36 enlargement of the look-up table caused a segmentation fault).
37 * Fixed another bug in the series expansion of powers, and a bug in
40 1.2.3 (13 August 2004)
41 * Added variants of dirac_trace() and color_trace() that take the trace over
42 more than one representation label by specifying a set or list of labels.
43 * diracgamma::contract_with() uses Chisholm identities in 4 dimensions to
44 produce more compact results.
45 * Fixed a bug in the series expansion of powers.
46 * The --enable-html-doc and --enable-ps-doc configure options are gone.
47 Documentation in HTML, DVI, PostScript, and PDF formats is now built with
48 "make html", "make dvi", "make ps", and "make pdf", respectively. The only
49 documentation that gets built by default are the tutorial in .info format
53 * Added const_preorder_iterator and const_postorder_iterator classes (and
54 associated methods ex::pre/postorder_begin/end()) providing tree traversal
56 * Fixed the LaTeX output of the varidx class.
57 * Fixed bugs in series expansion and complex conjugation.
58 * (p.i*p.i).get_free_indices() returns the correct result () instead of (.i).
59 * Symbolic functions without any eval(), evalf() etc. functions now work
61 * integer_content(), content(), and primpart() now also work for polynomials
62 with rational coefficients (calculating the LCM of coefficients'
64 * Added method matrix::rank().
65 * Added function resultant().
66 * Added integer_content() function to ginsh.
69 * Fixed infinite recursion in atan2_evalf() and improved atan2_eval().
70 * Added automatic evaluations for trigonometric functions with negative
71 arguments (e.g. sin(-2) -> -sin(2)).
72 * Fixed a static initialization order goof-up.
73 * Fixed various bugs in series expansion.
76 * Added a structure<T> template class for the easy creation of user-defined
78 * Added support for (acyclic) visitors, to allow cleaner implementations of
80 * Added a const_iterator class that can be used instead of op()/nops().
81 * Completely revamped the implementation of expression output. It is now
82 possible to add new output formats, to change the behavior of predefined
83 formats at run-time, and to have different output styles for algebraic
85 * Symbols can be made non-commutative.
86 * Added a method ex::conjugate() and a function conjugate() for complex
87 conjugation. Symbols can be declared as real or complex-valued.
88 * Improved the speed of subs(), normal(), to_rational() and to_polynomial()
89 by the use of maps instead of lists. The old forms
90 subs(const lst & ls, const lst & lr, unsigned options)
91 to_rational/to_polynomial(lst & repl)
92 are still available for compatibility, but using the new forms
93 subs(const exmap & m, unsigned options)
94 to_rational/to_polynomial(exmap & repl)
95 is more efficient, especially when the number of replacements is large.
96 * quo(), rem(), prem(), sprem(), decomp_rational(), unit(), content(),
97 primpart() and matrix::charpoly() now take a "const ex &" instead of a
99 * Redundant expressions (two ex pointing to different objects are found to be
100 equal in compare()) are now actively deleted/fused to conserve memory and
101 speed up subsequent comparisons. This behavior can be suppressed on a
102 per-object level with status_flags::not_shareable. Lists and matrices are
103 not shareable by default.
104 * Lots of internal streamlining and optimizations.
105 * Caveats for class implementors:
106 - basic::copy() and basic::destroy() are gone; classes derived from
107 basic can use the defaults for the assignment operator and copy
109 - basic::subs(), basic::normal(), basic::to_rational() and
110 basic::to_polynomial() take 'exmap' objects instead of lists.
111 - basic::subs() now descends into subexpressions (if accessible via
112 nops()/op()/let_op()). If you have a custom implementation of subs()
113 that calls basic::subs() after substituting subexpressions, this needs
114 to be changed to a call to subs_one_level().
115 - lst::thislst() and exprseq::thisexprseq() renamed to thiscontainer().
116 - thiscontainer() and associated constructors now take a std::auto_ptr.
117 - Overloading basic::print() is now deprecated. You should use
118 print_func<>() class options instead.
120 1.1.7 (11 March 2004)
121 * Fixed a bug in canonicalize_clifford().
122 * Series expansion now works predictably. All terms with the exponent of the
123 expansion variable smaller than the given order are calculated exactly. If
124 the series is not terminating, the Order function is (at least) of the given
127 1.1.6 (22 January 2004)
128 * Added a function option "dummy()" which means "no options". This simplifies
129 the implementation of symbolic functions which are not to be further
131 * Removed a bug in the numerical evaluation of Li() that caused the system
132 to hang for certain parameter combinations.
133 * Fixed a bug in the calculation of hash values for indices that could lead
134 to wrong results or bogus error messages from simplify_indexed().
135 * Fixed a bug in the evaluation of harmonic polylogarithms for complex
136 arguments with positive imaginary part.
138 1.1.5 (5 November 2003)
139 * Harmonic polylogarithms now numerically evaluate for arbitrary arguments
140 (parameter must still be positive integers).
141 * The zeta function now can also be given a lst as a parameter in which case
142 it becomes a multiple zeta value. The use of mZeta is deprecated.
143 * The order of parameters for the multiple polylogarithm has been corrected.
144 * Documentation for the nested sums functions zeta, harmonic polylog, multiple
145 polylog, etc. has been added.
147 1.1.4 (17 October 2003)
148 * Lists and matrices can now be initialized from comma-separated lists of
149 expressions, like this:
159 This is both faster and produces much smaller code than the old constructors
160 lst(ex, ex, ...) and matrix(unsigned, unsigned, lst), especially in the case
161 of matrices, and is now the recommended way to create these objects.
162 * The function mZeta now evaluates much faster for arbitrary parameters. The
163 harmonic and multiple polylogarithms evaluate considerably faster and check
164 for convergence. The order of parameters for the harmonic polylogarithm
167 1.1.3 (22 August 2003)
168 * Added new symbolic functions for better integration with nestedsums:
169 (multiple) polylogarithm Li(), Nielsen's generalized polylogarithm S(),
170 harmonic polylogarithm H(), and multiple zeta value mZeta().
171 * New exhashmap<T> template intended as a drop-in replacement for
172 std::map<ex, T, ex_is_less> using GiNaC's hashing algorithms.
174 1.1.2 (11 August 2003)
175 * Fixed a bug in the unarchiving of sums and products: terms were not
176 reordered in a canonical way.
177 * Fixed a bug in normal()/numer_denom(): denominator was not made unit
178 normal if it was a simple number.
179 * Improved the speed of subs() in some cases.
182 * lst (and exprseq) provide iterators for read-only element access. For
183 sequential access this is one order faster than using op().
184 * Implemented relational::subs() (this was done in 1.0.9 but inadvertently
185 omitted from the 1.1 branch).
186 * pole_error and do_taylor are available to library users.
187 * Added on-line help and Tab-completion for print(), iprint(), print_latex()
188 and print_csrc() in ginsh.
191 * Removed deprecated macros is_ex_a, is_ex_exactly_a and friends for good.
192 * The scalar_products mechanism allows the specification of an index dimension.
193 * Removed dirac_gamma6/7().
194 * Added ex::to_polynomial().
195 * subs() accepts an optional "options" argument. The option
196 subs_option::subs_algebraic enables "smart" substitutions in products and
198 * Added stream manipulators "dflt", "latex", "python", "python_repr", "tree",
199 "csrc", "csrc_float", "csrc_double", "csrc_cl_N", "index_dimensions" and
200 "no_index_dimensions" to control the output format. Calling basic::print()
201 directly is now deprecated.
202 * Made the hashing more simple and efficient.
203 * Caveats for class implementors:
204 - basic::subs(): third argument changed from "bool" to "unsigned"
205 - unarchiving constructor and basic::unarchive(): "const" removed from
207 - basic::let_op() should only be implemented if write access to
208 subexpressions is desired
209 - simplify_ncmul() renamed to eval_ncmul()
210 - simplified_ncmul() renamed to hold_ncmul()
211 - nonsimplified_ncmul() renamed to reeval_ncmul()
213 1.0.14 (1 March 2003)
214 * Improved the C-source output: complex numbers are printed correctly (using
215 the STL complex<> template or cln::complex()), rational numbers use cl_RA()
216 in the CLN output, and small integers are printed in a more compact format
217 (e.g. "2.0" instead of "2.0000000e+00").
218 * function_options::set_return_type() and function_options::do_not_evalf_params()
221 1.0.13 (27 January 2003)
222 * Contracting epsilon tensors with Euclidean indices now works.
223 * Improved dummy index symmetrization in sums.
224 * Added dirac_gammaL/R(), which can be used instead of dirac_gamma6/7()
225 but are single objects, to allow for a more compact notation of Dirac
227 * Powers with negative numeric exponents are printed as fractions in the
229 * Added symbolic_matrix() for the convenient creation of matrices filled
231 * Added collect_common_factors() which collects common factors from the
233 * simplify_indexed() converts "gamma~mu*p.mu" to "p\".
235 1.0.12 (30 October 2002)
236 * Fixed a bug in power::expand() that could produce invalid expressions.
237 * The input parser no longer ignores extra data following accepted input.
238 * Improved the CLN C-source output (integers are printed as integers, and
239 floating point numbers include the precision).
240 * Fixed a problem in the LaTeX-output of negative fractions.
241 * Added print_latex() and print_csrc() to ginsh.
242 * The sprem() function is now public.
244 1.0.11 (18 September 2002)
245 * Fixed a possible memory corruption in contractions of indexed objects with
246 delta or metric tensors.
247 * Computing the derivative of a power series object with respect to a symbol
248 that is not the expansion variable now works correctly.
249 * Several bugfixes in code generation.
251 1.0.10 (24 July 2002)
252 * Powers of indexed objects are now parenthesized correctly in LaTeX output.
253 * Input parser handles indices (they have to be specified in the same list
255 * Added some limited support for subspaces in the idx and tensor classes.
256 * Fixed a bug in canonicalize() (antisymmetric canonicalization of an already
257 sorted list containing two or more equal objects failed to return 0).
260 * simplify_indexed() now raises/lowers dummy indices to canonicalize the index
261 variance. This allows some simplifications that weren't possible before,
262 like eps~a.b~c~d*X.a*X~b -> 0 and X.a~a-X~a.a -> 0.
263 * Implemented relational::subs().
264 * Fixed bug in simplify_ncmul() for clifford objects.
266 1.0.8 (31 March 2002)
267 * Improvements in memory usage of the expand() methods.
269 1.0.7 (18 March 2002)
270 * Fixed LaTeX output of indexed and matrix objects.
271 * Fixed matrix::pow(n) for n==0 and added helper functions to create unit
272 matrices "ex unit_matrix(unsigned, unsigned)".
275 * "(x+1).subs(x==x-1)" now returns the correct result "x" instead of "x-1".
277 1.0.5 (27 January 2002)
278 * (l)degree(s), coeff(s, n) and collect(s) were extended to accept expressions
279 of any class (except add/mul/ncmul/numeric) for "s". They should even work
280 if "s" is a "power" object, as long as the exponent is non-integer, but with
281 some limitations. For example, you can "collect(a*2^x+b*2^x, 2^x)" to get
282 "(a+b)*2^x", but "degree(2^(3*x), 2^x)" yields 0 instead of 3).
283 * Fixed a small output bug.
285 1.0.4 (24 January 2002)
286 * Speedup in expand().
287 * Faster Bernoulli numbers (Markus Nullmeier).
288 * Some minor bugfixes and documentation updates.
290 1.0.3 (21 December 2001)
291 * Fixed a bug where quo() would call vector::reserve() with a negative
293 * Fix several bugs in code generation.
295 1.0.2 (19 December 2001)
296 * Input parser recognizes "sqrt()", which is also used in the output.
297 * divide(a,b,q) only modifies q if the division succeeds; also, divide(a,b,a)
299 * Fixed small bug in dummy index renaming which could cause it to not
300 recognize renamable indices in some cases.
301 * power::degree() and power::ldegree() throw an exception when encountering
302 a non-integer exponent.
303 * Add output-support for Python bindings.
305 1.0.1 (22 November 2001)
306 * Function sqrfree() handles a few more cases now.
307 * Class relational has real canonical ordering now.
308 * Handle obscene libreadline version numbers when building ginsh.
310 1.0.0 (6 November 2001)
311 * Some internal reorganization resulting in a general speed-up.
312 * The last 3 evaluated expressions in ginsh are now referred to with the
313 tokens '%', '%%' and '%%%'. The old '"', '""' and '"""' remain for
314 compatibility but may be removed in a future version of GiNaC.
316 0.9.4 (20 September 2001)
317 * Functions have better support for external scripting languages.
318 * Interface cleanups and bugfixes.
319 * Fix silly bug in evalf() that prevented things like 2^Pi being computed.
321 0.9.3 (16 August 2001)
322 * series expansion now much more consistent for small order expansion.
323 * lsolve() accepts algorithmic hint as parameter.
326 * Epsilon tensor is more functional.
327 * simplify_indexed() is better at detecting expressions that vanish for
329 * Several little bugfixes and consistency enhancements.
332 * Ctors of class numeric are not explicit any more. All built-in callers for
333 pseudofunctions are now templated and default to ex arguments which relaxes
334 the need for explicit ctors.
335 * New functions/methods:
337 - remove_first(), remove_last(), sort() and unique() for lists
338 - symmetrize_cyclic()
340 * Instead of just totally symmetric or antisymmetric, complex symmetries
341 can now be defined for indexed objects. Symmetries are described by a
342 tree of "symmetry" objects that is constructed with the sy_none(),
343 sy_symm(), sy_anti() and sy_cycl() functions. The symmetry of a function
344 with respect to its arguments can also be defined (this is currently
345 only used for the Beta function).
346 * Generalized map() to take a function object instead of a function pointer.
347 This allows passing an arbitrary number of additional state to the
348 function being called.
349 * color_trace(), dirac_trace(), diff(), expand(), evalf() and normal() work
350 better with container classes, e.g. using color_trace() on a relation will
351 take the trace on both sides, using diff() on a matrix differentiates every
353 * diff() works properly with non-commutative products and indexed objects.
354 * New option flag "expand_function_args" for expand().
355 * Supplement some (now deprecated) macros by inlined template functions:
356 - is_of_type(foo, type) -> is_a<type>(foo)
357 - is_ex_of_type(foo, type) -> is_a<type>(foo)
358 - is_exactly_of_type(foo, type) -> is_exactly_a<type>(foo)
359 - is_ex_exactly_of_type(foo, type) -> is_exactly_a<type>(foo)
360 - ex_to_foobar(baz) -> ex_to<foobar>(baz)
361 * rem(c, p[x], x) (c: numeric, p[x]: polynomial) erroneously returned p[x]
363 * Small bugfixes in pattern matching.
364 * Updated libtool to version 1.4.
367 * In the output and in ginsh, lists are now delimited by { } braces, and
368 matrices are delimited by single [ ] brackets.
369 * simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
370 simplified to "2*a.i*a.i" (or "2*a.j*a.j").
371 * New functions/methods:
372 - canonicalize_clifford() (helpful when comparing expressions containing
374 - symmetrize() and antisymmetrize()
375 - numer_denom() (return numerator and denominator in one call)
376 - map() (apply function to subexpressions)
377 - evalm() (evaluate sums, products and integer powers of matrices)
378 * Added a new function match() for performing pattern matching. subs() and
379 has() also accept patterns as arguments. A pattern can be any expression,
380 optionally containing wildcard objects. These are constructed with the
381 call "wild(<unsigned>)" and are denoted as "$0", "$1" etc. in the output
383 * Positive integer powers of non-commutative expressions (except matrices)
384 are automatically expanded.
385 * Removed cint subdirectory, ginaccint is a separate package now due to
386 packaging considerations.
387 * Several little bugfixes.
390 * color and clifford classes are functional and documented.
391 * New "spinidx" class for dotted/undotted indices.
392 * Predefined spinor metric tensor (created by spinor_metric()).
393 * Symbols can have a LaTeX name, e.g. symbol s("s", "\\sigma");
394 * LaTeX output of indexed objects is much nicer.
395 * Fixed some build problems (with recent libreadline).
396 * Semantics of arithmetic operators now follows the C++ rules more strictly.
398 0.8.2 (24 April 2001)
399 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() work with
400 non-symbols as the second argument in ginsh.
401 * the argument to collect() can be a list of objects in which case the
402 result is either a recursively collected polynomial, or a polynomial in
403 a distributed form with terms like coeff*x1^e1*...*xn^en, as specified by
404 the second argument to collect().
405 * Several bugfixes (including a nasty memory leak in .normal()).
406 * class matrix: solve() doesn't call algorithms redundantly any more and
407 inverse() falls back to solve() which works in more general cases.
409 0.8.1 (16 April 2001)
410 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() can now
411 be used with constants, functions and indexed expressions as well, so you
412 can use it to collect by powers of Pi or sin(x), or to find the coefficient
415 - it only works with symbols, constants, functions and indexed expressions,
416 trying to find the coefficient of, e.g., "x^2" or "x+y" won't work;
417 - it does not know about dummy index summations; the coefficient of
418 gamma~0 in p.mu*gamma~mu should be p.0 but is returned as 0;
419 - using coeff(), tcoeff(), lcoeff() or collect() on elements of
420 noncommutative products might return wrong or surprising results.
421 * subs() no longer only substitutes symbols and indices but performs a more
422 general "syntactic substitution", i.e. it substitutes whole objects in sub-
423 expressions. You can subs((a+b)^2,a+b==3) and get 9, but subs(a+b+c,a+b==3)
426 - substituting numerics (subs(expr, 2==4)) will not replace then in all
427 occurences; in general, you shouldn't substitute numerics, though.
428 * Added preliminary (re)implementations of color and clifford classes.
429 * simplify_indexed(): contraction of symmetric and antisymmetric tensors
431 * Replaced the various print*() member functions by a single print() that
432 takes a print_context object that determines the output formatting. This
433 should make it easier to add more output types, such as LaTeX output,
434 which is based on work by Stefan Weinzierl.
435 * Added functions to retrieve the properties stored in archive objects
436 outside of unarchive() (for printing or debugging purposes).
437 * Some bugfixes (indexed objects, archive writing).
438 * .collect() on non-polynomials is now algebraically correct.
440 0.8.0 (24 March 2001)
441 * Complete revamp of indexed objects. Instead of multiple classes for
442 indexed things and their indices there is now only one "indexed" class
443 and two types of indices: "idx" for simple indices and "varidx" for
444 indices with variance. There are predefined delta, epsilon and metric
445 tensors, and a function simplify_indexed() that performs canonicalization
446 and dummy index summations. Matrix objects can be indexed for doing simple
448 * Added an option "expand_indexed" to expand() to perform expansion of
449 indexed objects like (a+b).i -> a.i + b.i
450 * Renamed get_indices() to get_free_indices(), which no longer returns
451 dummy indices and checks the consistency of indices in sums.
452 * sqrfree() factorization fixed and improved syntactically.
453 * subs() works on matrices.
454 * Matrices can be constructed from flat list of elements; diagonal matrices
455 can be constructed from list of diagonal elements with diag_matrix().
456 * Fixed memory leak in expand().
457 * Operator% for objects of class ncmul has gone. Use operator* now for that
458 case too, which is much more natural.
460 0.7.3 (28 February 2001)
461 * Several bugfixes and minor performance tunings.
462 * Added a section to the tutorial about adding new algebraic classes to GiNaC.
463 * Closed many in-source documentation gaps.
465 0.7.2 (17 February 2001)
466 * Several bugfixes in power series expansion, one of them critical.
468 0.7.1 (7 February 2001)
469 * Fix problems with Cint that were caused by CLN's overloaded operator new.
470 * Fix compilation errors with GCC3.
471 * normal() handles large sums of fractions better and normalizes the exponent
472 of power expressions.
473 * expand() always expands the exponent and transforms x^(a+b) -> x^a*x^b.
474 * Some bugfixes of series expansion around branch cuts of special functions.
476 0.7.0 (15 December 2000)
477 * Requires CLN 1.1 now. Class numeric doesn't use an indirect pointer to the
478 actual representation any more. This is a speedup.
479 * mul::expand() was reengineered to not allocate excess temporary memory.
480 * Non-integer powers of a symbol are treated as constants by (l)degree() and
481 coeff(). Using these functions on an expression containing such powers used
482 to fail with an internal error message. The side-effect is that collect()
483 can be used on expressions which are not polynomials.
484 * Added a man page for the ginac-config script.
485 * Ctor of numeric from char* honors Digits.
487 0.6.4 (10 August 2000)
488 * Complete revamp of methods in class matrix. Some redundant (and poor)
489 implementations of elimination schemes were thrown out. The code is now
490 highly orthogonal, more flexible and much more efficient.
491 * Some long standing and quite nasty bugs were discovered and fixed in the
492 following functions: add::normal(), heur_gcd(), sr_gcd() and Order_eval().
495 * Derivatives are now assembled in a slightly different manner (i.e. they
496 might 'look' different on first sight). Under certain circumstances this
497 can result in a dramatic speedup because it gives hashing a better chance,
498 especially when computing higher derivatives.
499 * Some series expansions of built-in functions have been reengineered.
500 * The algorithm for computing determinants can be chosen by the user. See
501 ginac/flags.h and ginac/matrix.h.
502 * The Dilogarithm (Li2) now has floating point evaluation, derivative and a
503 proper series expansion.
504 * Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
505 * Some minor bugfixes, one major lsolve()-bugfix and documentation updates.
508 * ginaccint.bin is now launched by a binary program instead of by a scripts.
509 This allows us to write #!-scripts. A small test suite for GiNaC-cint was
511 * Several minor bugfixes.
514 * Cleanup in the interface to Cint. The required version is now Cint 5.14.38.
515 * Several bugfixes in target install.
518 * IMPORTANT: Several interface changes make programs written with GiNaC
519 much clearer but break compatibility with older versions:
520 - f(x).series(x,p[,o]) -> f(x).series(x==p,o)
521 - series(f(x),x,p[,o]) -> series(f(x),x==p,o)
522 - gamma() -> tgamma() (The true Gamma function, there is now also
523 log(tgamma()), called lgamma(), in accord with ISO/IEC 9899:1999.)
524 - EulerGamma -> Euler
525 * #include'ing ginac.h defines the preprocessor symbols GINACLIB_MAJOR_VERSION,
526 GINACLIB_MINOR_VERSION, and GINACLIB_MICRO_VERSION with the respective GiNaC
527 library version numbers.
528 * Expressions can be constructed from strings like this:
529 ex e("2*x+y", lst(x, y));
530 * ex::to_rational() provides a way to extend the domain of functions like
531 gcd() and divide() that only work on polynomials or rational functions (the
532 good old ex::subs() method reverses this process)
533 * Calling diff() on a function that has no derivative defined returns the
534 inert derivative function "Derivative".
535 * Several new timings in the check target. Some of them may be rather rude
536 at your machine, feel free to interrupt them.
538 0.5.4 (15 March 2000)
539 * Some algorithms in class matrix (notably determinant) were replaced by
540 less brain-dead ones and should now have much better performance.
541 * Checks were completely reorganized and split up into three parts:
542 a) exams (small regression tests with predefined input)
543 b) checks (lenghty coherence checks with random input)
544 c) timings (for coherence and crude benchmarking)
545 * Behaviour of .evalf() was changed: it doesn't .evalf() any exponents.
546 * Expanded expressions now remember they are expanded to prevent
547 superfluous expansions.
548 * Small bugfixes and improvements in the series expansion.
550 0.5.3 (23 February 2000)
551 * A more flexible scheme for registering functions was implemented,
552 allowing for remembering, too.
555 0.5.2 (16 February 2000)
556 * Mainly fixes a bug in the packaging of release 0.5.1.
558 0.5.1 (14 February 2000)
559 * Fixes a small number of bugs.
561 0.5.0 (7 February 2000)
562 * Expressions can be written ("archived") to files and read therefrom.
563 * Addition of GiNaC-cint, which lets you write complete programs in
564 an interactive shell-like manner in your favoured programming
567 0.4.1 (13 December 1999)
568 * Series Expansion of Gamma function and some other trigonometric
569 functions at their poles works now.
570 * Many more evaluations of special functions at points where
572 * info_flags::rational doesn't return true for complex extensions
573 any more---use info_flags::crational for the old behaviour.
574 info_flags::integer and -::cinteger work similarly, the same
575 holds for types like info_flags::rational_polynomial.
577 0.4.0 (26 November 1999)
578 * First public release.