1 This file records noteworthy changes.
4 * simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
5 simplified to "2*a.i*a.i".
6 * Fixed possible crash when calling subs() on expressions with non-commutative
10 * color and clifford classes are functional and documented.
11 * New "spinidx" class for dotted/undotted indices.
12 * Predefined spinor metric tensor (created by spinor_metric()).
13 * Symbols can have a LaTeX name, e.g. symbol s("s", "\\sigma");
14 * LaTeX output of indexed objects is much nicer.
15 * Fix some build-problems (with recent libreadline).
16 * Semantics of arithmetic operators now follows the C++ rules more strictly.
19 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() work with
20 non-symbols as the second argument in ginsh.
21 * the argument to collect() can be a list of objects in which case the
22 result is either a recursively collected polynomial, or a polynomial in
23 a distributed form with terms like coeff*x1^e1*...*xn^en, as specified by
24 the second argument to collect().
25 * Several bugfixes (including a nasty memory leak in .normal()).
26 * class matrix: solve() doesn't call algorithms redundantly any more and
27 inverse() falls back to solve() which works in more general cases.
30 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() can now
31 be used with constants, functions and indexed expressions as well, so you
32 can use it to collect by powers of Pi or sin(x), or to find the coefficient
35 - it only works with symbols, constants, functions and indexed expressions,
36 trying to find the coefficient of, e.g., "x^2" or "x+y" won't work;
37 - it does not know about dummy index summations; the coefficient of
38 gamma~0 in p.mu*gamma~mu should be p.0 but is returned as 0;
39 - using coeff(), tcoeff(), lcoeff() or collect() on elements of
40 noncommutative products might return wrong or surprising results.
41 * subs() no longer only substitutes symbols and indices but performs a more
42 general "syntactic substitution", i.e. it substitutes whole objects in sub-
43 expressions. You can subs((a+b)^2,a+b==3) and get 9, but subs(a+b+c,a+b==3)
46 - substituting numerics (subs(expr, 2==4)) will not replace then in all
47 occurences; in general, you shouldn't substitute numerics, though.
48 * Added preliminary (re)implementations of color and clifford classes.
49 * simplify_indexed(): contraction of symmetric and antisymmetric tensors
51 * Replaced the various print*() member functions by a single print() that
52 takes a print_context object that determines the output formatting. This
53 should make it easier to add more output types, such as LaTeX output,
54 which is based on work by Stefan Weinzierl.
55 * Added functions to retrieve the properties stored in archive objects
56 outside of unarchive() (for printing or debugging purposes).
57 * Some bugfixes (indexed objects, archive writing).
58 * .collect() on non-polynomials is now algebraically correct.
61 * Complete revamp of indexed objects. Instead of multiple classes for
62 indexed things and their indices there is now only one "indexed" class
63 and two types of indices: "idx" for simple indices and "varidx" for
64 indices with variance. There are predefined delta, epsilon and metric
65 tensors, and a function simplify_indexed() that performs canonicalization
66 and dummy index summations. Matrix objects can be indexed for doing simple
68 * Added an option "expand_indexed" to expand() to perform expansion of
69 indexed objects like (a+b).i -> a.i + b.i
70 * Renamed get_indices() to get_free_indices(), which no longer returns
71 dummy indices and checks the consistency of indices in sums.
72 * sqrfree() factorization fixed and improved syntactically.
73 * subs() works on matrices.
74 * Matrices can be constructed from flat list of elements; diagonal matrices
75 can be constructed from list of diagonal elements with diag_matrix().
76 * Fixed memory leak in expand().
77 * Operator% for objects of class ncmul has gone. Use operator* now for that
78 case too, which is much more natural.
80 0.7.3 (28 February 2001)
81 * Several bugfixes and minor performance tunings.
82 * Added a section to the tutorial about adding new algebraic classes to GiNaC.
83 * Closed many in-source documentation gaps.
85 0.7.2 (17 February 2001)
86 * Several bugfixes in power series expansion, one of them critical.
88 0.7.1 (7 February 2001)
89 * Fix problems with Cint that were caused by CLN's overloaded operator new.
90 * Fix compilation errors with GCC3.
91 * normal() handles large sums of fractions better and normalizes the exponent
93 * expand() always expands the exponent and transforms x^(a+b) -> x^a*x^b.
94 * Some bugfixes of series expansion around branch cuts of special functions.
96 0.7.0 (15 December 2000)
97 * Requires CLN 1.1 now. Class numeric doesn't use an indirect pointer to the
98 actual representation any more. This is a speedup.
99 * mul::expand() was reengineered to not allocate excess temporary memory.
100 * Non-integer powers of a symbol are treated as constants by (l)degree() and
101 coeff(). Using these functions on an expression containing such powers used
102 to fail with an internal error message. The side-effect is that collect()
103 can be used on expressions which are not polynomials.
104 * Added a man page for the ginac-config script.
105 * Ctor of numeric from char* honors Digits.
107 0.6.4 (10 August 2000)
108 * Complete revamp of methods in class matrix. Some redundant (and poor)
109 implementations of elimination schemes were thrown out. The code is now
110 highly orthogonal, more flexible and much more efficient.
111 * Some long standing and quite nasty bugs were discovered and fixed in the
112 following functions: add::normal(), heur_gcd(), sr_gcd() and Order_eval().
115 * Derivatives are now assembled in a slightly different manner (i.e. they
116 might 'look' different on first sight). Under certain circumstances this
117 can result in a dramatic speedup because it gives hashing a better chance,
118 especially when computing higher derivatives.
119 * Some series expansions of built-in functions have been reengineered.
120 * The algorithm for computing determinants can be chosen by the user. See
121 ginac/flags.h and ginac/matrix.h.
122 * The Dilogarithm (Li2) now has floating point evaluation, derivative and a
123 proper series expansion.
124 * Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
125 * Some minor bugfixes, one major lsolve()-bugfix and documentation updates.
128 * ginaccint.bin is now launched by a binary program instead of by a scripts.
129 This allows us to write #!-scripts. A small test suite for GiNaC-cint was
131 * Several minor bugfixes.
134 * Cleanup in the interface to Cint. The required version is now Cint 5.14.38.
135 * Several bugfixes in target install.
138 * IMPORTANT: Several interface changes make programs written with GiNaC
139 much clearer but break compatibility with older versions:
140 - f(x).series(x,p[,o]) -> f(x).series(x==p,o)
141 - series(f(x),x,p[,o]) -> series(f(x),x==p,o)
142 - gamma() -> tgamma() (The true Gamma function, there is now also
143 log(tgamma()), called lgamma(), in accord with ISO/IEC 9899:1999.)
144 - EulerGamma -> Euler
145 * #include'ing ginac.h defines the preprocessor symbols GINACLIB_MAJOR_VERSION,
146 GINACLIB_MINOR_VERSION, and GINACLIB_MICRO_VERSION with the respective GiNaC
147 library version numbers.
148 * Expressions can be constructed from strings like this:
149 ex e("2*x+y", lst(x, y));
150 * ex::to_rational() provides a way to extend the domain of functions like
151 gcd() and divide() that only work on polynomials or rational functions (the
152 good old ex::subs() method reverses this process)
153 * Calling diff() on a function that has no derivative defined returns the
154 inert derivative function "Derivative".
155 * Several new timings in the check target. Some of them may be rather rude
156 at your machine, feel free to interrupt them.
158 0.5.4 (15 March 2000)
159 * Some algorithms in class matrix (notably determinant) were replaced by
160 less brain-dead ones and should now have much better performance.
161 * Checks were completely reorganized and split up into three parts:
162 a) exams (small regression tests with predefined input)
163 b) checks (lenghty coherence checks with random input)
164 c) timings (for coherence and crude benchmarking)
165 * Behaviour of .evalf() was changed: it doesn't .evalf() any exponents.
166 * Expanded expressions now remember they are expanded to prevent
167 superfluous expansions.
168 * Small bugfixes and improvements in the series expansion.
170 0.5.3 (23 February 2000)
171 * A more flexible scheme for registering functions was implemented,
172 allowing for remembering, too.
175 0.5.2 (16 February 2000)
176 * Mainly fixes a bug in the packaging of release 0.5.1.
178 0.5.1 (14 February 2000)
179 * Fixes a small number of bugs.
181 0.5.0 (7 February 2000)
182 * Expressions can be written ("archived") to files and read therefrom.
183 * Addition of GiNaC-cint, which lets you write complete programs in
184 an interactive shell-like manner in your favoured programming
187 0.4.1 (13 December 1999)
188 * Series Expansion of Gamma function and some other trigonometric
189 functions at their poles works now.
190 * Many more evaluations of special functions at points where
192 * info_flags::rational doesn't return true for complex extensions
193 any more---use info_flags::crational for the old behaviour.
194 info_flags::integer and -::cinteger work similarly, the same
195 holds for types like info_flags::rational_polynomial.
197 0.4.0 (26 November 1999)
198 * First public release.