1 This file records noteworthy changes.
4 * In the output and in ginsh, lists are now delimited by { } braces, and
5 matrices are delimited by single [ ] brackets.
6 * simplify_indexed() renames dummy indices so, e.g., "a.i*a.i+a.j*a.j" gets
7 simplified to "2*a.i*a.i".
8 * New functions/methods:
9 - canonicalize_clifford() (helpful when comparing expressions containing
11 - symmetrize() and antisymmetrize()
12 - numer_denom() (return numerator and denominator)
13 - map() (apply function to subexpressions)
14 - evalm() (evaluate sums and products of matrices)
15 * Added a new function match() for performing pattern matching. subs() and
16 has() also accept patterns as arguments. A pattern can be any expression,
17 optionally containing wildcard objects. These are constructed with the
18 call "wild(<unsigned>)" and are denoted as "$0", "$1" etc. in the output
20 * Fixed possible crash when calling subs() on expressions with non-commutative
22 * Positive integer powers of non-commutative expressions are automatically
24 * Several little bugfixes.
27 * color and clifford classes are functional and documented.
28 * New "spinidx" class for dotted/undotted indices.
29 * Predefined spinor metric tensor (created by spinor_metric()).
30 * Symbols can have a LaTeX name, e.g. symbol s("s", "\\sigma");
31 * LaTeX output of indexed objects is much nicer.
32 * Fixed some build problems (with recent libreadline).
33 * Semantics of arithmetic operators now follows the C++ rules more strictly.
36 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() work with
37 non-symbols as the second argument in ginsh.
38 * the argument to collect() can be a list of objects in which case the
39 result is either a recursively collected polynomial, or a polynomial in
40 a distributed form with terms like coeff*x1^e1*...*xn^en, as specified by
41 the second argument to collect().
42 * Several bugfixes (including a nasty memory leak in .normal()).
43 * class matrix: solve() doesn't call algorithms redundantly any more and
44 inverse() falls back to solve() which works in more general cases.
47 * degree(), ldegree(), coeff(), lcoeff(), tcoeff() and collect() can now
48 be used with constants, functions and indexed expressions as well, so you
49 can use it to collect by powers of Pi or sin(x), or to find the coefficient
52 - it only works with symbols, constants, functions and indexed expressions,
53 trying to find the coefficient of, e.g., "x^2" or "x+y" won't work;
54 - it does not know about dummy index summations; the coefficient of
55 gamma~0 in p.mu*gamma~mu should be p.0 but is returned as 0;
56 - using coeff(), tcoeff(), lcoeff() or collect() on elements of
57 noncommutative products might return wrong or surprising results.
58 * subs() no longer only substitutes symbols and indices but performs a more
59 general "syntactic substitution", i.e. it substitutes whole objects in sub-
60 expressions. You can subs((a+b)^2,a+b==3) and get 9, but subs(a+b+c,a+b==3)
63 - substituting numerics (subs(expr, 2==4)) will not replace then in all
64 occurences; in general, you shouldn't substitute numerics, though.
65 * Added preliminary (re)implementations of color and clifford classes.
66 * simplify_indexed(): contraction of symmetric and antisymmetric tensors
68 * Replaced the various print*() member functions by a single print() that
69 takes a print_context object that determines the output formatting. This
70 should make it easier to add more output types, such as LaTeX output,
71 which is based on work by Stefan Weinzierl.
72 * Added functions to retrieve the properties stored in archive objects
73 outside of unarchive() (for printing or debugging purposes).
74 * Some bugfixes (indexed objects, archive writing).
75 * .collect() on non-polynomials is now algebraically correct.
78 * Complete revamp of indexed objects. Instead of multiple classes for
79 indexed things and their indices there is now only one "indexed" class
80 and two types of indices: "idx" for simple indices and "varidx" for
81 indices with variance. There are predefined delta, epsilon and metric
82 tensors, and a function simplify_indexed() that performs canonicalization
83 and dummy index summations. Matrix objects can be indexed for doing simple
85 * Added an option "expand_indexed" to expand() to perform expansion of
86 indexed objects like (a+b).i -> a.i + b.i
87 * Renamed get_indices() to get_free_indices(), which no longer returns
88 dummy indices and checks the consistency of indices in sums.
89 * sqrfree() factorization fixed and improved syntactically.
90 * subs() works on matrices.
91 * Matrices can be constructed from flat list of elements; diagonal matrices
92 can be constructed from list of diagonal elements with diag_matrix().
93 * Fixed memory leak in expand().
94 * Operator% for objects of class ncmul has gone. Use operator* now for that
95 case too, which is much more natural.
97 0.7.3 (28 February 2001)
98 * Several bugfixes and minor performance tunings.
99 * Added a section to the tutorial about adding new algebraic classes to GiNaC.
100 * Closed many in-source documentation gaps.
102 0.7.2 (17 February 2001)
103 * Several bugfixes in power series expansion, one of them critical.
105 0.7.1 (7 February 2001)
106 * Fix problems with Cint that were caused by CLN's overloaded operator new.
107 * Fix compilation errors with GCC3.
108 * normal() handles large sums of fractions better and normalizes the exponent
109 of power expressions.
110 * expand() always expands the exponent and transforms x^(a+b) -> x^a*x^b.
111 * Some bugfixes of series expansion around branch cuts of special functions.
113 0.7.0 (15 December 2000)
114 * Requires CLN 1.1 now. Class numeric doesn't use an indirect pointer to the
115 actual representation any more. This is a speedup.
116 * mul::expand() was reengineered to not allocate excess temporary memory.
117 * Non-integer powers of a symbol are treated as constants by (l)degree() and
118 coeff(). Using these functions on an expression containing such powers used
119 to fail with an internal error message. The side-effect is that collect()
120 can be used on expressions which are not polynomials.
121 * Added a man page for the ginac-config script.
122 * Ctor of numeric from char* honors Digits.
124 0.6.4 (10 August 2000)
125 * Complete revamp of methods in class matrix. Some redundant (and poor)
126 implementations of elimination schemes were thrown out. The code is now
127 highly orthogonal, more flexible and much more efficient.
128 * Some long standing and quite nasty bugs were discovered and fixed in the
129 following functions: add::normal(), heur_gcd(), sr_gcd() and Order_eval().
132 * Derivatives are now assembled in a slightly different manner (i.e. they
133 might 'look' different on first sight). Under certain circumstances this
134 can result in a dramatic speedup because it gives hashing a better chance,
135 especially when computing higher derivatives.
136 * Some series expansions of built-in functions have been reengineered.
137 * The algorithm for computing determinants can be chosen by the user. See
138 ginac/flags.h and ginac/matrix.h.
139 * The Dilogarithm (Li2) now has floating point evaluation, derivative and a
140 proper series expansion.
141 * Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
142 * Some minor bugfixes, one major lsolve()-bugfix and documentation updates.
145 * ginaccint.bin is now launched by a binary program instead of by a scripts.
146 This allows us to write #!-scripts. A small test suite for GiNaC-cint was
148 * Several minor bugfixes.
151 * Cleanup in the interface to Cint. The required version is now Cint 5.14.38.
152 * Several bugfixes in target install.
155 * IMPORTANT: Several interface changes make programs written with GiNaC
156 much clearer but break compatibility with older versions:
157 - f(x).series(x,p[,o]) -> f(x).series(x==p,o)
158 - series(f(x),x,p[,o]) -> series(f(x),x==p,o)
159 - gamma() -> tgamma() (The true Gamma function, there is now also
160 log(tgamma()), called lgamma(), in accord with ISO/IEC 9899:1999.)
161 - EulerGamma -> Euler
162 * #include'ing ginac.h defines the preprocessor symbols GINACLIB_MAJOR_VERSION,
163 GINACLIB_MINOR_VERSION, and GINACLIB_MICRO_VERSION with the respective GiNaC
164 library version numbers.
165 * Expressions can be constructed from strings like this:
166 ex e("2*x+y", lst(x, y));
167 * ex::to_rational() provides a way to extend the domain of functions like
168 gcd() and divide() that only work on polynomials or rational functions (the
169 good old ex::subs() method reverses this process)
170 * Calling diff() on a function that has no derivative defined returns the
171 inert derivative function "Derivative".
172 * Several new timings in the check target. Some of them may be rather rude
173 at your machine, feel free to interrupt them.
175 0.5.4 (15 March 2000)
176 * Some algorithms in class matrix (notably determinant) were replaced by
177 less brain-dead ones and should now have much better performance.
178 * Checks were completely reorganized and split up into three parts:
179 a) exams (small regression tests with predefined input)
180 b) checks (lenghty coherence checks with random input)
181 c) timings (for coherence and crude benchmarking)
182 * Behaviour of .evalf() was changed: it doesn't .evalf() any exponents.
183 * Expanded expressions now remember they are expanded to prevent
184 superfluous expansions.
185 * Small bugfixes and improvements in the series expansion.
187 0.5.3 (23 February 2000)
188 * A more flexible scheme for registering functions was implemented,
189 allowing for remembering, too.
192 0.5.2 (16 February 2000)
193 * Mainly fixes a bug in the packaging of release 0.5.1.
195 0.5.1 (14 February 2000)
196 * Fixes a small number of bugs.
198 0.5.0 (7 February 2000)
199 * Expressions can be written ("archived") to files and read therefrom.
200 * Addition of GiNaC-cint, which lets you write complete programs in
201 an interactive shell-like manner in your favoured programming
204 0.4.1 (13 December 1999)
205 * Series Expansion of Gamma function and some other trigonometric
206 functions at their poles works now.
207 * Many more evaluations of special functions at points where
209 * info_flags::rational doesn't return true for complex extensions
210 any more---use info_flags::crational for the old behaviour.
211 info_flags::integer and -::cinteger work similarly, the same
212 holds for types like info_flags::rational_polynomial.
214 0.4.0 (26 November 1999)
215 * First public release.