]> www.ginac.de Git - ginac.git/log
ginac.git
23 years agofriend void ginsh_get_ginac_functions(void) is now removed from
Alexander Frink [Tue, 16 May 2000 19:58:51 +0000 (19:58 +0000)]
friend void ginsh_get_ginac_functions(void) is now removed from
makecint with #ifndef __MAKECINT__ macros instead of dummies.cpp.

23 years agoSupport for cint versions older than 5.14.38 is discontinued.
Alexander Frink [Tue, 16 May 2000 19:57:43 +0000 (19:57 +0000)]
Support for cint versions older than 5.14.38 is discontinued.
Beginning with 5.14.38 namespaces seem to work reasonably well
so that it is no longer necessary to compile two separate libginacs.
Overloading resolution has been fixed in 5.14.37 so that the
declarations in dummies.* are no longer needed. libginac can now
be compiled without -D NO_NAMESPACE_GINAC and -D CINT_CONVERSION_WORKAROUND.

23 years ago- mentioned "Derivative" function
Christian Bauer [Thu, 11 May 2000 23:22:20 +0000 (23:22 +0000)]
- mentioned "Derivative" function

23 years ago- Updated Cint install status. release_0-6-0
Richard Kreckel [Thu, 11 May 2000 21:43:26 +0000 (21:43 +0000)]
- Updated Cint install status.

23 years ago- numeric::numeric(const char*): parse complex numbers.
Richard Kreckel [Thu, 11 May 2000 21:16:38 +0000 (21:16 +0000)]
- numeric::numeric(const char*): parse complex numbers.
- Documentation fixes.

23 years ago- Updated date.
Richard Kreckel [Thu, 11 May 2000 21:15:53 +0000 (21:15 +0000)]
- Updated date.

23 years agoMore features for GiNaC-cint:
Alexander Frink [Thu, 11 May 2000 21:12:18 +0000 (21:12 +0000)]
More features for GiNaC-cint:
restart environment
redirect output
updated man page

23 years ago- small efficiency fix in mul_series
Christian Bauer [Thu, 11 May 2000 20:39:18 +0000 (20:39 +0000)]
- small efficiency fix in mul_series

23 years agomore modifications for future cints with correct namespace handling
Alexander Frink [Wed, 10 May 2000 23:06:49 +0000 (23:06 +0000)]
more modifications for future cints with correct namespace handling

23 years ago#pragmas for MAKECINT in combination with namespaces
Alexander Frink [Wed, 10 May 2000 23:04:08 +0000 (23:04 +0000)]
#pragmas for MAKECINT in combination with namespaces

23 years agoenhanced ginaccint.bin.cpp (help, read from file, save commands)
Alexander Frink [Wed, 10 May 2000 23:03:35 +0000 (23:03 +0000)]
enhanced ginaccint.bin.cpp (help, read from file, save commands)
removed necessity for dummies.h for cint >= 5.14.37
some modifications for future cint-versions which will support namespaces (hopefully)

23 years ago- matrix::determinant_numeric(): deleted. It was duplicate code because we
Richard Kreckel [Sat, 6 May 2000 19:43:12 +0000 (19:43 +0000)]
- matrix::determinant_numeric(): deleted.  It was duplicate code because we
  have matrix::gauss_elimination().

23 years ago- normal() respects the "level" parameter to limit the recursion depth
Christian Bauer [Fri, 5 May 2000 17:09:04 +0000 (17:09 +0000)]
- normal() respects the "level" parameter to limit the recursion depth
- added sprem() function (sparse pseudo-remainder), as in Maple

23 years ago- introduced info_flag::algebraic.
Richard Kreckel [Sat, 29 Apr 2000 02:15:55 +0000 (02:15 +0000)]
- introduced info_flag::algebraic.
- matrix::fraction_free_elimination(bool) finished and used by matrix::determinant().
- introduced some comments.

23 years ago- documented info_flags::algebraic.
Richard Kreckel [Sat, 29 Apr 2000 02:14:19 +0000 (02:14 +0000)]
- documented info_flags::algebraic.

23 years ago- Extended check (FIXME: may under certain conditions rout oom).
Richard Kreckel [Sat, 29 Apr 2000 02:13:29 +0000 (02:13 +0000)]
- Extended check (FIXME: may under certain conditions rout oom).

24 years ago- numeric::to_rational(): fixed thinko.
Richard Kreckel [Fri, 21 Apr 2000 21:28:10 +0000 (21:28 +0000)]
- numeric::to_rational(): fixed thinko.

24 years ago- expairseq::to_rational(): Fixed thinko.
Richard Kreckel [Thu, 20 Apr 2000 21:41:35 +0000 (21:41 +0000)]
- expairseq::to_rational(): Fixed thinko.

24 years ago- expairseq.cpp: moved expairseq::to_rational to...
Richard Kreckel [Thu, 20 Apr 2000 04:16:23 +0000 (04:16 +0000)]
- expairseq.cpp: moved expairseq::to_rational to...
- ...normal.cpp.
- included new method matrix::determinant_bareiss, which doesn't rely on GCDs.
- log_series now honors the branch cut.

24 years ago- New figure classhierarchy.fig, which we all know, included in...
Richard Kreckel [Thu, 20 Apr 2000 04:14:42 +0000 (04:14 +0000)]
- New figure classhierarchy.fig, which we all know, included in...
- ... file ginac.texi.

24 years ago- Timings D and E: finer time resolution
Richard Kreckel [Thu, 20 Apr 2000 04:13:55 +0000 (04:13 +0000)]
- Timings D and E: finer time resolution
- Timings P, P', Q, Q': construct matrix with some loops instead of reading
  it from a const ex[][].  time_lw_w101n.h is not needed any more.
- check_matrices.cpp: check the new fraction free determinant.

24 years ago- fixed a typo
Christian Bauer [Thu, 20 Apr 2000 00:41:36 +0000 (00:41 +0000)]
- fixed a typo
- divide(0, something) now succeeds (quotient = 0)
- gcd(): variables are ordered by their maximum degree in the two polynomials,
  not the minimum
- implemented eu_gcd(), euprem_gcd(), peu_gcd() and red_gcd() (for testing)
- fixed numeric::to_rational (dropped denominator)

24 years ago- fixed typos
Christian Bauer [Thu, 20 Apr 2000 00:39:52 +0000 (00:39 +0000)]
- fixed typos

24 years ago- Overwritten: pseries::printtree().
Richard Kreckel [Tue, 18 Apr 2000 13:46:03 +0000 (13:46 +0000)]
- Overwritten: pseries::printtree().

24 years ago- '?time' now works
Christian Bauer [Thu, 13 Apr 2000 20:30:11 +0000 (20:30 +0000)]
- '?time' now works

24 years ago- numeric::archive(): fixed a typo.
Richard Kreckel [Tue, 11 Apr 2000 00:17:37 +0000 (00:17 +0000)]
- numeric::archive(): fixed a typo.

24 years ago- De-mystified the chapter about relationals.
Richard Kreckel [Mon, 10 Apr 2000 17:20:36 +0000 (17:20 +0000)]
- De-mystified the chapter about relationals.

24 years ago- normal() works for relationals: it normalizes both sides
Christian Bauer [Mon, 10 Apr 2000 16:49:07 +0000 (16:49 +0000)]
- normal() works for relationals: it normalizes both sides

24 years ago- modified printcsrc() example
Christian Bauer [Mon, 10 Apr 2000 16:37:33 +0000 (16:37 +0000)]
- modified printcsrc() example

24 years ago- extended the "Algorithms" chapter which has been renamed to "Methods and
Christian Bauer [Fri, 7 Apr 2000 20:28:13 +0000 (20:28 +0000)]
- extended the "Algorithms" chapter which has been renamed to "Methods and
  Functions"
- added a section about lists (class lst) to the "Basic Concepts" chapter

24 years ago- mentioned to_rational() and input parser
Christian Bauer [Fri, 7 Apr 2000 20:27:33 +0000 (20:27 +0000)]
- mentioned to_rational() and input parser

24 years ago- added ex::to_rational() to convert general expression to rational expression
Christian Bauer [Thu, 6 Apr 2000 19:58:15 +0000 (19:58 +0000)]
- added ex::to_rational() to convert general expression to rational expression
  by replacing all non-rational parts with temporary symbols, e.g.:
    ex a = pow(sin(x), 2) - pow(cos(x), 2);
    ex b = sin(x) + cos(x);
    ex d;
    lst l;
    divide(a.to_rational(l), b.to_rational(l), d);
    cout << d.subs(l) << endl;
  will print "sin(x)-cos(x)"

24 years ago- fixed a small oops: iprint() didn't reset the number output format to decimal
Christian Bauer [Wed, 5 Apr 2000 21:01:11 +0000 (21:01 +0000)]
- fixed a small oops: iprint() didn't reset the number output format to decimal

24 years ago- streamlined the code in numeric::numeric(archive_node &) and numeric::archive()
Christian Bauer [Wed, 5 Apr 2000 21:00:52 +0000 (21:00 +0000)]
- streamlined the code in numeric::numeric(archive_node &) and numeric::archive()
  with respect to sstream vs. strstream (avoided unnecessary code duplication)

24 years ago- implemented #b, #o, #x and #nR prefixes for binary, octal, hexadecimal, and
Christian Bauer [Wed, 5 Apr 2000 19:31:33 +0000 (19:31 +0000)]
- implemented #b, #o, #x and #nR prefixes for binary, octal, hexadecimal, and
  base-n integers

24 years ago- changed comment starter from # to //
Christian Bauer [Wed, 5 Apr 2000 19:30:47 +0000 (19:30 +0000)]
- changed comment starter from # to //
- implemented #b, #o, #x and #nR prefixes for binary, octal, hexadecimal, and
  base-n integers
- added iprint() command for decimal, octal, and hexadecimal output of integers
- updated man page

24 years ago- atanh_series(): corrected an oops.
Richard Kreckel [Mon, 3 Apr 2000 22:22:10 +0000 (22:22 +0000)]
- atanh_series(): corrected an oops.

24 years ago- Some corrections in series expansion of several functions.
Richard Kreckel [Mon, 3 Apr 2000 22:03:23 +0000 (22:03 +0000)]
- Some corrections in series expansion of several functions.

24 years ago- NOT building the documentation is the default now.
Richard Kreckel [Mon, 3 Apr 2000 22:02:54 +0000 (22:02 +0000)]
- NOT building the documentation is the default now.

24 years ago- inline int csgn(const numeric &): fixed a typo.
Richard Kreckel [Mon, 3 Apr 2000 22:02:17 +0000 (22:02 +0000)]
- inline int csgn(const numeric &): fixed a typo.

24 years ago- Fixed for new input parser scheme.
Richard Kreckel [Mon, 3 Apr 2000 22:01:22 +0000 (22:01 +0000)]
- Fixed for new input parser scheme.

24 years ago- the ex stream >> operator is gone
Christian Bauer [Fri, 31 Mar 2000 03:48:28 +0000 (03:48 +0000)]
- the ex stream >> operator is gone
- added a new ex constructor ex(string, ex) instead; use as follows:
    symbol x("x"), y("y");
    ex e("Pi*(4*x^2+y)^10", lst(x, y));
  or, for interactive input:
    symbol x("x");
    string s;
    cout <<  "Input something (may contain 'x'): ";
    getline(cin, s);
    try {
        e = ex(s, lst(x));
        cout << "You entered " << e << endl;
        cout << "e.diff(x) = " << e.diff(x) << endl;
    } catch (exception &c) {
        cerr << c.what() << endl;
    }
- symbols not specified in the list in the second argument to ex(string, ex)
  but encountered in the input expression will throw an exception

24 years ago- Documentation adjusted to new tgamma-lgamma-Euler-convention.
Richard Kreckel [Fri, 31 Mar 2000 00:30:20 +0000 (00:30 +0000)]
- Documentation adjusted to new tgamma-lgamma-Euler-convention.

24 years ago- dramatic speedup for characteristic polynomials of numerical matrices.
Richard Kreckel [Thu, 30 Mar 2000 00:09:13 +0000 (00:09 +0000)]
- dramatic speedup for characteristic polynomials of numerical matrices.
- speedup for matrix multiplication of somewhat sparse matrices.
- gamma -> Euler  (sorry 4 that, again).
- two new tests (Q and Q') from Lewis' and Wester's testsuite.
- fixed compilation desaster in timing M1 by using the new parser.
- had a glass of beer.

24 years agobug fixed in an incomplete assertion in power.cpp
Alexander Frink [Wed, 29 Mar 2000 22:22:14 +0000 (22:22 +0000)]
bug fixed in an incomplete assertion in power.cpp

24 years ago- parsed functions are now correctly evaluated (.eval() was missing)
Christian Bauer [Wed, 29 Mar 2000 20:22:41 +0000 (20:22 +0000)]
- parsed functions are now correctly evaluated (.eval() was missing)

24 years ago- expressions can now be read from streams; the input expression can contain
Christian Bauer [Wed, 29 Mar 2000 20:20:12 +0000 (20:20 +0000)]
- expressions can now be read from streams; the input expression can contain
  a list of predefined symbols to be used, before reading (and overwriting)
  the expression, e.g:
    symbol x("x"), y("y");
    e = lst(x, y);
    cout << "Enter an expression containing 'x' and 'y': ";
    try {
        cin >> e;
    } catch (exception &e) {
        cerr << e.what() << endl;
    }
    cout << "You entered " << e << endl;
    cout << "de/dx = " << e.diff(x) << endl;
    cout << "de/dy = " << e.diff(y) << endl;

24 years ago- remembering in divide_in_z() turned off (will eventually have to be replaced
Christian Bauer [Tue, 28 Mar 2000 18:58:57 +0000 (18:58 +0000)]
- remembering in divide_in_z() turned off (will eventually have to be replaced
  by the new remembering mechanism in remember.cpp)
- new #define options at top of source:
    USE_TRIAL_DIVISION makes divide_in_z() use trial division and Newton
      interpolation instead of recursive polynomial division (only useful
      for very large problems)
    STATISTICS enables some statistical output at the end of the program
- gcd(numeric, numeric) is now handled earlier; this should speed up normal()
- added new heuristic: gcd(p^n, p^m) with p a polynomial now immediately
  returns p^abs(n-m)
- if GCD is 1, cofactors are returned unexpanded

24 years ago- New tests from the Lewis-Wester Paper: I, J, K and L.
Richard Kreckel [Tue, 28 Mar 2000 00:51:17 +0000 (00:51 +0000)]
- New tests from the Lewis-Wester Paper: I, J, K and L.

24 years ago- Primitive heuristics to make determinants of sparse matrices work.
Richard Kreckel [Mon, 27 Mar 2000 17:43:54 +0000 (17:43 +0000)]
- Primitive heuristics to make determinants of sparse matrices work.

24 years ago- Finally: the test M1 from Lewis and Wester works.
Richard Kreckel [Mon, 27 Mar 2000 17:43:14 +0000 (17:43 +0000)]
- Finally: the test M1 from Lewis and Wester works.

24 years ago- As advertised: we are calling the Gamma function tgamma() now!
Richard Kreckel [Sun, 26 Mar 2000 22:57:01 +0000 (22:57 +0000)]
- As advertised: we are calling the Gamma function tgamma() now!

24 years ago- added test O1 from Lewis-Wester paper.
Richard Kreckel [Sun, 26 Mar 2000 18:45:39 +0000 (18:45 +0000)]
- added test O1 from Lewis-Wester paper.
- corrected a thinko in computing the time.

24 years ago- Two more tests from the Lewis-Wester paper.
Richard Kreckel [Sat, 25 Mar 2000 20:58:05 +0000 (20:58 +0000)]
- Two more tests from the Lewis-Wester paper.

24 years ago- Added some tests from the Lewis-Wester testsuite.
Richard Kreckel [Sat, 25 Mar 2000 20:35:48 +0000 (20:35 +0000)]
- Added some tests from the Lewis-Wester testsuite.

24 years ago- ginac.h includes version.h (created by configure script), which contains
Christian Bauer [Fri, 24 Mar 2000 19:45:38 +0000 (19:45 +0000)]
- ginac.h includes version.h (created by configure script), which contains
  library version defines

24 years ago- raised coefficient bit-length limit in heur_gcd() (CLN is good with large
Christian Bauer [Thu, 23 Mar 2000 21:50:38 +0000 (21:50 +0000)]
- raised coefficient bit-length limit in heur_gcd() (CLN is good with large
  numbers, so this improves performance)

24 years ago- We now write f(x).series(x==3,5) instead of f(x).series(x,3,5) and
Richard Kreckel [Thu, 23 Mar 2000 16:50:17 +0000 (16:50 +0000)]
- We now write f(x).series(x==3,5) instead of f(x).series(x,3,5) and
  f(x).series(x,4) instead of f(x).series(x,0,4).  We also don't allow
  default arguments any more.

24 years ago- removed inert Diff() function; only Derivative() remains
Christian Bauer [Wed, 22 Mar 2000 21:32:23 +0000 (21:32 +0000)]
- removed inert Diff() function; only Derivative() remains
- deriving Derivative() now works

24 years ago- gamma() -> Gamma().
Richard Kreckel [Tue, 21 Mar 2000 19:30:22 +0000 (19:30 +0000)]
- gamma() -> Gamma().
- EulerGamma -> gamma.
- beta() -> Beta().
- Version -> 0.6.0 in order to make it clear there were *real* changes.

24 years ago- 0^a -> 0^a instead of 0.
Richard Kreckel [Tue, 21 Mar 2000 17:07:17 +0000 (17:07 +0000)]
- 0^a -> 0^a instead of 0.

24 years ago- print 0 without parens.
Richard Kreckel [Tue, 21 Mar 2000 17:07:00 +0000 (17:07 +0000)]
- print 0 without parens.

24 years ago- Changes reflect beautifying of floating-point output.
Richard Kreckel [Mon, 20 Mar 2000 17:10:43 +0000 (17:10 +0000)]
- Changes reflect beautifying of floating-point output.

24 years ago- Output of floats is now in a more beautiful form.
Richard Kreckel [Mon, 20 Mar 2000 17:09:50 +0000 (17:09 +0000)]
- Output of floats is now in a more beautiful form.

24 years ago- fixed bug in normal(): normal(x^a) became (x^(-a))^(-1)
Christian Bauer [Mon, 20 Mar 2000 15:17:44 +0000 (15:17 +0000)]
- fixed bug in normal(): normal(x^a) became (x^(-a))^(-1)
- first shot at inert Diff() and Derivative() functions; still buggy!

24 years ago- Mention `warranty;' in splash-screen instead of file COPYING.
Richard Kreckel [Mon, 20 Mar 2000 15:14:39 +0000 (15:14 +0000)]
- Mention `warranty;' in splash-screen instead of file COPYING.

24 years ago- Updated the timings. release_0-5-4
Richard Kreckel [Wed, 15 Mar 2000 23:37:19 +0000 (23:37 +0000)]
- Updated the timings.

24 years ago- Updated for Version 0.5.4.
Richard Kreckel [Wed, 15 Mar 2000 22:44:12 +0000 (22:44 +0000)]
- Updated for Version 0.5.4.

24 years ago- update NEWS file
Christian Bauer [Wed, 15 Mar 2000 22:34:19 +0000 (22:34 +0000)]
- update NEWS file

24 years ago- Updated for 0.5.4.
Richard Kreckel [Wed, 15 Mar 2000 22:32:49 +0000 (22:32 +0000)]
- Updated for 0.5.4.

24 years ago- According to CLTL 0^I is undefined, 0^(I+epsilon) is 0 and 0^(I-epsilon)
Richard Kreckel [Wed, 15 Mar 2000 22:28:41 +0000 (22:28 +0000)]
- According to CLTL 0^I is undefined, 0^(I+epsilon) is 0 and 0^(I-epsilon)
  is an overflow.  Now, this should be honored.

24 years ago- The status_flags::expanded is now used on some occasions.
Richard Kreckel [Wed, 15 Mar 2000 21:22:41 +0000 (21:22 +0000)]
- The status_flags::expanded is now used on some occasions.

24 years ago- New file exam_inifcns.cpp holds some functions that really do not belong
Richard Kreckel [Wed, 15 Mar 2000 21:22:03 +0000 (21:22 +0000)]
- New file exam_inifcns.cpp holds some functions that really do not belong
  to the checks.

24 years ago- Doxygen'ed a bunch of comments.
Richard Kreckel [Tue, 14 Mar 2000 19:39:25 +0000 (19:39 +0000)]
- Doxygen'ed a bunch of comments.

24 years ago- Cleanup and clarifications.
Richard Kreckel [Tue, 14 Mar 2000 19:39:00 +0000 (19:39 +0000)]
- Cleanup and clarifications.
- Addition of sparse expression trees (to be used at a later stage).

24 years ago- ex::series() catches and re-throws exceptions happening during expansion
Christian Bauer [Mon, 13 Mar 2000 19:21:20 +0000 (19:21 +0000)]
- ex::series() catches and re-throws exceptions happening during expansion
  (almost always caused by expansion of unhandled poles)
- series(log(x),x) now returns log(x) instead of an error

24 years ago- More drastic performance boost on matrix stuff.
Richard Kreckel [Mon, 13 Mar 2000 17:15:05 +0000 (17:15 +0000)]
- More drastic performance boost on matrix stuff.
- Addition of some comments.

24 years ago- Two new timimgs that are interesting for optimizing.
Richard Kreckel [Mon, 13 Mar 2000 17:14:21 +0000 (17:14 +0000)]
- Two new timimgs that are interesting for optimizing.
- Readjusted some parameters.
- Cleaned up syntax in everything involving matrices to reflect the
  policy: "normal if in field, expand otherwise".  (It makes many things
  much clearer.)

24 years ago- New example in Tour of GiNaC about exact radicals.
Richard Kreckel [Mon, 13 Mar 2000 17:12:12 +0000 (17:12 +0000)]
- New example in Tour of GiNaC about exact radicals.

24 years ago- Added warning about too much optimism for Bernoulli numbers.
Richard Kreckel [Thu, 9 Mar 2000 18:29:17 +0000 (18:29 +0000)]
- Added warning about too much optimism for Bernoulli numbers.

24 years ago- Changed behaviour of .evalf() to not evalf any exponents.
Richard Kreckel [Thu, 2 Mar 2000 23:51:33 +0000 (23:51 +0000)]
- Changed behaviour of .evalf() to not evalf any exponents.

24 years ago- collect() does nothing
Christian Bauer [Thu, 2 Mar 2000 17:38:32 +0000 (17:38 +0000)]
- collect() does nothing
- evalf() only evalfs the coefficients

24 years ago- The paragraph about regression tests reflects the new 3-step scheme now.
Richard Kreckel [Thu, 2 Mar 2000 15:16:22 +0000 (15:16 +0000)]
- The paragraph about regression tests reflects the new 3-step scheme now.

24 years ago- Fixed a logic error in numeric::info().
Richard Kreckel [Wed, 1 Mar 2000 21:32:45 +0000 (21:32 +0000)]
- Fixed a logic error in numeric::info().
- Maybe added another world record: an extremely fast fibonacci routine
  (a nonrecursive variation of a method suggested by B. Haible).

24 years ago- pseries::coeff() now uses a binary instead of a linear search algorithm
Christian Bauer [Wed, 1 Mar 2000 20:51:01 +0000 (20:51 +0000)]
- pseries::coeff() now uses a binary instead of a linear search algorithm

24 years ago- Completely restructured the checks in subdir check/.
Richard Kreckel [Tue, 29 Feb 2000 17:28:13 +0000 (17:28 +0000)]
- Completely restructured the checks in subdir check/.
- Optimized Laplace expansion of determinants to not compute anything more
  than once.
- Moved int permutation_sign(vector<T> s) from matrix.cpp to utils.h.
- A handful of other fixes assiciated with the above.

24 years ago- Patch for static libreadline w/ libncurses problem, submitted by B.Haible.
Richard Kreckel [Tue, 29 Feb 2000 17:25:03 +0000 (17:25 +0000)]
- Patch for static libreadline w/ libncurses problem, submitted by B.Haible.

24 years ago- default maxargs for functions is 13 to make the current xloops compile
Christian Bauer [Thu, 24 Feb 2000 18:06:40 +0000 (18:06 +0000)]
- default maxargs for functions is 13 to make the current xloops compile
  out-of-the-box
- small fix for abs(): eval() function was not set properly

24 years ago- Speedup for macro that checks if doublefactorial is present in -lcln.
Richard Kreckel [Thu, 24 Feb 2000 15:05:53 +0000 (15:05 +0000)]
- Speedup for macro that checks if doublefactorial is present in -lcln.

24 years ago- Clarification. release_0-5-3
Richard Kreckel [Wed, 23 Feb 2000 20:08:30 +0000 (20:08 +0000)]
- Clarification.

24 years agobetter handling for comments in ginaccint.bin.cpp
Alexander Frink [Wed, 23 Feb 2000 19:56:05 +0000 (19:56 +0000)]
better handling for comments in ginaccint.bin.cpp

24 years ago- Version and date reflect 0.5.3
Richard Kreckel [Wed, 23 Feb 2000 18:38:27 +0000 (18:38 +0000)]
- Version and date reflect 0.5.3

24 years ago- Some minor clarifications.
Richard Kreckel [Wed, 23 Feb 2000 18:04:57 +0000 (18:04 +0000)]
- Some minor clarifications.

24 years ago- Version 0.5.2 -> 0.5.3.
Richard Kreckel [Wed, 23 Feb 2000 18:04:38 +0000 (18:04 +0000)]
- Version 0.5.2 -> 0.5.3.

24 years ago- A new test ripped shamelessly from the Mathematica book.
Richard Kreckel [Wed, 23 Feb 2000 17:36:06 +0000 (17:36 +0000)]
- A new test ripped shamelessly from the Mathematica book.

24 years ago- Version 5.0.2 -> 5.0.3
Richard Kreckel [Wed, 23 Feb 2000 17:35:42 +0000 (17:35 +0000)]
- Version 5.0.2 -> 5.0.3

24 years agoremember strategies least recently used and least frequently used,
Alexander Frink [Tue, 22 Feb 2000 19:53:27 +0000 (19:53 +0000)]
remember strategies least recently used and least frequently used,
code moved to remember.h/cpp

24 years agoregistered_function_info struct changed to function_options class
Alexander Frink [Mon, 21 Feb 2000 23:12:32 +0000 (23:12 +0000)]
registered_function_info struct changed to function_options class

24 years agocontainer.pl: can now generate constructors for an arbitary number
Alexander Frink [Mon, 21 Feb 2000 23:11:41 +0000 (23:11 +0000)]
container.pl: can now generate constructors for an arbitary number
    of arguments
function.pl: new concept for specifying options for functions,
    take a look at the function_options class
    first experimental support for remembering functions
    IMPORTANT: the syntax of the macro REGISTER_FUNCTION() has
    changed incompatibly (see inifcns.cpp for an example)
flags.h: flags for remembering strategies (lru, lfu,...)
inifcns*.cpp: use the new REGISTER_FUNCTION() macro syntax
numeric.h/cpp: hashing for numbers using CLN's cl_equal_hashcode()
    seems to give increased performance
simp_lor.h: simplify_simp_lor can be called without scalar products list
structure.pl: bug fix in regular expression
normal.cpp: bug fix, a_numeric.compare(_ex0)

24 years agosome words about the new function registration macro
Alexander Frink [Mon, 21 Feb 2000 23:04:35 +0000 (23:04 +0000)]
some words about the new function registration macro