Richard Kreckel [Sat, 23 Sep 2000 21:36:50 +0000 (21:36 +0000)]
- pseries::print(): did not insert parenthesis when needed for precedence.
- pseries::precedence: was missing (thus being inherited from basic).
- add::print(): was not honoring upper_precedence for rest.
Richard Kreckel [Thu, 10 Aug 2000 17:14:39 +0000 (17:14 +0000)]
- Complete revamp of methods in class matrix. Some redundant (and poor)
implementations of elimination schemes were thrown out. The code is now
highly orthogonal, more flexible and much more efficient. The function
lsolve() now falls back to matrix::solve(), which in turn calls different
elimination strategies. matrix::fraction_free_elim() was thrown out for
good. matrix::pivot() has now two integer arguments (ro and co) to allow
pivoting when eliminating to echelon form.
- Some new checks to guard against regressions in the new matrix code.
- Some minor bug-fixes in documentation.
- Bumped up version from 0.6.3 to 0.6.4.
- ...and the usual changes triggered by other changes.
Christian Bauer [Mon, 31 Jul 2000 21:09:44 +0000 (21:09 +0000)]
- I'm fired. Fixed a really stupid bug in sr_gcd() which called
divide_in_z(r, foo) with "var+1" as its last parameter, while the main
variable x=*var usually appears in r. This should never have worked...
Richard Kreckel [Sat, 22 Jul 2000 04:47:04 +0000 (04:47 +0000)]
- Fixed the bug that broke xloop's po_redux.
- Interface Change: No argument 'bool branchcut' for series methods,
instead 'unsigned options'. Sorry.
: ----------------------------------------------------------------------
Richard Kreckel [Fri, 21 Jul 2000 01:12:25 +0000 (01:12 +0000)]
- Derivatives are now assembled in a slightly different manner (i.e. they
'look' different on first sight). Under certain circumstances this can
result in a dramatic speedup because it gives hashing a better chance,
especially when computing higher derivatives.
- Series expansion accepts an additional bool argument now, telling it
whether branch cuts are to be honored (as Maple does it) or not (as
Mathematica does it). It defaults to true (i.e. the Maple way).
- Many series expansions of built-in functions have been reengineered.
- The Dilogarithm (Li2) now has floating point evaluation, derivative and a
proper series expansion.
- Namespace 'std' cleanly disentangled, as demanded by ISO/EIC 14882-1998(E).
- Some helpers updated from upstream.
Richard Kreckel [Thu, 6 Jul 2000 23:34:42 +0000 (23:34 +0000)]
- Introduced derivative of Li2.
- Introduced nontrivial part of series expansion of Li2 around 0. (Read the
comments in the source to see why it's nontrivial!) Missing: the branch
point and cut.
Richard Kreckel [Thu, 15 Jun 2000 22:46:10 +0000 (22:46 +0000)]
- cint/*: Revamped the launch-scheme for ginaccint.bin. It must be done
by a binary executable and not by a script to ensure users can write
shebang-scripts.
- cint/*: Added a testsuite which hopefully passes (some lucky day).
- cint/ginaccint.bin: For this to work we needed a --silent command line
switch.
- configure.in, Makefile.am, cint/Makefile.am: account for new launchers.
- ginac/matrix.cpp, ginac/matrix.h: removed confusing ffe_get() and ffe_set()
routines since they counted the non-standard way.
- config.guess, config.sub: Updated to new versions from FSF.
- others files: Changes triggered.
Richard Kreckel [Mon, 5 Jun 2000 13:45:14 +0000 (13:45 +0000)]
- ginac/operators.cpp: added some missing debugmsg()'s, which would have
come in handy if they had been around before Cint's show-stopper bug. ;-)
- INSTALL, cint/ginaccint.bin.cpp: Updated Cint-copyright.
- ginac/*: deleted some #ifdef CINT_CONVERSION_WORKAROUND's.
Richard Kreckel [Thu, 18 May 2000 14:14:21 +0000 (14:14 +0000)]
- Bumped up version to 0.6.1, incremented binary- and interface-age.
- Added a word of warning for --with-cint and GCC-2.95.x. (The conversion
discards qualifier fiasco in Cint.)
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.
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)