- CLN-list - ginac.de

Question on algorithm for pi
by Barry Levin 09 Jun '06

09 Jun '06

I am not a programmer (yet), and I am doing a project for a Calculus class on the derivation of pi. Since this software managed to produce a large amount of digits for me, and although the code is open, I cannot see what it does due to my own lack of skills, I would like to know how it calculates pi.

3 2

Questions?
by Wilson Castro 02 Jun '06

02 Jun '06

Hello CLN list-users, I want to experiment with ring of the form: (Z mod r) + (Z mod s)*sqrt( z ). where r,s > 1 in Z, and z in Z with abs(z) is squarefree. I have done following little program in order to leave in the root the part of z that is quadrat free. // --------------------- Begin ----------------------------------- // void squarefree_descomposition( cl_I n ) { bool bNegative; cl_I raiz; if( n < 0 ) bNegative = true; else bNegative = false; if( bNegative ) n = -n; if( cln::isprobprime( n ) ) { cout << n << " is quadrat frei " << "-- if( cln::isprobprime( n ) ) -- " << endl; } else { if( cln::sqrtp( n, &raiz) ) { cout << n << " is a perfect square and the exact square root is " << raiz << endl; } else { cl_I m; cl_I factor; cln::isqrt( n , &m); while( m > 1 ) { factor = cln::floor1( n, m*m); if( factor*(m*m) == n ) { cout << "sqrt( "<< n <<" ) = " << m << "*sqrt( "<< factor << " )" << endl; return ; } m = m - 1; } cout << n << " is quadrat frei" << endl; } } } // --------------------- End ----------------------------------- // when I invoke the function as I show in following main. // --------------------- Begin main ----------------------------------- // int main(int argc, char *argv[]) { cl_I n; n = "9999999999"; // 10 { CL_TIMING; squarefree_descomposition(n); } n = "99999999999"; // 11 { CL_TIMING; squarefree_descomposition(n); } n = "999999999999"; // 12 { CL_TIMING; squarefree_descomposition(n); } n = "9999999999999"; // 13 { CL_TIMING; squarefree_descomposition(n); } n = "99999999999999"; // 14 { CL_TIMING; squarefree_descomposition(n); } n = "999999999999999"; // 15 { CL_TIMING; squarefree_descomposition(n); } n = "999999999999999"; // 15 { CL_TIMING; squarefree_descomposition(n); } return EXIT_SUCCESS; } // --------------------- End main ----------------------------------- // I obtain the following result. sqrt( 9999999999 ) = 3*sqrt( 1111111111 ) real time: 0.131 s, run time: 0.132 s sqrt( 99999999999 ) = 3*sqrt( 11111111111 ) real time: 0.493 s, run time: 0.456 s sqrt( 999999999999 ) = 3*sqrt( 111111111111 ) real time: 1.533 s, run time: 1.500 s sqrt( 9999999999999 ) = 3*sqrt( 1111111111111 ) real time: 4.798 s, run time: 4.760 s sqrt( 99999999999999 ) = 3*sqrt( 11111111111111 ) real time: 15.180 s, run time: 15.093 s sqrt( 999999999999999 ) = 3*sqrt( 111111111111111 ) real time: 47.838 s, run time: 47.623 s sqrt( 999999999999999 ) = 3*sqrt( 111111111111111 ) real time: 48.248 s, run time: 47.811 s QUESTIONS: 1. because if the length of the two last calls are equal (length (n) = 15), the times reported by “CL_TIMING” are different? it seems to me that the complexity of my approach depends on the primality test of n. 2. There is some limitation in the length of the algument n in the function "cl_boolean isprobprime (const cl_I& n)"? thank you very much, -- Ms. C. Wilson Castro Rojas Foundations of Informatics Group, Department of Informatics, University of Kaiserslautern, Germany, PO Box: 3049, Building: 34/423, Phone: ++49 631 205 2155, Fax: ++49 631 205 2156,

2 3

fix m4 underquoting
by Ralf Wildenhues 30 May '06

30 May '06

Hello CLN authors, :-) Please apply this patch to fix some underquoted Autoconf macro. Cheers, Ralf * m4/gettimeofday.m4 (CL_GETTIMEOFDAY): Fix M4 quoting. Index: m4/gettimeofday.m4 =================================================================== RCS file: /home/cvs/cln/m4/gettimeofday.m4,v retrieving revision 1.1 diff -u -r1.1 gettimeofday.m4 --- m4/gettimeofday.m4 29 Aug 2005 13:18:40 -0000 1.1 +++ m4/gettimeofday.m4 30 May 2006 21:07:09 -0000 @@ -1,5 +1,5 @@ dnl -*- Autoconf -*- -dnl Copyright (C) 1993-2003 Free Software Foundation, Inc. +dnl Copyright (C) 1993-2003, 2006 Free Software Foundation, Inc. dnl This file is free software, distributed under the terms of the GNU dnl General Public License. As a special exception to the GNU General dnl Public License, this file may be distributed as part of a program @@ -22,7 +22,7 @@ ], [int gettimeofday (struct timeval * tp, struct timezone * tzp);], [int gettimeofday();], cl_cv_proto_gettimeofday_dots=no -cl_cv_proto_gettimeofday_arg2="struct timezone *", +cl_cv_proto_gettimeofday_arg2="struct timezone *", [ CL_PROTO_TRY([ #include <sys/types.h> #include <sys/time.h> @@ -31,7 +31,7 @@ cl_cv_proto_gettimeofday_dots=no cl_cv_proto_gettimeofday_arg2="void *", cl_cv_proto_gettimeofday_dots=yes -cl_cv_proto_gettimeofday_arg2="...")) +cl_cv_proto_gettimeofday_arg2="...")]) ], [extern int gettimeofday (struct timeval *, $cl_cv_proto_gettimeofday_arg2);]) if test $cl_cv_proto_gettimeofday_dots = yes; then AC_DEFINE(GETTIMEOFDAY_DOTS,,[declaration of gettimeofday() needs dots])

2 1

Python Interface to Class Library for Numbers
by Ray Buvel 06 May '06

06 May '06

The clnum package provides a Python interface to the Class Library for Numbers. There is now a binary installer and build procedure available for the Windows platform. For more information see http://calcrpnpy.sourceforge.net/clnumManual.html Ray Buvel

1 0

Re: Compiling CLN on Intel Macs
by Christian Fleck 28 Apr '06

28 Apr '06

Hi Richard, here is the backtrace: Program received signal EXC_BAD_ACCESS, Could not access memory. Reason: KERN_INVALID_ADDRESS at address: 0x06a0afe1 0xa0b02584 in std::wclog () (gdb) bt #0 0xa0b02584 in std::wclog () #1 0xa0b025f0 in std::wclog () #2 0x00009db7 in global constructors keyed to cl_module__cl_prin_globals__firstglobalfun () at ./base/output/ cl_prin_globals.cc:33 #3 0x00009c9a in __static_initialization_and_destruction_0 (__initialize_p=1, __priority=65535) at ../include/cln/io.h:91 #4 0x00009db7 in global constructors keyed to cl_module__cl_prin_globals__firstglobalfun () at ./base/output/ cl_prin_globals.cc:33 #5 0x000026f4 in __static_initialization_and_destruction_0 (__initialize_p=2, __priority=65535) at ../include/cln/io.h:91 #6 0x0000275d in global constructors keyed to _Z15test_elementaryv () at exam.cc:51 #7 0x8fe10168 in __dyld__ZN16ImageLoaderMachO16doInitializationERKN11ImageLoader11LinkCon textE () #8 0x8fe0a516 in __dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContextE () #9 0x8fe0ca08 in __dyld__ZN11ImageLoader15runInitializersERKNS_11LinkContextE () #10 0x8fe031b4 in __dyld__ZN4dyld24initializeMainExecutableEv () #11 0x0000264d in _call_mod_init_funcs () at /SourceCache/Csu/Csu-58/ crt.c:304 #12 0x00002579 in _start (argc=1, argv=0xbffff86c, envp=0xbffff874) at /SourceCache/Csu/Csu-58/crt.c:222 #13 0x00002529 in start () Christian

4 4

Use datarootdir
by Ralf Wildenhues 25 Apr '06

25 Apr '06

The other change for Autoconf-2.60: If you bootstrap CLN with Autoconf-2.59c, config.status will warn you: | config.status: creating Makefile | config.status: WARNING: ../cln/Makefile.in seems to ignore the --datarootdir setting | config.status: creating src/Makefile | config.status: WARNING: ../cln/src/Makefile.in seems to ignore the --datarootdir setting | config.status: creating tests/Makefile | config.status: creating benchmarks/Makefile | config.status: creating examples/Makefile | config.status: creating doc/Makefile | config.status: WARNING: ../cln/doc/Makefile.in seems to ignore the --datarootdir setting and turn on the backward compatibility setting for those Makefiles. The backward-compatible patch below fixes this. :-) Cheers, Ralf * Makefile.in: Set datarootdir. * doc/Makefile.in, src/Makefile.in: Likewise. Index: Makefile.in =================================================================== RCS file: /home/cvs/cln/Makefile.in,v retrieving revision 1.13 diff -u -r1.13 Makefile.in --- Makefile.in 10 Jun 2005 16:16:33 -0000 1.13 +++ Makefile.in 25 Apr 2006 16:05:08 -0000 @@ -8,6 +8,7 @@ exec_prefix = @exec_prefix@ srcdir = @srcdir@ bindir = @bindir@ +datarootdir = @datarootdir@ datadir = @datadir@ libdir = @libdir@ includedir = @includedir@ Index: doc/Makefile.in =================================================================== RCS file: /home/cvs/cln/doc/Makefile.in,v retrieving revision 1.7 diff -u -r1.7 Makefile.in --- doc/Makefile.in 26 Oct 2004 22:23:41 -0000 1.7 +++ doc/Makefile.in 25 Apr 2006 16:05:08 -0000 @@ -9,6 +9,7 @@ prefix = @prefix@ local_prefix = /usr/local exec_prefix = @exec_prefix@ +datarootdir = @datarootdir@ datadir = @datadir@ infodir = @infodir@ mandir = @mandir@ Index: src/Makefile.in =================================================================== RCS file: /home/cvs/cln/src/Makefile.in,v retrieving revision 1.18 diff -u -r1.18 Makefile.in --- src/Makefile.in 22 Oct 2005 09:09:02 -0000 1.18 +++ src/Makefile.in 25 Apr 2006 16:05:08 -0000 @@ -10,6 +10,7 @@ prefix = @prefix@ local_prefix = /usr/local exec_prefix = @exec_prefix@ +datarootdir = @datarootdir@ datadir = @datadir@ libdir = @libdir@ includedir = @includedir@

2 1

Compilation Error - `cl_2D_div.lo' is not a valid libtool object
by Sonia Singhal 22 Apr '06

22 Apr '06

Hi, These are my system specs >uname -a Linux frodo 2.6.11.4-20a-default #1 Wed Mar 23 21:52:37 UTC 2005 i686 unknown unknown GNU/Linux >g++ --version g++-3.4.3 (GCC) 3.4.3 I am using the latest version of CLN ie cln-1.1.11 while 'make'ing I get this error. Any help/inputs? I see that file is not compiled then why is it being linked ? /bin/sh ../libtool --mode=link g++ -o libcln.la -rpath /usr/local/lib -version-info 4:1:0 -objectlist libcln.objectlist libtool: link: `cl_2D_div.lo' is not a valid libtool object make[1]: *** [libcln.la] Error 1 make[1]: Leaving directory `/home/soniasin/software/cln-1.1.11/src' make: *** [all] Error 2 Thanks.

2 1

[patch] do not use `exit' in tests
by Ralf Wildenhues 19 Apr '06

19 Apr '06

Upcoming Autoconf-2.60 will not provide a declaration for exit; it has turned out that tracking MSVC incompatibilites here is a lot of work. Instead, returning from main is suggested as a viable alternative (the systems where that was not portable are not likely to be used any more). The following patch does that for the macros in CLN. Cheers, Ralf * m4/longdouble.m4 (CL_LONGDOUBLE): `return', rather than `exit' from main. * m4/longlong.m4 (CL_LONGLONG): Likewise. * m4/times.m4 (CL_TIMES_CLOCK): Likewise. * m4/general.m4 (CL_CC_WORKS): Likewise, do this for a C test to work when users set `$CC' to a C++ compiler. Index: m4/general.m4 =================================================================== RCS file: /home/cvs/cln/m4/general.m4,v retrieving revision 1.1 diff -u -r1.1 general.m4 --- m4/general.m4 29 Aug 2005 13:18:40 -0000 1.1 +++ m4/general.m4 11 Apr 2006 10:04:22 -0000 @@ -93,7 +93,7 @@ [AC_CACHE_CHECK(whether CC works at all, cl_cv_prog_cc_works, [ AC_LANG_SAVE() AC_LANG_C() -AC_TRY_RUN([int main() { exit(0); }], +AC_TRY_RUN([int main() { return 0; }], cl_cv_prog_cc_works=yes, cl_cv_prog_cc_works=no, AC_TRY_LINK([], [], cl_cv_prog_cc_works=yes, cl_cv_prog_cc_works=no)) AC_LANG_RESTORE() Index: m4/longdouble.m4 =================================================================== RCS file: /home/cvs/cln/m4/longdouble.m4,v retrieving revision 1.1 diff -u -r1.1 longdouble.m4 --- m4/longdouble.m4 29 Aug 2005 13:18:40 -0000 1.1 +++ m4/longdouble.m4 11 Apr 2006 10:04:22 -0000 @@ -12,7 +12,7 @@ AC_DEFUN([CL_LONGDOUBLE], [AC_CACHE_CHECK(for long double type, cl_cv_c_longdouble, [ AC_TRY_RUN([int main() -{ long double x = 2.7182818284590452354L; x = x*x; exit (x==0.0L); }], +{ long double x = 2.7182818284590452354L; x = x*x; return x==0.0L; }], cl_cv_c_longdouble=yes, cl_cv_c_longdouble=no, [ dnl When cross-compiling, use the test from gnulib. AC_TRY_COMPILE([ Index: m4/longlong.m4 =================================================================== RCS file: /home/cvs/cln/m4/longlong.m4,v retrieving revision 1.1 diff -u -r1.1 longlong.m4 --- m4/longlong.m4 29 Aug 2005 13:18:40 -0000 1.1 +++ m4/longlong.m4 11 Apr 2006 10:04:22 -0000 @@ -19,11 +19,11 @@ miscompiled. */ #if defined(__m68k__) || (defined(_IBMR2) || defined(__powerpc)) #if defined(__GNUC__) && (__GNUC__ == 2) && (__GNUC_MINOR__ <= 7) - exit(1); + return 1; #endif #endif { long x = 944938507; long y = 737962842; long z = 162359677; - exit(!(((long long) x)*((long long) y)>>32 == z)); + return !(((long long) x)*((long long) y)>>32 == z); } }], cl_cv_c_longlong=yes, cl_cv_c_longlong=no, [ Index: m4/times.m4 =================================================================== RCS file: /home/cvs/cln/m4/times.m4,v retrieving revision 1.1 diff -u -r1.1 times.m4 --- m4/times.m4 29 Aug 2005 13:18:40 -0000 1.1 +++ m4/times.m4 11 Apr 2006 10:04:22 -0000 @@ -29,12 +29,12 @@ clock_t result2; int ticks; result1 = times(&buffer); - if ((result1 == (clock_t)0) || (result1 == (clock_t)(-1))) exit(1); + if ((result1 == (clock_t)0) || (result1 == (clock_t)(-1))) return 1; sleep(1); result2 = times(&buffer); - if ((result2 == (clock_t)0) || (result2 == (clock_t)(-1))) exit(1); + if ((result2 == (clock_t)0) || (result2 == (clock_t)(-1))) return 1; ticks = result2 - result1; - exit(!((ticks >= CLK_TCK/2) && (ticks <= 3*CLK_TCK/2))); + return !((ticks >= CLK_TCK/2) && (ticks <= 3*CLK_TCK/2)); }], cl_cv_func_times_return=yes, cl_cv_func_times_return=no, dnl When cross-compiling, don't assume anything. cl_cv_func_times_return="guessing no")

2 1

Help - more info
by Wilson Castro Rojas 21 Mar '06

21 Mar '06

for complete the information: these are the data-types: cl_modint_ring modintRing; // Z/mZ cl_I intArg1; cl_MI intMod;

2 1

Help
by Wilson Castro Rojas 21 Mar '06

21 Mar '06

Hellow, I'm new with the cln library. Before the installation from the version 1.1.11 the follow code worked. ------------------ cl_I intRing = m_strValue.toAscii().constData(); intArg1 = m_sArg1.toAscii().constData(); modintRing = find_modint_ring( intArg1 ); intMod = modintRing->canonhom( intRing ); cout << intRing << "\n"; cout << intArg1 << "\n"; cout << intMod << "\n"; -------------------- but now in this code-line cout << intMod << "\n"; there is a "Segmentation fault" error report. Thanks and best regards, Wilson

1 0