October 2006 - GiNaC-list

amd64 - test failed
by Peper 22 Oct '06

22 Oct '06

Hello, test failed on amd64, any idea? GiNaC will now take an exam with specific input (like a pupils' exam): examining several historic failures just out of paranoia................ passed examining consistency of numeric types.....Error: caught exception syntax error, unexpected ')', expecting ']' or ',' at ) examining power laws..... passed examining consistency of symbolic functions.... passed examining consistency of nestedsums functions..................................Error: caught exception syntax error, unexpected T_EOF, expecting ']' or ',' at examining symbolic differentiationError: caught exception syntax error, unexpected T_EOF, expecting ']' or ',' at examining polynomial GCD computation......... passed examining rational function normalization..... passed examining series expansion............ passed examining symbolic matrix manipulations........ passed examining linear solve...... passed examining indexed objects........ passed examining color objects... passed examining clifford objects.......... passed examining archiving system. passed examining structure template.. passed examining hash maps............ passed examining miscellaneous other things........ passed Error: something went wrong. (3 individual failures) please check exams.out against exams.ref for more details. happy debugging! ./exams.ref exams.out differ: byte 90, line 4 FAIL: run_exams GiNaC will now run through some rather costly random consistency checks: checking consistency of numeric types.. passed checking consistency of symbolic functions.... passed checking symbolic matrix manipulations..... passed checking linear solve.......... passed PASS: run_checks GiNaC will now run through some basic timings: timing commutative expansion and substitution.... passed size: 50 100 200 400 time/s: 0.03 0.13 0.54 2.42 timing Laurent series expansion of Gamma function.... passed order: 10 15 20 25 time/s: 0.05 0.09 0.38 1.47 timing determinant of univariate symbolic Vandermonde matrices.... passed dim: 6x6 8x8 10x10 12x12 time/s: 0.006 0.05 0.5 4.61 timing determinant of polyvariate symbolic Toeplitz matricesError: caught exception syntax error, unexpected T_EOF, expecting ']' or ',' at timing hash map operations.... passed size: 10000 50000 100000 500000 insert/s: 0.02 0.09 0.19 0.86 find/s: 0.01 0.05 0.1 0.5 erase/s: 0.01 0.04 0.08 0.43 timing Lewis-Wester test A (divide factorials). passed 0.05s timing Lewis-Wester test B (sum of rational numbers). passed 0.003s timing Lewis-Wester test C (gcd of big integers). passed 0.024s timing Lewis-Wester test D (normalized sum of rational fcns). passed 0.08s timing Lewis-Wester test E (normalized sum of rational fcns). passed 0.04s timing Lewis-Wester test F (gcd of 2-var polys). passed 0.012s timing Lewis-Wester test G (gcd of 3-var polys). passed 0.38s timing Lewis-Wester test H (det of 80x80 Hilbert). passed 1.24s timing Lewis-Wester test I (invert rank 40 Hilbert). passed 0.37s timing Lewis-Wester test J (check rank 40 Hilbert). passed 0.22s timing Lewis-Wester test K (invert rank 70 Hilbert). passed 2.4s timing Lewis-Wester test L (check rank 70 Hilbert). passed 1.28s timing Lewis-Wester test M1 (26x26 sparse, det)Error: caught exception parser stack overflow at ^ timing Lewis-Wester test M2 (101x101 sparse, det) disabled timing Lewis-Wester test N (poly at rational fcns) disabled timing Lewis-Wester test O1 (three 15x15 dets)... passed 12.486s (average) timing Lewis-Wester test O2 (Resultant) disabled timing Lewis-Wester test P (det of sparse rank 101). passed 0.56s timing Lewis-Wester test P' (det of less sparse rank 101). passed 0.84s timing Lewis-Wester test Q (charpoly(P)). passed 19.01s timing Lewis-Wester test Q' (charpoly(P')). passed 37.2s timing computation of antipodes in Yukawa theory...... passed 21.85s (total) timing Fateman's polynomial expand benchmark. passed 47.85s Error: something went wrong. (2 individual failures) please check times.out against times.ref for more details. happy debugging! ./times.ref times.out differ: byte 261, line 8 FAIL: run_times ====================================== 2 of 3 tests failed Please report to <ginac-list(a)ginac.de> ====================================== make[3]: *** [check-TESTS] Error 1 make[3]: Leaving directory `/var/tmp/portage/sci-mathematics/ginac-1.3.4/work/ginac-1.3.4/check' make[2]: *** [check-am] Error 2 make[2]: Leaving directory `/var/tmp/portage/sci-mathematics/ginac-1.3.4/work/ginac-1.3.4/check' make[1]: *** [check-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/sci-mathematics/ginac-1.3.4/work/ginac-1.3.4' make: *** [check] Error 2

2 1

Suggestion for the FAQ.
by Chris Dams 17 Oct '06

17 Oct '06

Dear all, I think the following should added to the FAQ. Q: I am getting an error message that looks more or less like Internal error: statement in file real/conv/cl_R_to_double.cc, line 55 has been reached!! Please send the authors of the program a description how you produced this error! What should I do? A: This is an error message of the CLN library that GiNaC uses to handle its numerics. The problem usually is that an object was passed to a CLN-function that should have been a number but is not. Your best bet is to check for code that looks like ex_to<numeric>(x) and to check whether x really is a number. The just given code fragment cannot be right if is_a<numeric>(x) yields false. Thanks! Chris

4 4

hash tables and sets
by Marko Riedel 12 Oct '06

12 Oct '06

Hi folks, I would like to write a set implementation for expressions, or borrow one, if it has already been written. I have several questions. 1. What C++ type is best suited to representing sets? 2. Is there a hash function implemented for GiNaC expressions? 3. Can GiNaC expressions be put into some kind of canonical form? 4. (Requirement.) The hash function should be such that e.g. a + b + c * (d + e) and b + (e + d) * c + a hash to the same value. Thanks! Marko Riedel +-------------------------------------------------------------+ | Marko Riedel, EDV Neue Arbeit gGmbH, mriedel(a)lsi.upc.edu | | http://www.geocities.com/markoriedelde/index.html | +-------------------------------------------------------------+

3 7

Re: GiNaC error-message
by Chris Dams 11 Oct '06

11 Oct '06

Dear Martijn, On Tue, 10 Oct 2006, Martijn Frijlink wrote: > I am using GiNaC 1.3.3 and while running the executable of the attached > file blockrec.cc I got the following message : > > $ ./blockrec 11 > Discrete frequencies 0 ... 10 will be used > > Internal error: statement in file real/conv/cl_R_to_double.cc, line 55 > has been reached!! > Please send the authors of the program a description how you produced > this error! The problem is the line CosC[i] = ex_to<numeric>(Lsigma*a_m).to_double(); The expression Lsigma*a_m is not a numeric but e.g. something like 6*sqrt(3)*Pi^(-1). To evaluate this into a numeric you have to use evalf. Best wishes, Chris

1 0