Linux Certif

Toute la documentation sur la certification Linux LPI

ginsh

NAME

ginsh - GiNaC Interactive Shell

SYNPOSIS

ginsh [file...]

DESCRIPTION

ginsh is an interactive frontend for the GiNaC symbolic computation framework. It is intended as a tool for testing and experimenting with GiNaC's features, not as a replacement for traditional interactive computer algebra systems. Although it can do many things these traditional systems can do, ginsh provides no programming constructs like loops or conditional expressions. If you need this functionality you are advised to write your program in C++, using the "native" GiNaC class framework.

USAGE

INPUT FORMAT

After startup, ginsh displays a prompt ("> ") signifying that it is ready to accept your input. Acceptable input are numeric or symbolic expressions consisting of numbers (e.g. 42, 2/3 or 0.17), symbols (e.g. x or result), mathematical operators like + and *, and functions (e.g. sin or normal). Every input expression must be terminated with either a semicolon (;) or a colon (:). If terminated with a semicolon, ginsh will evaluate the expression and print the result to stdout. If terminated with a colon, ginsh will only evaluate the expression but not print the result. It is possible to enter multiple expressions on one line. Whitespace (spaces, tabs, newlines) can be applied freely between tokens. To quit ginsh, enter quit or exit, or type an EOF (Ctrl-D) at the prompt.

COMMENTS

Anything following a double slash (//) up to the end of the line, and all lines starting with a hash mark (#) are treated as a comment and ignored.

NUMBERS

ginsh accepts numbers in the usual decimal notations. This includes arbitrary precision integers and rationals as well as floating point numbers in standard or scientific notation (e.g. 1.2E6). The general rule is that if a number contains a decimal point (.), it is an (inexact) floating point number; otherwise it is an (exact) integer or rational. Integers can be specified in binary, octal, hexadecimal or arbitrary (2-36) base by prefixing them with #b, #o, #x, or #nR , respectively.

SYMBOLS

Symbols are made up of a string of alphanumeric characters and the underscore (_), with the first character being non-numeric. E.g. a and mu_1 are acceptable symbol names, while 2pi is not. It is possible to use symbols with the same names as functions (e.g. sin); ginsh is able to distinguish between the two.

Symbols can be assigned values by entering

symbol = expression;

To unassign the value of an assigned symbol, type

unassign('symbol');

Assigned symbols are automatically evaluated (= replaced by their assigned value) when they are used. To refer to the unevaluated symbol, put single quotes (') around the name, as demonstrated for the "unassign" command above.

Symbols are considered to be in the complex domain by default, i.e. they are treated as if they stand in for complex numbers. This behavior can be changed by using the keywords real_symbols and complex_symbols and affects all newly created symbols.

The following symbols are pre-defined constants that cannot be assigned a value by the user:

Pi

Archimedes' Constant

Catalan

Catalan's Constant

Euler

Euler-Mascheroni Constant

I

sqrt(-1)

FAIL

an object of the GiNaC "fail" class

There is also the special

Digits

symbol that controls the numeric precision of calculations with inexact numbers. Assigning an integer value to digits will change the precision to the given number of decimal places.

WILDCARDS

The has(), find(), match() and subs() functions accept wildcards as placeholders for expressions. These have the syntax

$number

for example $0, $1 etc.

LAST PRINTED EXPRESSIONS

ginsh provides the three special symbols

%, %% and %%%

that refer to the last, second last, and third last printed expression, respectively. These are handy if you want to use the results of previous computations in a new expression.

OPERATORS

ginsh provides the following operators, listed in falling order of precedence:

!

postfix factorial

^

powering

+

unary plus

-

unary minus

*

multiplication

/

division

+

addition

-

subtraction

<

less than

>

greater than

<=

less or equal

>=

greater or equal

==

equal

!=

not equal

=

symbol assignment

All binary operators are left-associative, with the exception of ^ and = which are right-associative. The result of the assignment operator (=) is its right-hand side, so it's possible to assign multiple symbols in one expression (e.g. a = b = c = 2;).

LISTS

Lists are used by the subs and lsolve functions. A list consists of an opening curly brace ({), a (possibly empty) comma-separated sequence of expressions, and a closing curly brace (}).

MATRICES

A matrix consists of an opening square bracket ([), a non-empty comma-separated sequence of matrix rows, and a closing square bracket (]). Each matrix row consists of an opening square bracket ([), a non-empty comma-separated sequence of expressions, and a closing square bracket (]). If the rows of a matrix are not of the same length, the width of the matrix becomes that of the longest row and shorter rows are filled up at the end with elements of value zero.

FUNCTIONS

A function call in ginsh has the form

name(arguments)

where arguments is a comma-separated sequence of expressions. ginsh provides a couple of built-in functions and also "imports" all symbolic functions defined by GiNaC and additional libraries. There is no way to define your own functions other than linking ginsh against a library that defines symbolic GiNaC functions.

ginsh provides Tab-completion on function names: if you type the first part of a function name, hitting Tab will complete the name if possible. If the part you typed is not unique, hitting Tab again will display a list of matching functions. Hitting Tab twice at the prompt will display the list of all available functions.

A list of the built-in functions follows. They nearly all work as the respective GiNaC methods of the same name, so I will not describe them in detail here. Please refer to the GiNaC documentation.

charpoly(matrix, symbol) - characteristic polynomial of a matrix
coeff(expression, object, number) - extracts coefficient of object^number from a polynomial
collect(expression, object-or-list) - collects coefficients of like powers (result in recursive form)
collect_distributed(expression, list) - collects coefficients of like powers (result in distributed form)
collect_common_factors(expression) - collects common factors from the terms of sums
conjugate(expression) - complex conjugation
content(expression, symbol) - content part of a polynomial
decomp_rational(expression, symbol) - decompose rational function into polynomial and proper rational function
degree(expression, object) - degree of a polynomial
denom(expression) - denominator of a rational function
determinant(matrix) - determinant of a matrix
diag(expression...) - constructs diagonal matrix
diff(expression, symbol [, number]) - partial differentiation
divide(expression, expression) - exact polynomial division
eval(expression [, level]) - evaluates an expression, replacing symbols by their assigned value
evalf(expression [, level]) - evaluates an expression to a floating point number
evalm(expression) - evaluates sums, products and integer powers of matrices
expand(expression) - expands an expression
factor(expression) - factorizes an expression (univariate)
find(expression, pattern) - returns a list of all occurrences of a pattern in an expression
fsolve(expression, symbol, number, number) - numerically find root of a real-valued function within an interval
gcd(expression, expression) - greatest common divisor
has(expression, pattern) - returns "1" if the first expression contains the pattern as a subexpression, "0" otherwise
integer_content(expression) - integer content of a polynomial
inverse(matrix) - inverse of a matrix
is(relation) - returns "1" if the relation is true, "0" otherwise (false or undecided)
lcm(expression, expression) - least common multiple
lcoeff(expression, object) - leading coefficient of a polynomial
ldegree(expression, object) - low degree of a polynomial
lsolve(equation-list, symbol-list) - solve system of linear equations
map(expression, pattern) - apply function to each operand; the function to be applied is specified as a pattern with the "$0" wildcard standing for the operands
match(expression, pattern) - check whether expression matches a pattern; returns a list of wildcard substitutions or "FAIL" if there is no match
nops(expression) - number of operands in expression
normal(expression [, level]) - rational function normalization
numer(expression) - numerator of a rational function
numer_denom(expression) - numerator and denumerator of a rational function as a list
op(expression, number) - extract operand from expression
power(expr1, expr2) - exponentiation (equivalent to writing expr1^expr2)
prem(expression, expression, symbol) - pseudo-remainder of polynomials
primpart(expression, symbol) - primitive part of a polynomial
quo(expression, expression, symbol) - quotient of polynomials
rank(matrix) - rank of a matrix
rem(expression, expression, symbol) - remainder of polynomials
resultant(expression, expression, symbol) - resultant of two polynomials with respect to symbol s
series(expression, relation-or-symbol, order) - series expansion
sprem(expression, expression, symbol) - sparse pseudo-remainder of polynomials
sqrfree(expression [, symbol-list]) - square-free factorization of a polynomial
sqrt(expression) - square root
subs(expression, relation-or-list)
subs(expression, look-for-list, replace-by-list) - substitute subexpressions (you may use wildcards)
tcoeff(expression, object) - trailing coefficient of a polynomial
time(expression) - returns the time in seconds needed to evaluate the given expression
trace(matrix) - trace of a matrix
transpose(matrix) - transpose of a matrix
unassign('symbol') - unassign an assigned symbol (mind the quotes, please!)
unit(expression, symbol) - unit part of a polynomial

SPECIAL COMMANDS

To exit ginsh, enter

quit

or

exit

ginsh can display a (short) help for a given topic (mostly about functions and operators) by entering

?topic

Typing

??

will display a list of available help topics.

The command

print(expression);

will print a dump of GiNaC's internal representation for the given expression. This is useful for debugging and for learning about GiNaC internals.

The command

print_latex(expression);

prints a LaTeX representation of the given expression.

The command

print_csrc(expression);

prints the given expression in a way that can be used in a C or C++ program.

The command

iprint(expression);

prints the given expression (which must evaluate to an integer) in decimal, octal, and hexadecimal representations.

Finally, the shell escape

! [command [arguments]]

passes the given command and optionally arguments to the shell for execution. With this method, you can execute shell commands from within ginsh without having to quit.

EXAMPLES

 > a = x^2-x-2;
 -2-x+x^2
 > b = (x+1)^2;
 (x+1)^2
 > s = a/b;
 (x+1)^(-2)*(-2-x+x^2)
 > diff(s, x);
 (2*x-1)*(x+1)^(-2)-2*(x+1)^(-3)*(-x+x^2-2)
 > normal(s);
 (x-2)*(x+1)^(-1)
 > x = 3^50;
 717897987691852588770249
 > s;
 717897987691852588770247/717897987691852588770250
 > Digits = 40;
 40
 > evalf(s);
 0.999999999999999999999995821133292704384960990679
 > unassign('x');
 x
 > s;
 (x+1)^(-2)*(-x+x^2-2)
 > series(sin(x),x==0,6);
 1*x+(-1/6)*x^3+1/120*x^5+Order(x^6)
 > lsolve({3*x+5*y == 7}, {x, y});
 {x==-5/3*y+7/3,y==y}
 > lsolve({3*x+5*y == 7, -2*x+10*y == -5}, {x, y});
 {x==19/8,y==-1/40}
 > M = [ [a, b], [c, d] ];
 [[-x+x^2-2,(x+1)^2],[c,d]]
 > determinant(M);
 -2*d-2*x*c-x^2*c-x*d+x^2*d-c
 > collect(%, x);
 (-d-2*c)*x+(d-c)*x^2-2*d-c
 > solve quantum field theory;
 parse error at quantum
 > quit
 

DIAGNOSTICS

parse error at foo

You entered something which ginsh was unable to parse. Please check the syntax of your input and try again.

argument num to function must be a type

The argument number num to the given function must be of a certain type (e.g. a symbol, or a list). The first argument has number 0, the second argument number 1, etc.

AUTHOR

The GiNaC Group:

Christian Bauer <Christian.Bauer@uni-mainz.de>
Alexander Frink <Alexander.Frink@uni-mainz.de>
Richard Kreckel <Richard.Kreckel@uni-mainz.de>
Jens Vollinga <vollinga@thep.physik.uni-mainz.de>

SEE ALSO

GiNaC Tutorial - An open framework for symbolic computation within the C++ programming language

CLN - A Class Library for Numbers, Bruno Haible

COPYRIGHT

Copyright © 1999-2009 Johannes Gutenberg Universit:at Mainz, Germany

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.