File: facexp.dem

package info (click to toggle)
maxima 5.42.1-1
  • links: PTS
  • area: main
  • in suites: buster
  • size: 150,192 kB
  • sloc: lisp: 382,565; fortran: 14,666; perl: 14,365; tcl: 11,123; sh: 4,622; makefile: 2,688; ansic: 444; xml: 23; awk: 17; sed: 17
file content (155 lines) | stat: -rw-r--r-- 6,832 bytes parent folder | download | duplicates (16)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
/*  First load the necessary file: */
load("facexp");
/* Here is an expression to work with: */
exp1:d*e*f^2+c*e*f^2-d*e+c*e+b*c*d+a*c*d;
/* The function FACSUM can be used to cast an expression
	into a form in which it is fully expanded with respect
	to a specified set of variables, but factored with
	respect to all other variables: */
facsum(exp1,e,f);
/*  For the next example, we use EXP2: */
exp2:c*d*(x*y*z+w*y*z+u*v^2*z-u*z+u*v^2*y+u*y)
	+d*e*f^2+c*e*f^2-d*e+c*e+b*c*d;
/* Another mode in which FACSUM is useful results in a form
	 in which a second set of variables plays the same role as
	 above, but one level deeper in the expression.  These var-
	 iables are given to FACSUM in a list. */
facsum(exp2,c,d,[u,v]);
/*  The list can occur in any position in the argument list: */
facsum(exp2,[u,v],c,d);
/*  Or it can be split up: */
facsum(exp2,c,[u],d,[v]);
/*  List arguments can be nested arbitrarily deeply,
	 depending upon the requirements of the situation: */
exp3:c*d*q*x*y*z+2*c*d*m*p*x*y*z+c*d*m*n*x*y*z
	+c*d*q*w*y*z+2*c*d*m*p*w*y*z+c*d*m*n*w*y*z+c*d*q*u*v^2*z
	+2*c*d*m*p*u*v^2*z+c*d*m*n*u*v^2*z-c*d*q*u*z-2*c*d*m*p*u*z
	-c*d*m*n*u*z+c*d*u*v^2*y+c*d*u*y+d*e*f^2+c*e*f^2
	-d*e+c*e+b*c*d;
facsum(exp3,c,d,[u,v,[z,[m]]]);
/*  The arguments of FACSUM need not be atomic. */
exp4:subst(log(m+n),c,exp2);
facsum(exp4,log(m+n),[e,f]);
/*  FACSUM also recognizes in its argument the special
	 form OPERATOR.  This form can be used to indicate to
	 FACSUM that all structures within its argument that 
	 have certain indicated leading operators are to be
	 specially treated.
	 For example, consider the expression EXP5: */
exp5:x+(a.b)*(c+d)+(b.a)*(d+c)+log(a*b)*2*(c+d)+log(a/b)*2*(d+c);
/* First FACTOR EXP5, to obtain an expression
	 to work on: */
exp5_factored:factor(exp5);
/* To demonstrate the use of OPERATOR, we recover the original
	 form of EXP5: */
facsum(exp5_factored,operator(log,"."));
is(%=exp5);
/* This form also works: */
facsum(exp5_factored,operator(log),operator("."));
/* Another function that is related to FACSUM is
	 FACTORFACSUM.  FACTORFACSUM is similar to FACSUM, except
	 that it first factors the expression, then calls FACSUM 
	 on each of the factors.  It can take all the arguments
	 that FACSUM can, including the nested lists and the form
	 OPERATOR.
		To get an expression to work on, we use EXP6: */
exp6:exp5*2*(e+(h+f)*(e.f));
/* And EXPAND it: */
exp6_expanded:expand(exp6);
/* To illustrate the use of FACTORFACSUM, we recover the 
	 original form of EXP6: */
factorfacsum(exp6_expanded,operator(".",log));
is(%=exp6);
/* There is a switch NEXTLAYERFACTOR[FALSE]
	 which can be used in two ways to control the behavior
	 of FACSUM.  By setting NEXTLAYERFACTOR:TRUE one can force
	 FACSUM to FACTOR the coefficients of the variables specified
	 at any level before it calls itself recursively on the
	 factors of those coefficients:  */
exp7:f*h+f*g-2*c*d*f^2+2*c*d*e^2;
facsum(exp7,c,[e]),nextlayerfactor:true;
facsum(exp7,c,[e]);
/* The second method for controlling the behavior of
	 FACSUM is to include the atom NEXTLAYERFACTOR in the
	 in the argument list of FACSUM: */
facsum(exp7,c,'nextlayerfactor,[e]);
/* Notice that this method produced the same result as
	 simply setting NEXTLAYERFACTOR:TRUE.  The difference
	 between the two methods is that the second method allows
	 one to achieve the effect of NEXTLAYERFACTOR:TRUE for
	 only a few specified levels of the expression, instead
	 of for all levels.  For example, consider EXP8: */
exp8:-2*c*d*f^2*h^2*l^2+f*h^2*l^2-4*c*d*f^2*h^2*k*l
	 +2*f*h^2*k*l-4*c*d*f^2*g*h*l+2*f*g*h*l-2*c*d*f^2*h^2*k^2
	 +f*h^2*k^2-4*c*d*f^2*g*h*k+2*f*g*h*k-2*c*d*f^2*g^2
	 +f*g^2+2*c*d*e^2;
facsum(exp8,c,'nextlayerfactor,[f]);
facsum(exp8,c,[f,'nextlayerfactor]);

/* Another switch FACSUM_COMBINE[TRUE] controls the form
	 returned by FACSUM when its argument has a denominator.
	 If FACSUM_COMBINE is TRUE then the form returned will
	 be a ratio of polynomials.  If FALSE, then the denominator
	 factors will be distributed over the terms of the numerator.
	 In either case, both the numerator and denominator will
	 be processed according to the arguments of FACSUM. */
exp1/(c*(e+f)+d*e);
facsum(exp1/(c*(e+f)+d*e),e,f,[c,d]);
facsum(exp1/(c*(e+f)+d*e),e,f,[c,d]),facsum_combine:false;
/* It is also possible to control the form of the coefficients
         of products of quantities specified in the argument list of
	 FACSUM.  (Normally, these coefficients are factored, unless 
         they are numbers, and the function that is used for this
         purpose is called NONUMFACTOR.)  But it is possible to use
         other functions in place of NONUMFACTOR by changing the
         AUTOMATIC property of FACSUM.  */
get('facsum,'automatic);
/*  Let us illustrate this procedure by changing from
         NONUMFACTOR to SQFR. */
put('facsum,'sqfr,'automatic);
/*  Now compare the behavior of FACSUM to its former 
         behavior.  Here is what it did with NONUMFACTOR AUTOMATIC: */
playback([5,6])$
/* And now with SQFR AUTOMATIC: */
facsum(exp2,c,d,[u,v]);
/*  Since this particular choice for AUTOMATIC is so
         useful, it is available as a separate function, SQFRFACSUM.
         More complicated choices are possible, depending on the
         requirements of the situation.  One possibility is illustrated
         below.
           The AUTOMATIC function can be declared FORMAL.  If this
         is done, then the function will not be applied, but will
         be introduced into the expression in the places where it
         would have been applied.  This capability is useful for
         constructing expressions that one intends to use later
         in function definitions.  We illustrate with SQFR:
         (Maxima: NOUN does the same thing, doesn't it. You can define
         an alias for it if you want to, see genut.mac). */
/* DECLARE(SQFR,FORMAL)$ */
declare(sqfr,noun)$
facsum(exp2,c,d,[u,v]);
/*  Now restore the default AUTOMATIC property: */
put('facsum,'nonumfactor,'automatic);
/* Sometimes one must combine large expressions that
	 have already been processed with FACSUM, perhaps in different
	 Macsymas for reasons of space.  To combine these expressions
	 it is not necessary to FACSUM their sum.  An alternative is
	 to use COLLECTTERMS.  To illustrate the use of COLLECTTERMS,
	 we use EXP9 and EXP10. */
exp9:a*u+b;
exp10:c*u+d;
/* COLLECTERMS(exp, var1, var2..) collects terms of exp that contain
	 like powers of the vari.  */
collectterms(exp9+exp10,u);
e*u^2+f/u +b+d;
exp11:expand(%+exp10*u);
collectterms(%,u);
/* Another more complex example: */
u^2*v+exp11+subst(v,u,exp11);
collectterms(%,u,v);
/* COLLECTTERMS also accepts arguments in the same form
	 as FACSUM: */
exp2;
collectterms(exp2,c,d,[u,v]);
exp5_factored;
collectterms(exp5_factored,operator(log,"."));