epilectrik
/
pyNgSpice
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Browse Source

Make doxygen documentation complete as an example.

pre-master-46
Holger Vogt 3 years ago
parent
70f38ec25f
commit
638a09b718
1 changed files with 79 additions and 27 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 106
      src/maths/cmaths/cmath1.c

106
src/maths/cmaths/cmath1.c
View File

@ -4,7 +4,7 @@ Author: 1985 Wayne A. Christopher, U. C. Berkeley CAD Group
**********/
/** \file cmath1.c
\brief functions for the control language parser: mag, ph, cph, unwrap, j, real, conj, pos, db, log10, log, exp, sqrt, sin, sinh, cos, coh, tan, tanh, atan, sortorder
\brief Functions for the control language parser: mag, ph, cph, unwrap, j, real, conj, pos, db, log10, log, exp, sqrt, sin, sinh, cos, coh, tan, tanh, atan, sortorder
Routines to do complex mathematical functions. These routines require
the -lm libraries. We sacrifice a lot of space to be able
@ -14,7 +14,9 @@ Author: 1985 Wayne A. Christopher, U. C. Berkeley CAD Group
Complex functions are called as follows:
cx_something(data, type, length, &newlength, &newtype),
and return a char * that is cast to complex or double.
and return a void* that has to be cast to complex or double.
Integers newlength and newtype contain the newly resulting length
of the void* vector and its new type (REAL or COMPLEX).
*/
@ -32,12 +34,16 @@ Author: 1985 Wayne A. Christopher, U. C. Berkeley CAD Group
#define win_x_fprintf fprintf
#endif
/* This flag determines whether degrees or radians are used. The radtodeg
* and degtorad macros are no-ops if this is FALSE.
/**This global flag determines whether degrees or radians are used. The radtodeg
* and degtorad macros are no-ops if this is FALSE. It will be set to TRUE in options.c
* if variable (option) 'unit' is equal to 'degree'.
*/
bool cx_degrees = FALSE;
/** Magnitude of real and complex vectors:
fabs() for real,
hypot() for complex.
*/
void *
cx_mag(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -57,6 +63,10 @@ cx_mag(void *data, short int type, int length, int *newlength, short int *newtyp
return ((void *) d);
}
/** Phase of vectors:
0 for real,
atan2() for complex.
*/
void *
cx_ph(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -74,7 +84,10 @@ cx_ph(void *data, short int type, int length, int *newlength, short int *newtype
return ((void *) d);
}
/* SJdV Modified from above to find closest from +2pi,0, -2pi */
/** Continuous phase of vectors:
0 for real,
atan2() for complex.
Modified from cx_ph to find closest from +2pi,0, -2pi. */
void *
cx_cph(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -97,7 +110,9 @@ cx_cph(void *data, short int type, int length, int *newlength, short int *newtyp
return ((void *) d);
}
/* Modified from above but with real phase vector in degrees as input */
/** Modified from cx_cph(), but with real phase vector in degrees as input.
Currently not in use.
*/
void *
cx_unwrap(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -120,7 +135,7 @@ cx_unwrap(void *data, short int type, int length, int *newlength, short int *new
return ((void *) d);
}
/* If this is pure imaginary we might get real, but never mind... */
/** Multiply by i (imaginary unit). */
void *cx_j(void *data, short int type, int length, int *newlength,
short int *newtype)
{
@ -147,6 +162,7 @@ void *cx_j(void *data, short int type, int length, int *newlength,
return (void *) c;
}
/** Return the real part of the vector. */
void *cx_real(void *data, short int type, int length, int *newlength,
short int *newtype)
{
@ -171,6 +187,7 @@ void *cx_real(void *data, short int type, int length, int *newlength,
return (void *) d;
}
/** Return the imaginary part of the vector. */
void *cx_imag(void *data, short int type, int length, int *newlength,
short int *newtype)
{
@ -197,7 +214,7 @@ void *cx_imag(void *data, short int type, int length, int *newlength,
/* Create complex conjugate of data */
/** Create complex conjugate of data. */
void *cx_conj(void *data, short int type, int length,
int *p_newlength, short int *p_newtype)
{
@ -224,8 +241,9 @@ void *cx_conj(void *data, short int type, int length,
/* This is obsolete... */
/* Return a vector with 1s for positive and 0 for negative element values of the input vector.
Currently not in use.
*/
void *
cx_pos(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -245,6 +263,10 @@ cx_pos(void *data, short int type, int length, int *newlength, short int *newtyp
return ((void *) d);
}
/** Calculatue values in db as 20.0 * log10.
Prior to this use macro rcheck() to check for input values being positive.
Return NULL if not.
*/
void *cx_db(void *data, short int type, int length,
int *newlength, short int *newtype)
{
@ -291,7 +313,11 @@ EXITPOINT:
} /* end of function cx_db */
/** Return the common logarithm.
Prior to this use macro rcheck() to check for input values being positive or 0.
Return -log10(HUGE) when magnitude is 0.
Return NULL if negative.
*/
void *cx_log10(void *data, short int type, int length,
int *newlength, short int *newtype)
{
@ -353,7 +379,11 @@ EXITPOINT:
} /* end of function cx_log10 */
/** Return the natural logarithm.
Prior to this use macro rcheck() to check for input values being positive or 0.
Return -log(HUGE) when magnitude is 0.
Return NULL if negative.
*/
void *cx_log(void *data, short int type, int length,
int *newlength, short int *newtype)
{
@ -412,7 +442,10 @@ EXITPOINT:
} /* end of function cx_log */
/** Return the exponential of a vector:
exp() for real,
exp(realpart)*cos(imagpart), exp(realpart)*sin(imagpart) for imaginary.
*/
void *
cx_exp(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -445,6 +478,10 @@ cx_exp(void *data, short int type, int length, int *newlength, short int *newtyp
}
}
/** Square root of a complex vector:
Determine if the result vector is real or complex (due to input being negative or already complex).
Distinction of cases: Input complex, then real part pos., neg. or 0. Input real negative or real positive.
*/
void *
cx_sqrt(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -529,6 +566,9 @@ cx_sqrt(void *data, short int type, int length, int *newlength, short int *newty
}
}
/** sin of a complex vector:
sin(realpart)*cosh(imagpart), cos(realpart)*sinh(imagpart)
*/
void *
cx_sin(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -560,6 +600,8 @@ cx_sin(void *data, short int type, int length, int *newlength, short int *newtyp
}
}
/** sinh of a complex vector:
sinh(x+iy) = sinh(x)*cos(y) + i * cosh(x)*sin(y) */
void *
cx_sinh(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -593,7 +635,9 @@ cx_sinh(void *data, short int type, int length, int *newlength, short int *newty
}
}
/** cos of a complex vector:
cos(realpart)*cosh(imagpart), -sin(realpart)*sinh(imagpart)
*/
void *
cx_cos(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -625,7 +669,9 @@ cx_cos(void *data, short int type, int length, int *newlength, short int *newtyp
}
}
/**cosh of a complex vector:
cosh(x+iy) = cosh(x)*cos(y) + i * sinh(x)*sin(y)
*/
void *
cx_cosh(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -661,7 +707,9 @@ cx_cosh(void *data, short int type, int length, int *newlength, short int *newty
}
/** tan for real valued vectors. Used by cx_tanh.
Prior to this use macro rcheck() to check for input values not being 0.
Return NULL if 0.*/
static double *d_tan(double *dd, int length)
{
int xrc = 0;
@ -683,7 +731,7 @@ EXITPOINT:
} /* end of function d_tan */
/** tanh for real valued vectors. Used by cx_tanh. */
static double *
d_tanh(double *dd, int length)
{
@ -697,7 +745,9 @@ d_tanh(double *dd, int length)
return d;
}
/* See https://proofwiki.org/wiki/Tangent_of_Complex_Number (formulation 4) among
/** tan of a complex vector.
* Used by cx_tan
* See https://proofwiki.org/wiki/Tangent_of_Complex_Number (formulation 4) among
* others for the tangent formula:
* sin z = sin(x + iy) = sin x cos(iy) + cos x sin(iy) = sin x cosh y + i cos x sinh y
@ -741,10 +791,10 @@ static ngcomplex_t *c_tan(ngcomplex_t *cc, int length)
/* complex tanh function, uses tanh(z) = -i * tan(i * z).
* Unfortunately, it did so very inefficiently (a memory allocations per
* value calculated!) and leaked memory badly (all of the aforementioned
* allocations.) These issues were fixed 2020-03-04 */
/**complex tanh function:
uses tanh(z) = -i * tan(i * z).
Used by cx_tanh.
*/
static ngcomplex_t *c_tanh(ngcomplex_t *cc, int length)
{
ngcomplex_t * const tmp = alloc_c(length); /* i * z */
@ -784,7 +834,7 @@ static ngcomplex_t *c_tanh(ngcomplex_t *cc, int length)
} /* end of function c_tanh */
/** tan of a complex vector */
void *
cx_tan(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -798,6 +848,7 @@ cx_tan(void *data, short int type, int length, int *newlength, short int *newtyp
}
}
/** tanh of a complex vector */
void *
cx_tanh(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -811,6 +862,7 @@ cx_tanh(void *data, short int type, int length, int *newlength, short int *newty
}
}
/** atan of a complex vector: return atan of the real part. */
void *
cx_atan(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -835,7 +887,7 @@ cx_atan(void *data, short int type, int length, int *newlength, short int *newty
return ((void *) d);
}
/* Struct to store and order the values of the amplitudes preserving the index in the original array */
/** Struct to store and order the values of the amplitudes preserving the index in the original array */
typedef struct {
double amplitude;
int index;
@ -843,11 +895,10 @@ typedef struct {
static int compare_structs (const void *a, const void *b);
/*
/**
* Returns the positions of the elements in a real vector
* after they have been sorted into increasing order using a stable method (qsort).
*/
void *
cx_sortorder(void *data, short int type, int length, int *newlength, short int *newtype)
{
@ -879,6 +930,7 @@ cx_sortorder(void *data, short int type, int length, int *newlength, short int *
return ((void *) d);
}
/** Compares ampplitudes of vector elements. Input to qsort. */
static int
compare_structs(const void *a, const void *b)
{

Write Preview
Loading…
Cancel
Save