d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Minor cleanups to allow handing vast datasets.

Browse Source 
Submitted by: dds
This commit is contained in:
phk 2020-02-03 20:46:31 +00:00
parent 2f7b86ee2c
commit 5b020e7a0f
1 changed files with 29 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

53
usr.bin/ministat/ministat.c
View File

@ -142,9 +142,9 @@ static char symbol[MAX_DS] = { ' ', 'x', '+', '*', '%', '#', '@', 'O' };
struct dataset { struct dataset {
char *name; char *name;
double *points; double *points;
unsigned lpoints; size_t lpoints;
double sy, syy; double sy, syy;
unsigned n; size_t n;
}; };
static struct dataset * static struct dataset *
@ -202,7 +202,7 @@ Avg(const struct dataset *ds)
static double static double
Median(const struct dataset *ds) Median(const struct dataset *ds)
{ {
const unsigned m = ds->n / 2; const size_t m = ds->n / 2;
if ((ds->n % 2) == 0) if ((ds->n % 2) == 0)
return ((ds->points[m] + (ds->points[m - 1])) / 2); return ((ds->points[m] + (ds->points[m - 1])) / 2);
@ -212,13 +212,13 @@ Median(const struct dataset *ds)
static double static double
Var(struct dataset *ds) Var(struct dataset *ds)
{ {
unsigned n; size_t z;
const double a = Avg(ds); const double a = Avg(ds);
if (isnan(ds->syy)) { if (isnan(ds->syy)) {
ds->syy = 0.0; ds->syy = 0.0;
for (n = 0; n < ds->n; n++) for (z = 0; z < ds->n; z++)
ds->syy += (ds->points[n] - a) * (ds->points[n] - a); ds->syy += (ds->points[z] - a) * (ds->points[z] - a);
} }
return (ds->syy / (ds->n - 1.0)); return (ds->syy / (ds->n - 1.0));
@ -242,7 +242,7 @@ static void
Vitals(struct dataset *ds, int flag) Vitals(struct dataset *ds, int flag)
{ {
printf("%c %3d %13.8g %13.8g %13.8g %13.8g %13.8g", symbol[flag], printf("%c %3zu %13.8g %13.8g %13.8g %13.8g %13.8g", symbol[flag],
ds->n, Min(ds), Max(ds), Median(ds), Avg(ds), Stddev(ds)); ds->n, Min(ds), Max(ds), Median(ds), Avg(ds), Stddev(ds));
printf("\n"); printf("\n");
} }
@ -252,13 +252,13 @@ Relative(struct dataset *ds, struct dataset *rs, int confidx)
{ {
double spool, s, d, e, t; double spool, s, d, e, t;
double re; double re;
int i; size_t z;
i = ds->n + rs->n - 2; z = ds->n + rs->n - 2;
if (i > NSTUDENT) if (z > NSTUDENT)
t = student[0][confidx]; t = student[0][confidx];
else else
t = student[i][confidx]; t = student[z][confidx];
spool = (ds->n - 1) * Var(ds) + (rs->n - 1) * Var(rs); spool = (ds->n - 1) * Var(ds) + (rs->n - 1) * Var(rs);
spool /= ds->n + rs->n - 2; spool /= ds->n + rs->n - 2;
spool = sqrt(spool); spool = sqrt(spool);
@ -272,7 +272,6 @@ Relative(struct dataset *ds, struct dataset *rs, int confidx)
re = t * sqrt(re); re = t * sqrt(re);
if (fabs(d) > e) { if (fabs(d) > e) {
printf("Difference at %.1f%% confidence\n", studentpct[confidx]); printf("Difference at %.1f%% confidence\n", studentpct[confidx]);
printf(" %g +/- %g\n", d, e); printf(" %g +/- %g\n", d, e);
printf(" %g%% +/- %g%%\n", d * 100 / Avg(rs), re * 100 / Avg(rs)); printf(" %g%% +/- %g%%\n", d * 100 / Avg(rs), re * 100 / Avg(rs));
@ -290,7 +289,7 @@ struct plot {
int width; int width;
double x0, dx; double x0, dx;
int height; size_t height;
char *data; char *data;
char **bar; char **bar;
int separate_bars; int separate_bars;
@ -343,9 +342,11 @@ static void
PlotSet(struct dataset *ds, int val) PlotSet(struct dataset *ds, int val)
{ {
struct plot *pl; struct plot *pl;
int i, j, m, x; int i, x;
unsigned n; size_t m, j, z;
size_t n;
int bar; int bar;
double av, sd;
pl = &plot; pl = &plot;
if (pl->span == 0) if (pl->span == 0)
@ -370,6 +371,7 @@ PlotSet(struct dataset *ds, int val)
m = 1; m = 1;
i = -1; i = -1;
j = 0; j = 0;
/* Set m to max(j) + 1, to allocate required memory */
for (n = 0; n < ds->n; n++) { for (n = 0; n < ds->n; n++) {
x = (ds->points[n] - pl->x0) / pl->dx; x = (ds->points[n] - pl->x0) / pl->dx;
if (x == i) { if (x == i) {
@ -400,18 +402,20 @@ PlotSet(struct dataset *ds, int val)
} }
pl->data[j * pl->width + x] |= val; pl->data[j * pl->width + x] |= val;
} }
if (!isnan(Stddev(ds))) { av = Avg(ds);
x = ((Avg(ds) - Stddev(ds)) - pl->x0) / pl->dx; sd = Stddev(ds);
m = ((Avg(ds) + Stddev(ds)) - pl->x0) / pl->dx; if (!isnan(sd)) {
x = ((av - sd) - pl->x0) / pl->dx;
m = ((av + sd) - pl->x0) / pl->dx;
pl->bar[bar][m] = '|'; pl->bar[bar][m] = '|';
pl->bar[bar][x] = '|'; pl->bar[bar][x] = '|';
for (i = x + 1; i < m; i++) for (z = x + 1; z < m; z++)
if (pl->bar[bar][i] == 0) if (pl->bar[bar][z] == 0)
pl->bar[bar][i] = '_'; pl->bar[bar][z] = '_';
} }
x = (Median(ds) - pl->x0) / pl->dx; x = (Median(ds) - pl->x0) / pl->dx;
pl->bar[bar][x] = 'M'; pl->bar[bar][x] = 'M';
x = (Avg(ds) - pl->x0) / pl->dx; x = (av - pl->x0) / pl->dx;
pl->bar[bar][x] = 'A'; pl->bar[bar][x] = 'A';
} }
@ -420,6 +424,7 @@ DumpPlot(void)
{ {
struct plot *pl; struct plot *pl;
int i, j, k; int i, j, k;
size_t z;
pl = &plot; pl = &plot;
if (pl->span == 0) { if (pl->span == 0) {
@ -432,10 +437,10 @@ DumpPlot(void)
putchar('-'); putchar('-');
putchar('+'); putchar('+');
putchar('\n'); putchar('\n');
for (i = 1; i < pl->height; i++) { for (z = 1; z < pl->height; z++) {
putchar('|'); putchar('|');
for (j = 0; j < pl->width; j++) { for (j = 0; j < pl->width; j++) {
k = pl->data[(pl->height - i) * pl->width + j]; k = pl->data[(pl->height - z) * pl->width + j];
if (k >= 0 && k < MAX_DS) if (k >= 0 && k < MAX_DS)
putchar(symbol[k]); putchar(symbol[k]);
else else