Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Exposed td_reset() api #9

Merged
merged 1 commit into from
Feb 28, 2021
Merged

Exposed td_reset() api #9

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 21 additions & 16 deletions src/tdigest.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,23 +36,15 @@ static size_t td_required_buf_size(double compression) {

int td_init(double compression, td_histogram_t **result) {
size_t buf_size = td_required_buf_size(compression);
td_histogram_t *h = (td_histogram_t *)((char *)(__td_malloc(buf_size)));
if (!h) {
return NULL;
td_histogram_t *histogram = (td_histogram_t *)((char *)(__td_malloc(buf_size)));
if (!histogram) {
return 1;
}
bbzero((void *)(h), buf_size);
*h = (td_histogram_t){
.compression = compression,
.cap = (buf_size - sizeof(td_histogram_t)) / sizeof(node_t),
.min = __DBL_MAX__,
.max = __DBL_MIN__,
.total_compressions = 0,
.merged_nodes = 0,
.merged_weight = 0,
.unmerged_nodes = 0,
.unmerged_weight = 0,
};
*result = h;
bbzero((void *)(histogram), buf_size);
histogram->cap = (buf_size - sizeof(td_histogram_t)) / sizeof(node_t);
histogram->compression = compression;
td_reset(histogram);
*result = histogram;
return 0;
}

Expand All @@ -75,6 +67,19 @@ void td_merge(td_histogram_t *into, td_histogram_t *from) {

int td_centroid_count(td_histogram_t *h) { return next_node(h); }

void td_reset(td_histogram_t *h) {
if (h == NULL) {
return;
}
h->min = __DBL_MAX__;
h->max = __DBL_MIN__;
h->merged_nodes = 0;
h->merged_weight = 0;
h->unmerged_nodes = 0;
h->unmerged_weight = 0;
h->total_compressions = 0;
}

double td_size(td_histogram_t *h) { return h->merged_weight + h->unmerged_weight; }

double td_cdf(td_histogram_t *h, double val) {
Expand Down
101 changes: 98 additions & 3 deletions tests/td_test.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,42 @@

#define STREAM_SIZE 1000000

#define MAX(x, y) (((x) > (y)) ? (x) : (y))
#define MIN(x, y) (((x) < (y)) ? (x) : (y))

static double randfrom(double M, double N) { return M + (rand() / (RAND_MAX / (N - M))); }

/**
* Reference implementations for cdf if we have all data.
*/
static double dist_cdf(double x, double *data, int data_length) {
double n1 = 0;
double n2 = 0;
for (size_t i = 0; i < data_length; i++) {
const double v = data[i];
n1 += (v < x) ? 1 : 0;
n2 += (v == x) ? 1 : 0;
}
return (n1 + n2 / 2.0) / data_length;
}

/**
* Reference implementations for quantile if we have all data.
*/
static double dist_quantile(double q, double *data, int data_length) {
if (data_length == 0) {
return NAN;
}
double index = q * data_length;
if (index < 0) {
index = 0;
}
if (index > data_length - 1) {
index = data_length - 1;
}
return data[(int)floor(index)];
}

int tests_run = 0;

td_histogram_t *histogram = NULL;
Expand Down Expand Up @@ -55,8 +89,27 @@ MU_TEST(test_basic) {
mu_assert(td_compression(t) < t->cap, "False: buffer size < compression");
mu_assert_double_eq(0.0, td_quantile(t, .0));
mu_assert_double_eq(0.0, td_quantile(t, .1));
// mu_assert_double_eq(10.0, td_quantile(t, .5));
mu_assert_double_eq(10.0, td_quantile(t, .99));
td_reset(t);
td_reset(NULL);
td_free(t);
}

MU_TEST(test_quantile_interpolations) {
td_histogram_t *t = td_new(10);
mu_assert(t != NULL, "created_histogram");
mu_assert_double_eq(0, t->unmerged_weight);
mu_assert_double_eq(0, t->merged_weight);
td_add(t, 5.0, 2);
mu_assert_double_eq(1, t->unmerged_weight);
// with one data point, all quantiles lead to Rome
mu_assert_double_eq(0.0, td_quantile(t, .0));
mu_assert_double_eq(0.0, td_quantile(t, 0.5));
td_compress(t);
mu_assert_double_eq(0, t->unmerged_weight);
mu_assert_double_eq(2.0, t->merged_weight);
td_add(t, 100.0, 1);
// we know that there are at least two centroids now
td_free(t);
}

Expand Down Expand Up @@ -187,19 +240,61 @@ MU_TEST(test_two_interp) {
td_add(t, 1, 1);
td_add(t, 10, 1);
mu_assert(isfinite(td_quantile(t, .9)), "test_two_interp: value at .9");
td_reset(t);
// if the left centroid has more than one sample, we still know
// that one sample occurred at min so we can do some interpolation
td_add(t, 1, 10);
td_add(t, 10, 1);
mu_assert_double_eq(1.0, td_quantile(t, .1));
td_reset(t);
// if the right-most centroid has more than one sample, we still know
// that one sample occurred at max so we can do some interpolation
td_add(t, 1, 1);
td_add(t, 10, 10);
mu_assert_double_eq(10.0, td_quantile(t, .9));
td_reset(t);
// in between extremes we interpolate between centroids
td_add(t, 1, 1);
td_add(t, 5, 1);
td_add(t, 10, 1);
// centroids i and i+1 bracket our current point
// check for unit weight
// within the singleton's sphere
// left
mu_assert_double_eq(5.0, td_quantile(t, .5));
td_reset(t);
// in between extremes we interpolate between centroids
td_add(t, 1, 1); // q0
td_add(t, 4, 1); // q20
td_add(t, 8, 1); // q40
td_add(t, 12, 1); // q60
td_add(t, 16, 1); // q80
td_add(t, 20, 1); // q100
// centroids i and i+1 bracket our current point
// check for unit weight
// within the singleton's sphere
// TODO: check for right
// mu_assert_double_eq(4.0, td_quantile(t, .20) );
// mu_assert_double_eq(8.0, td_quantile(t, .40) );
// mu_assert_double_eq(12.0, td_quantile(t, .60) );
// mu_assert_double_eq(7.0, td_quantile(t, .70) );
// mu_assert_double_eq(8.0, td_quantile(t, .75) );
td_free(t);
}

MU_TEST(test_cdf) {
td_histogram_t *t = td_new(100);
mu_assert(isnan(td_cdf(t, 1.1)), "no data to examine");
// interpolate if somehow we have weight > 0 and max != min
td_add(t, 1, 1);
// bellow lower bound
mu_assert_double_eq(0, td_cdf(t, 0));
// exactly one centroid, should have max==min
// min and max are too close together to do any viable interpolation
mu_assert_double_eq(0.5, td_cdf(t, 1));
// above upper bound
mu_assert_double_eq(1.0, td_cdf(t, 2));
td_add(t, 10, 1);
mu_assert_double_eq(0, td_cdf(t, .99));
mu_assert_double_eq(1, td_cdf(t, 10.01));
mu_assert_double_eq(.25, td_cdf(t, 1));
mu_assert_double_eq(.5, td_cdf(t, 5.5));
// // TODO: fix this
Expand Down