This commit is contained in:
Andras Schmelczer 2026-02-02 20:10:32 +00:00
parent 9179acd4cd
commit 2c613dc0d1
14 changed files with 376 additions and 188 deletions

View file

@ -8,7 +8,7 @@ use axum::response::IntoResponse;
use serde::Deserialize;
use tracing::{info, warn};
use crate::consts::{ENUM_NULL, H3_REQUEST_MAX, H3_REQUEST_MIN, HISTOGRAM_BINS};
use crate::consts::{ENUM_NULL, H3_PRECOMPUTE_MAX, H3_REQUEST_MAX, H3_REQUEST_MIN, HISTOGRAM_BINS};
use crate::filter::{parse_filters, row_passes_filters};
use crate::state::AppState;
@ -19,6 +19,10 @@ pub struct HexagonStatsParams {
pub h3: String,
pub resolution: u8,
pub filters: Option<String>,
/// Comma-separated feature names to include in stats response.
/// When present (even if empty), only listed features are computed.
/// When absent, all features are returned (backward compatible).
pub fields: Option<String>,
}
pub async fn get_hexagon_stats(
@ -45,8 +49,6 @@ pub async fn get_hexagon_stats(
),
));
}
let resolution_idx = resolution as usize;
let h3_str = params.h3.clone();
let filters_str = params.filters.clone();
let (parsed_filters, parsed_enum_filters) = parse_filters(
@ -56,42 +58,58 @@ pub async fn get_hexagon_stats(
);
let num_filters = parsed_filters.len() + parsed_enum_filters.len();
// Parse optional `fields` param into sets of feature names.
// None = include all, Some = only include listed features.
let field_set: Option<std::collections::HashSet<String>> = params.fields.as_ref().map(|fields_str| {
fields_str
.split(',')
.map(|field| field.trim().to_string())
.filter(|field| !field.is_empty())
.collect()
});
let result = tokio::task::spawn_blocking(move || {
let start_time = std::time::Instant::now();
let precomputed: Option<&[u64]> = state
.h3_cells
.get(resolution_idx)
.filter(|cells| !cells.is_empty())
.map(|cells| cells.as_slice());
let precomputed = &state.h3_cells;
let h3_res = h3o::Resolution::try_from(resolution)
.map_err(|err| format!("Invalid H3 resolution {}: {}", resolution, err))?;
let need_parent = resolution < H3_PRECOMPUTE_MAX;
let num_features = state.data.num_features;
let num_enums = state.data.num_enums;
let feature_data = &state.data.feature_data;
let enum_data = &state.data.enum_data;
let enum_features = &state.data.enum_features;
let (min_lat, min_lon, max_lat, max_lon) = h3_cell_bounds(cell, 0.001);
// Resolve cell at requested resolution from precomputed max-resolution cell
let cell_for_row = |row: usize| -> u64 {
let max_cell = precomputed[row];
if !need_parent || max_cell == 0 {
return max_cell;
}
h3o::CellIndex::try_from(max_cell)
.ok()
.and_then(|ci| ci.parent(h3_res))
.map(u64::from)
.unwrap_or(0)
};
// Collect matching rows
let mut matching_rows: Vec<usize> = Vec::new();
state
.grid
.for_each_in_bounds(min_lat, min_lon, max_lat, max_lon, |row_idx| {
let row = row_idx as usize;
let row_cell = if let Some(h3_data) = precomputed {
h3_data[row]
} else {
h3o::LatLng::new(state.data.lat[row] as f64, state.data.lon[row] as f64)
.map(|coord| u64::from(coord.to_cell(h3_res)))
.unwrap_or(0)
};
if row_cell == cell_u64
if cell_for_row(row) == cell_u64
&& row_passes_filters(
row,
&parsed_filters,
&parsed_enum_filters,
feature_data,
num_features,
enum_features,
enum_data,
num_enums,
)
{
matching_rows.push(row);
@ -109,6 +127,12 @@ pub async fn get_hexagon_stats(
output.push_str(",\"numeric_features\":[");
let mut first_numeric = true;
for (feature_index, feature_name) in state.data.feature_names.iter().enumerate() {
// Skip features not in the requested set (when fields param is present)
if let Some(ref set) = field_set {
if !set.contains(feature_name.as_str()) {
continue;
}
}
let global_stats = &state.data.feature_stats[feature_index];
let histogram_min = global_stats.histogram.min;
let histogram_max = global_stats.histogram.max;
@ -178,15 +202,20 @@ pub async fn get_hexagon_stats(
output.push_str("],\"enum_features\":[");
let mut first_enum = true;
for enum_feature in enum_features {
// Skip enum features not in the requested set
if let Some(ref set) = field_set {
if !set.contains(enum_feature.name.as_str()) {
continue;
}
}
let enum_index = match state.enum_name_to_idx.get(&enum_feature.name) {
Some(&index) => index,
None => continue,
};
let enum_data = &state.data.enum_features[enum_index];
let mut value_counts = vec![0u64; enum_data.values.len()];
let mut value_counts = vec![0u64; enum_feature.values.len()];
for &row in &matching_rows {
let value = enum_data.data[row];
let value = enum_data[row * num_enums + enum_index];
if value != ENUM_NULL && (value as usize) < value_counts.len() {
value_counts[value as usize] += 1;
}
@ -215,7 +244,7 @@ pub async fn get_hexagon_stats(
output.push(',');
}
first_value = false;
write_json_string(&mut output, &enum_data.values[value_index]);
write_json_string(&mut output, &enum_feature.values[value_index]);
write!(output, ":{}", count).unwrap();
}
output.push_str("}}");