mc-utils

this is general utilities for usage in Map Colonies project.

included components

• http client
• footprint to tile ranges

usage

footprint to tile ranges

footprintToTileRanges(footprint, { minZoom, maxZoom }) lazily computes the tile ranges intersecting a footprint (Polygon / MultiPolygon / Feature, holes supported) for every zoom level in the span, on the WGS84 geodetic 2:1 grid.

it uses hierarchical descent (see docs/adr/0001-hierarchical-precompute-in-mc-utils.md): fully-interior subtrees are emitted as compressed ranges with no per-tile work, so memory stays flat regardless of how many tiles the footprint covers. a tile touching the footprint with zero overlap area counts as intersecting. ranges of the same zoom are disjoint; their order is deterministic but unspecified.

typical use is tile-cache invalidation: expand each range to your cache's key format and batch-delete.

import { footprintToTileRanges } from '@map-colonies/mc-utils';

for (const range of footprintToTileRanges(footprint, { minZoom: 0, maxZoom: 18 })) {
  for (let x = range.minX; x <= range.maxX; x++) {
    for (let y = range.minY; y <= range.maxY; y++) {
      actionOnTile(range.zoom, x, y); // e.g. batch UNLINK keys in redis
    }
  }
}

http client

this is abstract base class for sending http request with logging and request retries.

this class constructor requires the following parameters:

• ILogger instance.
• base url to use for all requests.
• optional name for target service to be used in logs.
• optional retry configuration.

the class have the following protected attributes -axiosOptions the options used by axios when sending requests

the class have the protected methods for sending http requests:

• protected async get<T>(url: string, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async post<T>(url: string, body?: unknown, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async put<T>(url: string, body?: unknown, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async delete<T>(url: string, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async head<T>(url: string, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async options<T>(url: string, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>
• protected async patch<T>(url: string, body?: unknown, queryParams?: Record<string, unknown>, retryConfig?: IAxiosRetryConfig, auth?: AxiosBasicCredentials, headers?: unknown): Promise<T>

function parameters list:

• url: url path to send the request to (not including the base url).
• body: optional object to be used as request body (in relevant request types).
• queryParams: optional dictionary with query parameters and value.
• retryConfig: optional override to the class retry configuration.
• auth: optional basic authentication object (username, password).
• headers: optional headers to proceed to the request.

usage example:

class myServiceClient extends HttpClient {
  public constructor(logger: ILogger){
    super(logger,'https://myService.com','myService',{
      attempts: 3,
      delay: 'exponential',
      shouldResetTimeout: true
    })
  }

  public async getName(): string {
    const name = await this.get<string>('api/name',{queryParam1: 'name'});
    return name;
  }
}

models

ILogger

logger interface with the flowing log methods

• error(message:string)
• warn(message:string)
• info(message: string)
• debug(message: string)

IHttpRetryConfig

http requests retry configuration interface with the following attributes:

• attempts - the number of request to send until valid response (not 500+ status code). this value must be integer and greater then 0.
• delay - the amount of time in ms to wait between attempts (for constant delay) or 'exponential' for exponential backoff.
• shouldResetTimeout boolean value to indicate if the request timeout should be for each request (true) or global for all attempts (false)

Shapefile Processing with State Management

The ShapefileChunkReader provides robust shapefile processing with built-in state management and progress tracking capabilities.

Features

• Resumable Processing: Automatically saves processing state and can resume from the last processed chunk
• Progress Tracking: Real-time progress information including features processed, vertices counted, and time estimates
• Metrics Collection: Optional performance metrics collection for monitoring resource usage
• Chunk-based Processing: Processes large shapefiles in configurable chunks to manage memory usage

Basic Usage

import { ShapefileChunkReader, ReaderOptions, ChunkProcessor } from '@map-colonies/mc-utils';

// Define your chunk processor
const processor: ChunkProcessor = {
  process: async (chunk) => {
    console.log(`Processing chunk ${chunk.id} with ${chunk.features.length} features`);
    // Your processing logic here
  }
};

// Configure the reader
const options: ReaderOptions = {
  maxVerticesPerChunk: 10000,
  generateFeatureId: false, // Optional: for generating an UUID v4 for missing IDs
  logger: myLogger,
  stateManager: myStateManager, // Optional: for resumable processing
  metricsCollector: myMetricsCollector, // Optional: for performance monitoring
};

// Create and use the reader
const reader = new ShapefileChunkReader(options);
await reader.readAndProcess('/path/to/shapefile.shp', processor);

State Management

Implement the StateManager interface to enable resumable processing:

import { StateManager, ProcessingState } from '@map-colonies/mc-utils';

class FileStateManager implements StateManager {
  async saveState(state: ProcessingState): Promise<void> {
    // Save state to file, database, etc.
    await fs.writeFile('processing-state.json', JSON.stringify(state));
  }

  async loadState(): Promise<ProcessingState | null> {
    try {
      const data = await fs.readFile('processing-state.json', 'utf8');
      return JSON.parse(data);
    } catch {
      return null; // No previous state found
    }
  }
}

Progress Tracking

The reader automatically tracks processing progress and provides detailed information:

// Progress information is included in the saved state
const progressInfo = state.progress; // Contains:
// - processedFeatures, totalFeatures
// - processedChunks, totalChunks  
// - processedVertices, totalVertices
// - percentage, elapsedTimeMs, estimatedRemainingTimeMs
// - Processing speeds (features/sec, vertices/sec, chunks/sec)

Metrics Collection

Implement the MetricsCollector interface to monitor performance:

import { MetricsCollector, ChunkMetrics, FileMetrics } from '@map-colonies/mc-utils';

const metricsCollector: MetricsCollector = {
  onChunkMetrics: (metrics: ChunkMetrics) => {
    console.log(`Chunk ${metrics.chunkIndex}: ${metrics.featuresCount} features, ${metrics.totalTimeMs}ms`);
  },
  onFileMetrics: (metrics: FileMetrics) => {
    console.log(`File complete: ${metrics.totalFeatures} features, ${metrics.totalTimeMs}ms total`);
  }
};

Configuration Options

Option	Type	Default	Description
maxVerticesPerChunk	number	Required	Maximum vertices per chunk to control memory usage
generateFeatureId	boolean	false	Optional determines whether a unique feature identifier should be automatically generated for each feature missing an indentifier
logger	Logger	undefined	Optional logger for debugging and monitoring
stateManager	StateManager	undefined	Optional state manager for resumable processing
metricsCollector	MetricsCollector	undefined	Optional metrics collector for performance monitoring