19768106976

Committed 28 Nov 2025 03:32PM UTC coverage: 61.675% (-0.09%) from 61.76%

Build # 19768106976

Build Type

push

github

Committed by

web-flow

Commit Message

chore: patch release 3.2.3 (#3250)

* draft

Signed-off-by: Ihor Dykhta <dikhta.igor@gmail.com>

* patch

Signed-off-by: Ihor Dykhta <dikhta.igor@gmail.com>

* fix eslint during release

Signed-off-by: Ihor Dykhta <dikhta.igor@gmail.com>

---------

Signed-off-by: Ihor Dykhta <dikhta.igor@gmail.com>

Run Details

6352 of 12229 branches covered (51.94%)

Branch coverage included in aggregate %.

13043 of 19218 relevant lines covered (67.87%)

81.74 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

89.19

/src/processors/src/file-handler.ts

// SPDX-License-Identifier: MIT
// Copyright contributors to the kepler.gl project

import {parseInBatches} from '@loaders.gl/core';
import {JSONLoader, _JSONPath} from '@loaders.gl/json';
import {CSVLoader} from '@loaders.gl/csv';
import {GeoArrowLoader} from '@loaders.gl/arrow';
import {ParquetWasmLoader} from '@loaders.gl/parquet';
import {Loader} from '@loaders.gl/loader-utils';
import {
  isPlainObject,
  generateHashIdFromString,
  getApplicationConfig,
  getError,
  isArrowTable
} from '@kepler.gl/utils';
import {generateHashId} from '@kepler.gl/common-utils';
import {DATASET_FORMATS} from '@kepler.gl/constants';
import {AddDataToMapPayload, Feature, LoadedMap, ProcessorResult} from '@kepler.gl/types';
import {KeplerTable} from '@kepler.gl/table';
import {FeatureCollection} from '@turf/helpers';

import {
  processArrowBatches,
  processGeojson,
  processKeplerglJSON,
  processRowObject
} from './data-processor';

import {FileCacheItem, ValidKeplerGlMap} from './types';

const BATCH_TYPE = {
  METADATA: 'metadata',
  PARTIAL_RESULT: 'partial-result',
  FINAL_RESULT: 'final-result'
};

const CSV_LOADER_OPTIONS = {
  shape: 'object-row-table',
  dynamicTyping: false // not working for now
};

const ARROW_LOADER_OPTIONS = {
  shape: 'arrow-table',
  batchDebounceMs: 10 // time to delay between batches, for incremental loading
};

const PARQUET_LOADER_OPTIONS = {
  shape: 'arrow-table'
};

const JSON_LOADER_OPTIONS = {
  shape: 'object-row-table',
  // instruct loaders.gl on what json paths to stream
  jsonpaths: [
    '$', // JSON Row array
    '$.features', // GeoJSON
    '$.datasets' // KeplerGL JSON
  ]
};

export type ProcessFileDataContent = {
  data: unknown;
  fileName: string;
  length?: number;
  progress?: {rowCount?: number; rowCountInBatch?: number; percent?: number};
  /**  metadata e.g. for arrow data, metadata could be the schema.fields */
  metadata?: Map<string, string>;
};

export {isArrowTable};

/**
 * check if data is an ArrowData object, which is an array of RecordBatch
 * @param data - object to check
 * @returns {boolean} - true if data is an ArrowData object type guarded
 */
export function isArrowData(data: any): boolean {
  return Array.isArray(data) && Boolean(data.length && data[0].data && data[0].schema);
}

export function isGeoJson(json: unknown): json is Feature | FeatureCollection {
  // json can be feature collection
  // or single feature
  return isPlainObject(json) && (isFeature(json) || isFeatureCollection(json));
}

export function isFeature(json: unknown): json is Feature {
  return isPlainObject(json) && json.type === 'Feature' && Boolean(json.geometry);
}

export function isFeatureCollection(json: unknown): json is FeatureCollection {
  return isPlainObject(json) && json.type === 'FeatureCollection' && Boolean(json.features);
}

export function isRowObject(json: any): boolean {
  return Array.isArray(json) && isPlainObject(json[0]);
}

export function isKeplerGlMap(json: unknown): json is ValidKeplerGlMap {
  return Boolean(
    isPlainObject(json) &&
      json.datasets &&
      json.config &&
      json.info &&
      isPlainObject(json.info) &&
      json.info.app === 'kepler.gl'
  );
}

export async function* makeProgressIterator(
  asyncIterator: AsyncIterable<any>,
  info: {size: number}
): AsyncGenerator {
  let rowCount = 0;

  for await (const batch of asyncIterator) {
    // the length could be stored in `batch.length` for arrow batch
    const rowCountInBatch = (batch.data && (batch.data.length || batch.length)) || 0;
    rowCount += rowCountInBatch;
    const percent = Number.isFinite(batch.bytesUsed) ? batch.bytesUsed / info.size : null;

    // Update progress object
    const progress = {
      rowCount,
      rowCountInBatch,
      ...(Number.isFinite(percent) ? {percent} : {})
    };

    yield {...batch, progress};
  }
}

// eslint-disable-next-line complexity
export async function* readBatch(
  asyncIterator: AsyncIterable<any>,
  fileName: string
): AsyncGenerator {
  let result = null;
  const batches = <any>[];
  for await (const batch of asyncIterator) {
    // Last batch will have this special type and will provide all the root
    // properties of the parsed document.
    // Only json parse will have `FINAL_RESULT`
    if (batch.batchType === BATCH_TYPE.FINAL_RESULT) {
      if (batch.container) {
        result = {...batch.container};
      }
      // Set the streamed data correctly is Batch json path is set
      // and the path streamed is not the top level object (jsonpath = '$')
      if (batch.jsonpath && batch.jsonpath.length > 1) {
        const streamingPath = new _JSONPath(batch.jsonpath);
        streamingPath.setFieldAtPath(result, batches);
      } else if (batch.jsonpath && batch.jsonpath.length === 1) {
        // The streamed object is a ROW JSON-batch (jsonpath = '$')
        // row objects
        result = batches;
      }
    } else {
      const batchData = isArrowTable(batch.data) ? batch.data.batches : batch.data;
      for (let i = 0; i < batchData?.length; i++) {
        batches.push(batchData[i]);
      }
    }

    yield {
      ...batch,
      ...(batch.schema ? {headers: Object.keys(batch.schema)} : {}),
      fileName,
      // if dataset is CSV, data is set to the raw batches
      data: result ? result : batches
    };
  }
}

export async function readFileInBatches({
  file,
  loaders = [],
  loadOptions = {}
}: {
  file: File;
  fileCache: FileCacheItem[];
  loaders: Loader[];
  loadOptions: any;
}): Promise<AsyncGenerator> {
  loaders = [JSONLoader, CSVLoader, GeoArrowLoader, ParquetWasmLoader, ...loaders];
  loadOptions = {
    csv: CSV_LOADER_OPTIONS,
    arrow: ARROW_LOADER_OPTIONS,
    json: JSON_LOADER_OPTIONS,
    parquet: PARQUET_LOADER_OPTIONS,
    metadata: true,
    ...loadOptions
  };

  const batchIterator = await parseInBatches(file, loaders, loadOptions);
  const progressIterator = makeProgressIterator(batchIterator, {size: file.size});

  return readBatch(progressIterator, file.name);
}

export async function processFileData({
  content,
  fileCache
}: {
  content: ProcessFileDataContent;
  fileCache: FileCacheItem[];
}): Promise<FileCacheItem[]> {
  const {fileName, data} = content;
  let format: string | undefined;
  let processor: ((data: any) => ProcessorResult | LoadedMap | null) | undefined;
  // generate unique id with length of 4 using fileName string
  const id = generateHashIdFromString(fileName);
  // decide on which table class to use based on application config
  const table = getApplicationConfig().table ?? KeplerTable;

  if (typeof table.getFileProcessor === 'function') {
    // use custom processors from table class
    const processorResult = table.getFileProcessor(data);
    format = processorResult.format;
    processor = processorResult.processor;
  } else {
    // use default processors
    if (isArrowData(data)) {
      format = DATASET_FORMATS.arrow;
      processor = processArrowBatches;
    } else if (isKeplerGlMap(data)) {
      format = DATASET_FORMATS.keplergl;
      processor = processKeplerglJSON;
    } else if (isRowObject(data)) {
      // csv file goes here
      format = DATASET_FORMATS.row;
      processor = processRowObject;
    } else if (isGeoJson(data)) {
      format = DATASET_FORMATS.geojson;
      processor = processGeojson;
    }
  }
  if (format && processor) {
    // eslint-disable-next-line no-useless-catch
    let result;
    try {
      result = await processor(data);
    } catch (error) {
      throw new Error(`Can not process uploaded file, ${getError(error as Error)}`);
    }

    return [
      ...fileCache,
      {
        data: result,
        info: {
          id,
          label: content.fileName,
          format
        }
      }
    ];
  } else {
    throw new Error('Can not process uploaded file, unknown file format');
  }
}

export function filesToDataPayload(fileCache: FileCacheItem[]): AddDataToMapPayload[] {
  // seperate out files which could be a single datasets. or a keplergl map json
  const collection = fileCache.reduce<{
    datasets: FileCacheItem[];
    keplerMaps: AddDataToMapPayload[];
  }>(
    (accu, file) => {
      const {data, info} = file;
      if (info?.format === DATASET_FORMATS.keplergl) {
        // if file contains a single kepler map dataset & config
        accu.keplerMaps.push({
          ...data,
          options: {
            centerMap: !(data.config && data.config.mapState)
          }
        });
      } else if (DATASET_FORMATS[info?.format]) {
        // if file contains only data
        const newDataset = {
          data,
          info: {
            id: info?.id || generateHashId(4),
            ...(info || {})
          }
        };
        accu.datasets.push(newDataset);
      }
      return accu;
    },
    {datasets: [], keplerMaps: []}
  );

  // add kepler map first with config
  // add datasets later in one add data call
  return collection.keplerMaps.concat({datasets: collection.datasets});
}

1	// SPDX-License-Identifier: MIT
2	// Copyright contributors to the kepler.gl project
3
4	import {parseInBatches} from '@loaders.gl/core';
5	import {JSONLoader, _JSONPath} from '@loaders.gl/json';
6	import {CSVLoader} from '@loaders.gl/csv';
7	import {GeoArrowLoader} from '@loaders.gl/arrow';
8	import {ParquetWasmLoader} from '@loaders.gl/parquet';
9	import {Loader} from '@loaders.gl/loader-utils';
10	import {
11	isPlainObject,
12	generateHashIdFromString,
13	getApplicationConfig,
14	getError,
15	isArrowTable
16	} from '@kepler.gl/utils';
17	import {generateHashId} from '@kepler.gl/common-utils';
18	import {DATASET_FORMATS} from '@kepler.gl/constants';
19	import {AddDataToMapPayload, Feature, LoadedMap, ProcessorResult} from '@kepler.gl/types';
20	import {KeplerTable} from '@kepler.gl/table';
21	import {FeatureCollection} from '@turf/helpers';
22
23	import {
24	processArrowBatches,
25	processGeojson,
26	processKeplerglJSON,
27	processRowObject
28	} from './data-processor';
29
30	import {FileCacheItem, ValidKeplerGlMap} from './types';
31
32	const BATCH_TYPE = {	13✔
33	METADATA: 'metadata',
34	PARTIAL_RESULT: 'partial-result',
35	FINAL_RESULT: 'final-result'
36	};
37
38	const CSV_LOADER_OPTIONS = {	13✔
39	shape: 'object-row-table',
40	dynamicTyping: false // not working for now
41	};
42
43	const ARROW_LOADER_OPTIONS = {	13✔
44	shape: 'arrow-table',
45	batchDebounceMs: 10 // time to delay between batches, for incremental loading
46	};
47
48	const PARQUET_LOADER_OPTIONS = {	13✔
49	shape: 'arrow-table'
50	};
51
52	const JSON_LOADER_OPTIONS = {	13✔
53	shape: 'object-row-table',
54	// instruct loaders.gl on what json paths to stream
55	jsonpaths: [
56	'$', // JSON Row array
57	'$.features', // GeoJSON
58	'$.datasets' // KeplerGL JSON
59	]
60	};
61
62	export type ProcessFileDataContent = {
63	data: unknown;
64	fileName: string;
65	length?: number;
66	progress?: {rowCount?: number; rowCountInBatch?: number; percent?: number};
67	/** metadata e.g. for arrow data, metadata could be the schema.fields */
68	metadata?: Map<string, string>;
69	};
70
71	export {isArrowTable};
72
73	/**
74	* check if data is an ArrowData object, which is an array of RecordBatch
75	* @param data - object to check
76	* @returns {boolean} - true if data is an ArrowData object type guarded
77	*/
78	export function isArrowData(data: any): boolean {
79	return Array.isArray(data) && Boolean(data.length && data[0].data && data[0].schema);	5!
80	}
81
82	export function isGeoJson(json: unknown): json is Feature \| FeatureCollection {
83	// json can be feature collection
84	// or single feature
85	return isPlainObject(json) && (isFeature(json) \|\| isFeatureCollection(json));	2✔
86	}
87
88	export function isFeature(json: unknown): json is Feature {
89	return isPlainObject(json) && json.type === 'Feature' && Boolean(json.geometry);	2✔
90	}
91
92	export function isFeatureCollection(json: unknown): json is FeatureCollection {
93	return isPlainObject(json) && json.type === 'FeatureCollection' && Boolean(json.features);	1✔
94	}
95
96	export function isRowObject(json: any): boolean {
97	return Array.isArray(json) && isPlainObject(json[0]);	4✔
98	}
99
100	export function isKeplerGlMap(json: unknown): json is ValidKeplerGlMap {
101	return Boolean(	8✔
102	isPlainObject(json) &&	22✔
103	json.datasets &&
104	json.config &&
105	json.info &&
106	isPlainObject(json.info) &&
107	json.info.app === 'kepler.gl'
108	);
109	}
110
111	export async function* makeProgressIterator(
112	asyncIterator: AsyncIterable<any>,
113	info: {size: number}
114	): AsyncGenerator {
115	let rowCount = 0;	6✔
116
117	for await (const batch of asyncIterator) {	6✔
118	// the length could be stored in `batch.length` for arrow batch
119	const rowCountInBatch = (batch.data && (batch.data.length \|\| batch.length)) \|\| 0;	18✔
120	rowCount += rowCountInBatch;	18✔
121	const percent = Number.isFinite(batch.bytesUsed) ? batch.bytesUsed / info.size : null;	18✔
122
123	// Update progress object
124	const progress = {	18✔
125	rowCount,
126	rowCountInBatch,
127	...(Number.isFinite(percent) ? {percent} : {})	18✔
128	};
129
130	yield {...batch, progress};	18✔
131	}
132	}
133
134	// eslint-disable-next-line complexity
135	export async function* readBatch(
136	asyncIterator: AsyncIterable<any>,
137	fileName: string
138	): AsyncGenerator {
139	let result = null;	5✔
140	const batches = <any>[];	5✔
141	for await (const batch of asyncIterator) {	5✔
142	// Last batch will have this special type and will provide all the root
143	// properties of the parsed document.
144	// Only json parse will have `FINAL_RESULT`
145	if (batch.batchType === BATCH_TYPE.FINAL_RESULT) {	16✔
146	if (batch.container) {	4!
147	result = {...batch.container};	4✔
148	}
149	// Set the streamed data correctly is Batch json path is set
150	// and the path streamed is not the top level object (jsonpath = '$')
151	if (batch.jsonpath && batch.jsonpath.length > 1) {	4✔
152	const streamingPath = new _JSONPath(batch.jsonpath);	2✔
153	streamingPath.setFieldAtPath(result, batches);	2✔
154	} else if (batch.jsonpath && batch.jsonpath.length === 1) {	2✔
155	// The streamed object is a ROW JSON-batch (jsonpath = '$')
156	// row objects
157	result = batches;	1✔
158	}
159	} else {
160	const batchData = isArrowTable(batch.data) ? batch.data.batches : batch.data;	12!
161	for (let i = 0; i < batchData?.length; i++) {	12✔
162	batches.push(batchData[i]);	48✔
163	}
164	}
165
166	yield {	16✔
167	...batch,
168	...(batch.schema ? {headers: Object.keys(batch.schema)} : {}),	16✔
169	fileName,
170	// if dataset is CSV, data is set to the raw batches
171	data: result ? result : batches	16✔
172	};
173	}
174	}
175
176	export async function readFileInBatches({
177	file,
178	loaders = [],	5✔
179	loadOptions = {}	5✔
180	}: {
181	file: File;
182	fileCache: FileCacheItem[];
183	loaders: Loader[];
184	loadOptions: any;
185	}): Promise<AsyncGenerator> {
186	loaders = [JSONLoader, CSVLoader, GeoArrowLoader, ParquetWasmLoader, ...loaders];	5✔
187	loadOptions = {	5✔
188	csv: CSV_LOADER_OPTIONS,
189	arrow: ARROW_LOADER_OPTIONS,
190	json: JSON_LOADER_OPTIONS,
191	parquet: PARQUET_LOADER_OPTIONS,
192	metadata: true,
193	...loadOptions
194	};
195
196	const batchIterator = await parseInBatches(file, loaders, loadOptions);	5✔
197	const progressIterator = makeProgressIterator(batchIterator, {size: file.size});	5✔
198
199	return readBatch(progressIterator, file.name);	5✔
200	}
201
202	export async function processFileData({
203	content,
204	fileCache
205	}: {
206	content: ProcessFileDataContent;
207	fileCache: FileCacheItem[];
208	}): Promise<FileCacheItem[]> {
209	const {fileName, data} = content;	5✔
210	let format: string \| undefined;
211	let processor: ((data: any) => ProcessorResult \| LoadedMap \| null) \| undefined;
212	// generate unique id with length of 4 using fileName string
213	const id = generateHashIdFromString(fileName);	5✔
214	// decide on which table class to use based on application config
215	const table = getApplicationConfig().table ?? KeplerTable;	5✔
216
217	if (typeof table.getFileProcessor === 'function') {	5!
218	// use custom processors from table class
219	const processorResult = table.getFileProcessor(data);	×
220	format = processorResult.format;	×
221	processor = processorResult.processor;	×
222	} else {
223	// use default processors
224	if (isArrowData(data)) {	5!
225	format = DATASET_FORMATS.arrow;	×
226	processor = processArrowBatches;	×
227	} else if (isKeplerGlMap(data)) {	5✔
228	format = DATASET_FORMATS.keplergl;	1✔
229	processor = processKeplerglJSON;	1✔
230	} else if (isRowObject(data)) {	4✔
231	// csv file goes here
232	format = DATASET_FORMATS.row;	2✔
233	processor = processRowObject;	2✔
234	} else if (isGeoJson(data)) {	2!
235	format = DATASET_FORMATS.geojson;	2✔
236	processor = processGeojson;	2✔
237	}
238	}
239	if (format && processor) {	5!
240	// eslint-disable-next-line no-useless-catch
241	let result;
242	try {	5✔
243	result = await processor(data);	5✔
244	} catch (error) {
245	throw new Error(`Can not process uploaded file, ${getError(error as Error)}`);	×
246	}
247
248	return [	5✔
249	...fileCache,
250	{
251	data: result,
252	info: {
253	id,
254	label: content.fileName,
255	format
256	}
257	}
258	];
259	} else {
260	throw new Error('Can not process uploaded file, unknown file format');	×
261	}
262	}
263
264	export function filesToDataPayload(fileCache: FileCacheItem[]): AddDataToMapPayload[] {
265	// seperate out files which could be a single datasets. or a keplergl map json
266	const collection = fileCache.reduce<{	1✔
267	datasets: FileCacheItem[];
268	keplerMaps: AddDataToMapPayload[];
269	}>(
270	(accu, file) => {
271	const {data, info} = file;	2✔
272	if (info?.format === DATASET_FORMATS.keplergl) {	2✔
273	// if file contains a single kepler map dataset & config
274	accu.keplerMaps.push({	1✔
275	...data,
276	options: {
277	centerMap: !(data.config && data.config.mapState)	2✔
278	}
279	});
280	} else if (DATASET_FORMATS[info?.format]) {	1!
281	// if file contains only data
282	const newDataset = {	1✔
283	data,
284	info: {
285	id: info?.id \|\| generateHashId(4),	2✔
286	...(info \|\| {})	1!
287	}
288	};
289	accu.datasets.push(newDataset);	1✔
290	}
291	return accu;	2✔
292	},
293	{datasets: [], keplerMaps: []}
294	);
295
296	// add kepler map first with config
297	// add datasets later in one add data call
298	return collection.keplerMaps.concat({datasets: collection.datasets});	1✔
299	}

keplergl / kepler.gl / 19768106976

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous