@ellipticltd/aml-utils 0.15.20 → 0.15.22

package/lib/structured-file-parser/structured-file-parser.ts

@@ -1,129 +0,0 @@
-/* eslint-disable no-await-in-loop */
-/* eslint-disable no-restricted-syntax */
-import parseRow from './parse-row';
-import { InvalidFileError, TooManyRowsError, NoRowsError, InvalidHeadersError } from './errors';
-import sanitizeRows from './sanitize-rows';
-
-export type ProcessingEntry = {
-  isValid: boolean;
-};
-
-type Dependencies<T extends ProcessingEntry> = {
-  processRow: (cells: string[], headers: string[]) => T;
-};
-
-type CompleteDependencies<T extends ProcessingEntry, U extends string[]> = {
-  processRow: (cells: string[], headers: U) => T;
-  headers: U;
-};
-
-function chunk<T>(array: T[], size: number): T[][] {
-  if (!array.length) {
-    return [];
-  }
-  const head = array.slice(0, size);
-  const tail = array.slice(size);
-
-  return [head, ...chunk(tail, size)];
-}
-
-async function processChunk<T extends ProcessingEntry, U extends string[]>(
-  parsedRows: string[][],
-  dependencies: CompleteDependencies<T, U>
-): Promise<T[]> {
-  const { processRow, headers } = dependencies;
-  return new Promise((resolve: (value: T[] | undefined) => void, reject: (error: Error) => void): void => {
-    setTimeout(() => {
-      try {
-        const processedRows = parsedRows.map((row) => processRow(row, headers));
-        resolve(processedRows);
-      } catch (ex) {
-        reject(ex);
-      }
-    }, 0);
-  });
-}
-
-// a row is said to contain headers if two valid header names are present
-export const hasHeaderRow = (rows: string[][], requiredHeaderNames: string[]): boolean => {
-  const firstRow = rows[0];
-  if (!firstRow) {
-    return false;
-  }
-  const foundIndex = firstRow.findIndex((value) => requiredHeaderNames.includes(value.toLowerCase()));
-  if (foundIndex < 0) {
-    return false;
-  }
-  const containsSecondHeader = firstRow
-    .filter((_val, i) => i !== foundIndex)
-    .some((value) => requiredHeaderNames.includes(value.toLowerCase()));
-  return containsSecondHeader;
-};
-
-export const isHeaderRowValid = (rows: string[][], requiredHeaderNames: string[]): boolean => {
-  const firstRow = rows[0];
-  if (!firstRow) {
-    return false;
-  }
-  return (
-    requiredHeaderNames.every((header) => firstRow.map((value) => value.toLowerCase()).includes(header)) &&
-    requiredHeaderNames.length === firstRow.length
-  );
-};
-
-async function parseFile<T extends ProcessingEntry>(
-  fileContent: string,
-  maxRows: number,
-  requiredHeaderNames: string[],
-  dependencies: Dependencies<T>
-): Promise<T[]> {
-  if (fileContent === 'invalidFileType') {
-    throw new InvalidFileError();
-  }
-
-  const rows = fileContent.split('\n');
-  const parsedRows = rows.map(parseRow);
-
-  const sanitizedRows = sanitizeRows(parsedRows);
-
-  if (sanitizedRows.length > maxRows) {
-    throw new TooManyRowsError();
-  }
-
-  if (sanitizedRows.length === 0) {
-    throw new NoRowsError();
-  }
-
-  const containsHeaders = hasHeaderRow(sanitizedRows, requiredHeaderNames);
-  if (containsHeaders && !isHeaderRowValid(sanitizedRows, requiredHeaderNames)) {
-    throw new InvalidHeadersError();
-  }
-
-  const firstSanitizedRow = sanitizedRows[0];
-
-  if (!firstSanitizedRow) {
-    throw new NoRowsError();
-  }
-
-  const headers = containsHeaders ? firstSanitizedRow.map((value) => value.toLowerCase()) : requiredHeaderNames;
-
-  const dataRows = containsHeaders ? sanitizedRows.splice(1) : sanitizedRows;
-
-  if (dataRows.length === 0) {
-    throw new NoRowsError();
-  }
-
-  const parsedRowChunks = chunk(dataRows, 30);
-  let results: T[] = [];
-  for (const currentRowChunk of parsedRowChunks) {
-    const chunkResult = await processChunk(currentRowChunk, {
-      ...dependencies,
-      headers,
-    });
-    results = results.concat(chunkResult);
-  }
-
-  return results;
-}
-
-export default parseFile;