// Original definitions in https://github.com/DefinitelyTyped/DefinitelyTyped by: David Muller <https://github.com/davidm77>

/// <reference types="node" />

import * as stream from "stream";

export type Callback<T = string[]> = (
  err: CsvError | undefined,
  records: T[],
  info?: InfoCallback,
) => void;

export class Parser extends stream.Transform {
  constructor(options: Options);

  // __push(line: T): CsvError | undefined;
  // __push(line: any): CsvError | undefined;

  // __write(chars: any, end: any, callback: any): any;

  readonly options: OptionsNormalized;

  readonly info: Info;
}

export interface Info {
  /**
   * The number of processed bytes.
   */
  readonly bytes: number;
  /**
   * The number of processed bytes until the last successfully parsed and emitted records.
   */
  readonly bytes_records: number;
  /**
   * The number of lines being fully commented.
   */
  readonly comment_lines: number;
  /**
   * The number of processed empty lines.
   */
  readonly empty_lines: number;
  /**
   * The number of non uniform records when `relax_column_count` is true.
   */
  readonly invalid_field_length: number;
  /**
   * The number of lines encountered in the source dataset, start at 1 for the first line.
   */
  readonly lines: number;
  /**
   * The number of processed records.
   */
  readonly records: number;
}

export interface InfoCallback extends Info {
  /**
   * Normalized version of `options.columns` when `options.columns` is true, boolean otherwise.
   */
  readonly columns: boolean | { name: string }[] | { disabled: true }[];
}

export interface InfoDataSet extends Info {
  readonly column: number | string;
}

export interface InfoRecord extends InfoDataSet {
  readonly error: CsvError;
  readonly header: boolean;
  readonly index: number;
  readonly raw: string | undefined;
}

export interface InfoField extends InfoRecord {
  readonly quoting: boolean;
}

/**
 * @deprecated Use the InfoField interface instead, the interface will disappear in future versions.
 */
// eslint-disable-next-line
export interface CastingContext extends InfoField {}

export type CastingFunction = (value: string, context: InfoField) => unknown;

export type CastingDateFunction = (value: string, context: InfoField) => Date;

export type ColumnOption<K = string> =
  K | undefined | null | false | { name: K };

type ColumnKey<T> = T extends string[]
  ? string
  : unknown extends T
    ? string
    : string | keyof T;

// Keep columns from overriding record types inferred from options such as raw.
type NoInferColumnRecord<T> = [T][T extends unknown ? 0 : never];

export interface ScoringFunctionInfo {
  /**
   * The character code of the delimiter candidate being scored.
   */
  readonly char_code: number;
  /**
   * The number of occurrences of the candidate in each line.
   */
  readonly lines: number[];
  /**
   * Whether the candidate is listed in the `preferred` option.
   */
  readonly preferred: boolean;
  /**
   * The standard deviation of the occurrences across the lines.
   */
  readonly std: number;
  /**
   * The total number of occurrences of the candidate.
   */
  readonly total: number;
}

export type ScoringFunction = (
  info: ScoringFunctionInfo,
  options: ScoringFunctionOptions,
) => number;

export interface ScoringFunctionOptions {
  preferred: Record<string, number>;
  score: ScoringFunction;
  size: number;
}

export interface OptionsNormalized<T = string[], U = T> {
  auto_parse?: boolean | CastingFunction;
  auto_parse_date?: boolean | CastingDateFunction;
  /**
   * If true, detect and exclude the byte order mark (BOM) from the CSV input if present.
   */
  bom?: boolean;
  /**
   * If true, the parser will attempt to convert input string to native types.
   * If a function, receive the value as first argument, a context as second argument and return a new value. More information about the context properties is available below.
   */
  cast?: boolean | CastingFunction;
  /**
   * If true, the parser will attempt to convert input string to dates.
   * If a function, receive the value as argument and return a new value. It requires the "auto_parse" option. Be careful, it relies on Date.parse.
   */
  cast_date?: boolean | CastingDateFunction;
  /**
   * Internal property string the function to
   */
  cast_first_line_to_header?: (
    record: string[],
  ) => ColumnOption<ColumnKey<U>>[];
  /**
   * List of fields as an array, a user defined callback accepting the first
   * line and returning the column names or true if autodiscovered in the first
   * CSV line, default to null, affect the result data set in the sense that
   * records will be objects instead of arrays. The callback receives the raw
   * header fields as strings, while returned names may use keys from the typed
   * input record.
   */
  columns: boolean | ColumnOption<ColumnKey<U>>[];
  /**
   * Treat all the characters after this one as a comment, default to '' (disabled).
   */
  comment: string | null;
  /**
   * Restrict the definition of comments to a full line. Comment characters
   * defined in the middle of the line are not interpreted as such. The
   * option require the activation of comments.
   */
  comment_no_infix: boolean;
  /**
   * Set the field delimiter. One character only, defaults to comma.
   */
  delimiter: Buffer[];
  /**
   * Discover the field delimiter.
   */
  delimiter_auto: ScoringFunctionOptions;
  /**
   * Set the source and destination encoding, a value of `null` returns buffer instead of strings.
   */
  encoding: BufferEncoding | null;
  /**
   * Set the escape character, one character only, defaults to double quotes.
   */
  escape: null | Buffer;
  /**
   * Start handling records from the requested number of records.
   */
  from: number;
  /**
   * Start handling records from the requested line number.
   */
  from_line: number;
  /**
   * Convert values into an array of values when columns are activated and
   * when multiple columns of the same name are found.
   */
  group_columns_by_name: boolean;
  /**
   * Don't interpret delimiters as such in the last field according to the number of fields calculated from the number of columns, the option require the presence of the `column` option when `true`.
   */
  ignore_last_delimiters: boolean | number;
  /**
   * Generate two properties `info` and `record` where `info` is a snapshot of the info object at the time the record was created and `record` is the parsed array or object.
   */
  info: boolean;
  /**
   * If true, ignore whitespace immediately following the delimiter (i.e. left-trim all fields), defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  ltrim: boolean;
  /**
   * Maximum number of characters to be contained in the field and line buffers before an exception is raised,
   * used to guard against a wrong delimiter or record_delimiter,
   * default to 128000 characters.
   */
  max_record_size: number;
  /**
   * Name of header-record title to name objects by.
   */
  objname: number | string | undefined;
  /**
   * Alter and filter records by executing a user defined function.
   */
  on_record?: (record: U, context: InfoRecord) => T | null | undefined;
  /**
   * Function called when an error occurred if the `skip_records_with_error`
   * option is activated.
   */
  on_skip?: (err: CsvError | undefined, raw: string | undefined) => undefined;
  /**
   * Optional character surrounding a field, one character only, defaults to double quotes.
   */
  quote?: Buffer | null;
  /**
   * Generate two properties raw and row where raw is the original CSV row content and row is the parsed array or object.
   */
  raw: boolean;
  /**
   * One or multiple characters used to delimit record rows; defaults to auto discovery if not provided.
   * Supported auto discovery method are Linux ("\n"), Apple ("\r") and Windows ("\r\n") row delimiters.
   */
  record_delimiter: Buffer[];
  /**
   * Discard inconsistent columns count, default to false.
   */
  relax_column_count: boolean;
  /**
   * Discard inconsistent columns count when the record contains less fields than expected, default to false.
   */
  relax_column_count_less: boolean;
  /**
   * Discard inconsistent columns count when the record contains more fields than expected, default to false.
   */
  relax_column_count_more: boolean;
  /**
   * Preserve quotes inside unquoted field.
   */
  relax_quotes: boolean;
  /**
   * If true, ignore whitespace immediately preceding the delimiter (i.e. right-trim all fields), defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  rtrim: boolean;
  /**
   * Dont generate empty values for empty lines.
   * Defaults to false
   */
  skip_empty_lines: boolean;
  /**
   * Don't generate records for lines containing empty column values (column matching /\s*\/), defaults to false.
   */
  skip_records_with_empty_values: boolean;
  /**
   * Skip a line with error found inside and directly go process the next line.
   */
  skip_records_with_error: boolean;
  /**
   * Stop handling records after the requested number of records.
   */
  to: number;
  /**
   * Stop handling records after the requested line number.
   */
  to_line: number;
  /**
   * If true, ignore whitespace immediately around the delimiter, defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  trim: boolean;
}

/*
Note, could not `extends stream.TransformOptions` because encoding can be
BufferEncoding and undefined as well as null which is not defined in the
extended type.
*/
export interface Options<T = string[], U = T> {
  /**
   * If true, the parser will attempt to convert read data types to native types.
   * @deprecated Use {@link cast}
   */
  auto_parse?: boolean | CastingFunction;
  autoParse?: boolean | CastingFunction;
  /**
   * If true, the parser will attempt to convert read data types to dates. It requires the "auto_parse" option.
   * @deprecated Use {@link cast_date}
   */
  auto_parse_date?: boolean | CastingDateFunction;
  autoParseDate?: boolean | CastingDateFunction;
  /**
   * If true, detect and exclude the byte order mark (BOM) from the CSV input if present.
   */
  bom?: OptionsNormalized["bom"];
  /**
   * If true, the parser will attempt to convert input string to native types.
   * If a function, receive the value as first argument, a context as second argument and return a new value. More information about the context properties is available below.
   */
  cast?: OptionsNormalized["cast"];
  /**
   * If true, the parser will attempt to convert input string to dates.
   * If a function, receive the value as argument and return a new value. It requires the "auto_parse" option. Be careful, it relies on Date.parse.
   */
  cast_date?: OptionsNormalized["cast_date"];
  castDate?: OptionsNormalized["cast_date"];
  /**
   * List of fields as an array,
   * a user defined callback accepting the first line and returning the column names or true if autodiscovered in the first CSV line,
   * default to null,
   * affect the result data set in the sense that records will be objects instead of arrays. The callback receives the raw header fields as strings, while returned names may use keys from the typed input record.
   */
  columns?:
    | OptionsNormalized<T, U>["columns"]
    | ((record: string[]) => ColumnOption<ColumnKey<U>>[]);
  /**
   * Treat all the characters after this one as a comment, default to '' (disabled).
   */
  comment?: OptionsNormalized["comment"] | boolean;
  /**
   * Restrict the definition of comments to a full line. Comment characters
   * defined in the middle of the line are not interpreted as such. The
   * option require the activation of comments.
   */
  comment_no_infix?: OptionsNormalized["comment_no_infix"] | null;
  /**
   * Set the field delimiter. One character only, defaults to comma.
   */
  delimiter?: OptionsNormalized["delimiter"] | string | string[] | Buffer;
  /**
   * Discover the field delimiter
   */
  delimiter_auto?: boolean | Partial<OptionsNormalized["delimiter_auto"]>;
  /**
   * Set the source and destination encoding, a value of `null` returns buffer instead of strings.
   */
  encoding?: OptionsNormalized["encoding"] | boolean | undefined;
  /**
   * Set the escape character, one character only, defaults to double quotes.
   */
  escape?: OptionsNormalized["escape"] | string | boolean;
  /**
   * Start handling records from the requested number of records.
   */
  from?: OptionsNormalized["from"] | string;
  /**
   * Start handling records from the requested line number.
   */
  from_line?: OptionsNormalized["from_line"] | null | string;
  fromLine?: OptionsNormalized["from_line"] | null | string;
  /**
   * Convert values into an array of values when columns are activated and
   * when multiple columns of the same name are found.
   */
  group_columns_by_name?: OptionsNormalized["group_columns_by_name"];
  groupColumnsByName?: OptionsNormalized["group_columns_by_name"];
  /**
   * Don't interpret delimiters as such in the last field according to the number of fields calculated from the number of columns, the option require the presence of the `column` option when `true`.
   */
  ignore_last_delimiters?: OptionsNormalized["ignore_last_delimiters"];
  /**
   * Generate two properties `info` and `record` where `info` is a snapshot of the info object at the time the record was created and `record` is the parsed array or object.
   */
  info?: OptionsNormalized["info"];
  /**
   * If true, ignore whitespace immediately following the delimiter (i.e. left-trim all fields), defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  ltrim?: OptionsNormalized["ltrim"] | null;
  /**
   * Maximum number of characters to be contained in the field and line buffers before an exception is raised,
   * used to guard against a wrong delimiter or record_delimiter,
   * default to 128000 characters.
   */
  max_record_size?: OptionsNormalized["max_record_size"] | null | string;
  maxRecordSize?: OptionsNormalized["max_record_size"];
  /**
   * Name of header-record title to name objects by.
   */
  objname?: OptionsNormalized["objname"] | Buffer | null;
  /**
   * Alter and filter records by executing a user defined function.
   */
  on_record?: (record: U, context: InfoRecord) => T | null | undefined | U;
  onRecord?: (record: U, context: InfoRecord) => T | null | undefined | U;
  /**
   * Function called when an error occurred if the `skip_records_with_error`
   * option is activated.
   */
  on_skip?: OptionsNormalized["on_skip"];
  onSkip?: OptionsNormalized["on_skip"];
  /**
   * Optional character surrounding a field, one character only, defaults to double quotes.
   */
  quote?: OptionsNormalized["quote"] | string | boolean;
  /**
   * Generate two properties raw and row where raw is the original CSV row content and row is the parsed array or object.
   */
  raw?: OptionsNormalized["raw"] | null;
  /**
   * One or multiple characters used to delimit record rows; defaults to auto discovery if not provided.
   * Supported auto discovery method are Linux ("\n"), Apple ("\r") and Windows ("\r\n") row delimiters.
   */
  record_delimiter?:
    | OptionsNormalized["record_delimiter"]
    | string
    | Buffer
    | null
    | (string | null)[];
  recordDelimiter?:
    | OptionsNormalized["record_delimiter"]
    | string
    | Buffer
    | null
    | (string | null)[];
  /**
   * Discard inconsistent columns count, default to false.
   */
  relax_column_count?: OptionsNormalized["relax_column_count"] | null;
  relaxColumnCount?: OptionsNormalized["relax_column_count"] | null;
  /**
   * Discard inconsistent columns count when the record contains less fields than expected, default to false.
   */
  relax_column_count_less?: OptionsNormalized["relax_column_count_less"] | null;
  relaxColumnCountLess?: OptionsNormalized["relax_column_count_less"] | null;
  /**
   * Discard inconsistent columns count when the record contains more fields than expected, default to false.
   */
  relax_column_count_more?: OptionsNormalized["relax_column_count_more"] | null;
  relaxColumnCountMore?: OptionsNormalized["relax_column_count_more"] | null;
  /**
   * Preserve quotes inside unquoted field.
   */
  relax_quotes?: OptionsNormalized["relax_quotes"] | null;
  relaxQuotes?: OptionsNormalized["relax_quotes"] | null;
  /**
   * If true, ignore whitespace immediately preceding the delimiter (i.e. right-trim all fields), defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  rtrim?: OptionsNormalized["rtrim"] | null;
  /**
   * Dont generate empty values for empty lines.
   * Defaults to false
   */
  skip_empty_lines?: OptionsNormalized["skip_empty_lines"] | null;
  skipEmptyLines?: OptionsNormalized["skip_empty_lines"] | null;
  /**
   * Don't generate records for lines containing empty column values (column matching /\s*\/), defaults to false.
   */
  skip_records_with_empty_values?:
    OptionsNormalized["skip_records_with_empty_values"] | null;
  skipRecordsWithEmptyValues?:
    OptionsNormalized["skip_records_with_empty_values"] | null;
  /**
   * Skip a line with error found inside and directly go process the next line.
   */
  skip_records_with_error?: OptionsNormalized["skip_records_with_error"] | null;
  skipRecordsWithError?: OptionsNormalized["skip_records_with_error"] | null;
  /**
   * Stop handling records after the requested number of records.
   */
  to?: OptionsNormalized["to"] | null | string;
  /**
   * Stop handling records after the requested line number.
   */
  to_line?: OptionsNormalized["to_line"] | null | string;
  toLine?: OptionsNormalized["to_line"] | null | string;
  /**
   * If true, ignore whitespace immediately around the delimiter, defaults to false.
   * Does not remove whitespace in a quoted field.
   */
  trim?: OptionsNormalized["trim"] | null;
}

export type OptionsWithColumns<T, U = T> = Omit<Options<T, U>, "columns"> & {
  columns: Exclude<
    Options<NoInferColumnRecord<T>, NoInferColumnRecord<U>>["columns"],
    undefined | false
  >;
};

declare function parse<T = unknown, U = T>(
  input: string | Buffer | Uint8Array,
  options: OptionsWithColumns<T, U>,
  callback?: Callback<T>,
): Parser;
declare function parse(
  input: string | Buffer | Uint8Array,
  options: Options,
  callback?: Callback,
): Parser;

declare function parse<T = unknown, U = T>(
  options: OptionsWithColumns<T, U>,
  callback?: Callback<T>,
): Parser;
declare function parse(options: Options, callback?: Callback): Parser;

declare function parse(
  input: string | Buffer | Uint8Array,
  callback?: Callback,
): Parser;
declare function parse(callback?: Callback): Parser;

export { parse };

/////////////////////////////////////////////////////// CsvError

export type CsvErrorCode =
  | "CSV_INVALID_ARGUMENT"
  | "CSV_INVALID_CLOSING_QUOTE"
  | "CSV_INVALID_COLUMN_DEFINITION"
  | "CSV_INVALID_COLUMN_MAPPING"
  | "CSV_INVALID_OPTION_BOM"
  | "CSV_INVALID_OPTION_CAST"
  | "CSV_INVALID_OPTION_CAST_DATE"
  | "CSV_INVALID_OPTION_COLUMNS"
  | "CSV_INVALID_OPTION_COMMENT"
  | "CSV_INVALID_OPTION_DELIMITER"
  | "CSV_INVALID_OPTION_GROUP_COLUMNS_BY_NAME"
  | "CSV_INVALID_OPTION_ON_RECORD"
  | "CSV_MAX_RECORD_SIZE"
  | "CSV_NON_TRIMABLE_CHAR_AFTER_CLOSING_QUOTE"
  | "CSV_OPTION_COLUMNS_MISSING_NAME"
  | "CSV_QUOTE_NOT_CLOSED"
  | "CSV_RECORD_INCONSISTENT_FIELDS_LENGTH"
  | "CSV_RECORD_INCONSISTENT_COLUMNS"
  | "CSV_UNKNOWN_ERROR"
  | "INVALID_OPENING_QUOTE";

export class CsvError extends Error {
  readonly code: CsvErrorCode;
  [key: string]: unknown;

  constructor(
    code: CsvErrorCode,
    message: string | string[],
    options?: OptionsNormalized,
    ...contexts: unknown[]
  );
}

/////////////////////////////////////////////////////// normalize_options

declare function normalize_options(opts: Options): OptionsNormalized;
export { normalize_options };