1
0
Fork 0
mirror of https://github.com/denoland/deno.git synced 2024-11-04 08:54:20 -05:00
denoland-deno/std/encoding/csv.ts

359 lines
9.8 KiB
TypeScript
Raw Normal View History

// Ported from Go:
// https://github.com/golang/go/blob/go1.12.5/src/encoding/csv/
// Copyright 2011 The Go Authors. All rights reserved. BSD license.
// https://github.com/golang/go/blob/master/LICENSE
2020-01-02 15:13:47 -05:00
// Copyright 2018-2020 the Deno authors. All rights reserved. MIT license.
import { BufReader } from "../io/bufio.ts";
import { TextProtoReader } from "../textproto/mod.ts";
import { StringReader } from "../io/readers.ts";
import { assert } from "../testing/asserts.ts";
const INVALID_RUNE = ["\r", "\n", '"'];
export const ERR_BARE_QUOTE = 'bare " in non-quoted-field';
export const ERR_QUOTE = 'extraneous or missing " in quoted-field';
export const ERR_INVALID_DELIM = "Invalid Delimiter";
export const ERR_FIELD_COUNT = "wrong number of fields";
export class ParseError extends Error {
StartLine: number;
Line: number;
constructor(start: number, line: number, message: string) {
super(message);
this.StartLine = start;
this.Line = line;
}
}
/**
* @property comma - Character which separates values. Default: ','
* @property comment - Character to start a comment. Default: '#'
* @property trimLeadingSpace - Flag to trim the leading space of the value.
* Default: 'false'
* @property lazyQuotes - Allow unquoted quote in a quoted field or non double
* quoted quotes in quoted field Default: 'false'
* @property fieldsPerRecord - Enabling the check of fields for each row.
Fix typos across the repo (#5295) Corrections made: * cli/js/tests/README.md:44:7: corrected "discoveres" to "discovers" * cli/js/tests/chown_test.ts:111:37: corrected "priviledge" to "privilege" * cli/worker.rs:231:56: corrected "decendants" to "descendants" * deno_typescript/lib.rs:136:50: corrected "emmited" to "emitted" * core/es_isolate.rs:492:67: corrected "registerd" to "registered" * core/isolate.rs:103:28: corrected "initalize" to "initialize" * docs/runtime.md:29:14: corrected "ect" to "etc" * docs/tools/debugger.md:122:16: corrected "implementes" to "implements" * std/encoding/_yaml/dumper/dumper_state.ts:57:63: corrected "everwhere" to "everywhere" * std/encoding/csv.ts:37:43: corrected "referal" to "referral" * std/fmt/sprintf.ts:209:20: corrected "unusuable" to "unusable" * std/fmt/README.md:21:40: corrected "Alternativly" to "Alternatively" * std/fmt/README.md:35:68: corrected "seperated" to "separated" * std/fmt/README.md:179:59: corrected "provded" to "provided" * std/mime/multipart.ts:581:46: corrected "writen" to "written" * std/path/_globrex.ts:19:52: corrected "equivelant" to "equivalent" * std/node/events_test.ts:447:9: corrected "asyncronous" to "asynchronous" * std/node/events_test.ts:475:9: corrected "asyncronous" to "asynchronous" * std/node/events_test.ts:500:29: corrected "asyncronous" to "asynchronous" * std/node/events_test.ts:530:40: corrected "asyncronous" to "asynchronous" * std/node/events_test.ts:555:9: corrected "asyncronous" to "asynchronous" * tools/deno_tcp_proxy.ts:1:42: corrected "perfromance" to "performance" * std/node/module.ts:1003:18: corrected "existend" to "existed"
2020-05-14 00:38:42 -04:00
* If == 0, first row is used as referral for the number of fields.
*/
export interface ReadOptions {
comma?: string;
comment?: string;
trimLeadingSpace?: boolean;
lazyQuotes?: boolean;
fieldsPerRecord?: number;
}
function chkOptions(opt: ReadOptions): void {
if (!opt.comma) {
opt.comma = ",";
}
if (!opt.trimLeadingSpace) {
opt.trimLeadingSpace = false;
}
if (
INVALID_RUNE.includes(opt.comma) ||
(typeof opt.comment === "string" && INVALID_RUNE.includes(opt.comment)) ||
opt.comma === opt.comment
) {
throw new Error(ERR_INVALID_DELIM);
}
}
async function readRecord(
Startline: number,
reader: BufReader,
opt: ReadOptions = { comma: ",", trimLeadingSpace: false }
): Promise<string[] | null> {
const tp = new TextProtoReader(reader);
const lineIndex = Startline;
let line = await readLine(tp);
if (line === null) return null;
if (line.length === 0) {
return [];
}
// line starting with comment character is ignored
if (opt.comment && line[0] === opt.comment) {
return [];
}
assert(opt.comma != null);
let quoteError: string | null = null;
const quote = '"';
const quoteLen = quote.length;
const commaLen = opt.comma.length;
let recordBuffer = "";
const fieldIndexes = [] as number[];
parseField: for (;;) {
if (opt.trimLeadingSpace) {
line = line.trimLeft();
}
if (line.length === 0 || !line.startsWith(quote)) {
// Non-quoted string field
const i = line.indexOf(opt.comma);
let field = line;
if (i >= 0) {
field = field.substring(0, i);
}
// Check to make sure a quote does not appear in field.
if (!opt.lazyQuotes) {
const j = field.indexOf(quote);
if (j >= 0) {
quoteError = ERR_BARE_QUOTE;
break parseField;
}
}
recordBuffer += field;
fieldIndexes.push(recordBuffer.length);
if (i >= 0) {
line = line.substring(i + commaLen);
continue parseField;
}
break parseField;
} else {
// Quoted string field
line = line.substring(quoteLen);
for (;;) {
const i = line.indexOf(quote);
if (i >= 0) {
// Hit next quote.
recordBuffer += line.substring(0, i);
line = line.substring(i + quoteLen);
if (line.startsWith(quote)) {
// `""` sequence (append quote).
recordBuffer += quote;
line = line.substring(quoteLen);
} else if (line.startsWith(opt.comma)) {
// `","` sequence (end of field).
line = line.substring(commaLen);
fieldIndexes.push(recordBuffer.length);
continue parseField;
} else if (0 === line.length) {
// `"\n` sequence (end of line).
fieldIndexes.push(recordBuffer.length);
break parseField;
} else if (opt.lazyQuotes) {
// `"` sequence (bare quote).
recordBuffer += quote;
} else {
// `"*` sequence (invalid non-escaped quote).
quoteError = ERR_QUOTE;
break parseField;
}
} else if (line.length > 0 || !(await isEOF(tp))) {
// Hit end of line (copy all data so far).
recordBuffer += line;
const r = await readLine(tp);
if (r === null) {
if (!opt.lazyQuotes) {
quoteError = ERR_QUOTE;
break parseField;
}
fieldIndexes.push(recordBuffer.length);
break parseField;
}
recordBuffer += "\n"; // preserve line feed (This is because TextProtoReader removes it.)
line = r;
} else {
// Abrupt end of file (EOF on error).
if (!opt.lazyQuotes) {
quoteError = ERR_QUOTE;
break parseField;
}
fieldIndexes.push(recordBuffer.length);
break parseField;
}
}
}
}
if (quoteError) {
throw new ParseError(Startline, lineIndex, quoteError);
}
const result = [] as string[];
let preIdx = 0;
for (const i of fieldIndexes) {
result.push(recordBuffer.slice(preIdx, i));
preIdx = i;
}
return result;
}
async function isEOF(tp: TextProtoReader): Promise<boolean> {
return (await tp.r.peek(0)) === null;
}
async function readLine(tp: TextProtoReader): Promise<string | null> {
let line: string;
const r = await tp.readLine();
if (r === null) return null;
line = r;
// For backwards compatibility, drop trailing \r before EOF.
if ((await isEOF(tp)) && line.length > 0 && line[line.length - 1] === "\r") {
line = line.substring(0, line.length - 1);
}
// Normalize \r\n to \n on all input lines.
if (
line.length >= 2 &&
line[line.length - 2] === "\r" &&
line[line.length - 1] === "\n"
) {
line = line.substring(0, line.length - 2);
line = line + "\n";
}
return line;
}
export async function readMatrix(
reader: BufReader,
opt: ReadOptions = {
comma: ",",
trimLeadingSpace: false,
lazyQuotes: false,
}
): Promise<string[][]> {
const result: string[][] = [];
let _nbFields: number | undefined;
let lineResult: string[];
let first = true;
let lineIndex = 0;
chkOptions(opt);
for (;;) {
const r = await readRecord(lineIndex, reader, opt);
if (r === null) break;
lineResult = r;
lineIndex++;
// If fieldsPerRecord is 0, Read sets it to
// the number of fields in the first record
if (first) {
first = false;
if (opt.fieldsPerRecord !== undefined) {
if (opt.fieldsPerRecord === 0) {
_nbFields = lineResult.length;
} else {
_nbFields = opt.fieldsPerRecord;
}
}
}
if (lineResult.length > 0) {
if (_nbFields && _nbFields !== lineResult.length) {
throw new ParseError(lineIndex, lineIndex, ERR_FIELD_COUNT);
}
result.push(lineResult);
}
}
return result;
}
/**
* HeaderOptions provides the column definition
* and the parse function for each entry of the
* column.
*/
export interface HeaderOptions {
name: string;
parse?: (input: string) => unknown;
}
export interface ParseOptions extends ReadOptions {
header: boolean | string[] | HeaderOptions[];
/** Parse function for rows.
* Example:
* const r = await parseFile('a,b,c\ne,f,g\n', {
* header: ["this", "is", "sparta"],
* parse: (e: Record<string, unknown>) => {
* return { super: e.this, street: e.is, fighter: e.sparta };
* }
* });
* // output
* [
* { super: "a", street: "b", fighter: "c" },
* { super: "e", street: "f", fighter: "g" }
* ]
*/
parse?: (input: unknown) => unknown;
}
/**
* Csv parse helper to manipulate data.
* Provides an auto/custom mapper for columns and parse function
* for columns and rows.
* @param input Input to parse. Can be a string or BufReader.
* @param opt options of the parser.
*/
export async function parse(
input: string | BufReader,
opt: ParseOptions = {
header: false,
}
): Promise<unknown[]> {
let r: string[][];
if (input instanceof BufReader) {
r = await readMatrix(input, opt);
} else {
r = await readMatrix(new BufReader(new StringReader(input)), opt);
}
if (opt.header) {
let headers: HeaderOptions[] = [];
let i = 0;
if (Array.isArray(opt.header)) {
if (typeof opt.header[0] !== "string") {
headers = opt.header as HeaderOptions[];
} else {
const h = opt.header as string[];
headers = h.map(
(e): HeaderOptions => {
return {
name: e,
};
}
);
}
} else {
const head = r.shift();
assert(head != null);
headers = head.map(
(e): HeaderOptions => {
return {
name: e,
};
}
);
i++;
}
return r.map((e): unknown => {
if (e.length !== headers.length) {
throw `Error number of fields line:${i}`;
}
i++;
const out: Record<string, unknown> = {};
for (let j = 0; j < e.length; j++) {
const h = headers[j];
if (h.parse) {
out[h.name] = h.parse(e[j]);
} else {
out[h.name] = e[j];
2019-10-09 17:22:22 -04:00
}
}
if (opt.parse) {
return opt.parse(out);
}
return out;
});
}
if (opt.parse) {
return r.map((e: string[]): unknown => {
assert(opt.parse, "opt.parse must be set");
return opt.parse(e);
});
}
return r;
}