deno.land / std@0.224.0 / csv / parse_test.ts
parse_test.ts
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909// Test ported from Golang// https://github.com/golang/go/blob/2cc15b1/src/encoding/csv/reader_test.go// Copyright 2011 The Go Authors. All rights reserved. BSD license.// https://github.com/golang/go/blob/master/LICENSE// Copyright 2018-2024 the Deno authors. All rights reserved. MIT license.
import { assert, assertEquals, assertThrows } from "../assert/mod.ts";import { parse, ParseError, type ParseOptions } from "./parse.ts";import type { AssertTrue, IsExact } from "../testing/types.ts";
const BYTE_ORDER_MARK = "\ufeff";
Deno.test({ name: "parse() handles various inputs", async fn(t) { await t.step({ name: "Simple", fn() { const input = "a,b,c\n"; assertEquals( parse(input), [["a", "b", "c"]], ); }, }); await t.step({ name: "CRLF", fn() { const input = "a,b\r\nc,d\r\n"; assertEquals( parse(input), [ ["a", "b"], ["c", "d"], ], ); }, });
await t.step({ name: "BareCR", fn() { const input = "a,b\rc,d\r\n"; assertEquals( parse(input), [["a", "b\rc", "d"]], ); }, });
await t.step({ name: "RFC4180test", fn() { const input = '#field1,field2,field3\n"aaa","bbb","ccc"\n"a,a","bbb","ccc"\nzzz,yyy,xxx'; assertEquals( parse(input), [ ["#field1", "field2", "field3"], ["aaa", "bbb", "ccc"], ["a,a", `bbb`, "ccc"], ["zzz", "yyy", "xxx"], ], ); }, }); await t.step({ name: "NoEOLTest", fn() { const input = "a,b,c"; assertEquals( parse(input), [["a", "b", "c"]], ); }, });
await t.step({ name: "Semicolon", fn() { const input = "a;b;c\n"; assertEquals( parse(input, { separator: ";" }), [["a", "b", "c"]], ); }, });
await t.step({ name: "MultiLine", fn() { const input = '"two\nline","one line","three\nline\nfield"'; assertEquals( parse(input), [["two\nline", "one line", "three\nline\nfield"]], ); }, });
await t.step({ name: "BlankLine", fn() { const input = "a,b,c\n\nd,e,f\n\n"; assertEquals( parse(input), [ ["a", "b", "c"], ["d", "e", "f"], ], ); }, });
await t.step({ name: "BlankLineFieldCount", fn() { const input = "a,b,c\n\nd,e,f\n\n"; assertEquals( parse(input, { fieldsPerRecord: 0 }), [ ["a", "b", "c"], ["d", "e", "f"], ], ); }, });
await t.step({ name: "TrimSpace", fn() { const input = " a, b, c\n"; assertEquals( parse(input, { trimLeadingSpace: true }), [["a", "b", "c"]], ); }, });
await t.step({ name: "LeadingSpace", fn() { const input = " a, b, c\n"; const output = [[" a", " b", " c"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "Comment", fn() { const input = "#1,2,3\na,b,c\n#comment"; const output = [["a", "b", "c"]]; assertEquals(parse(input, { comment: "#" }), output); }, }); await t.step({ name: "NoComment", fn() { const input = "#1,2,3\na,b,c"; const output = [ ["#1", "2", "3"], ["a", "b", "c"], ]; assertEquals(parse(input), output); }, }); await t.step({ name: "LazyQuotes", fn() { const input = `a "word","1"2",a","b`; const output = [[`a "word"`, `1"2`, `a"`, `b`]]; assertEquals(parse(input, { lazyQuotes: true }), output); }, }); await t.step({ name: "BareQuotes", fn() { const input = `a "word","1"2",a"`; const output = [[`a "word"`, `1"2`, `a"`]]; assertEquals(parse(input, { lazyQuotes: true }), output); }, }); await t.step({ name: "BareDoubleQuotes", fn() { const input = `a""b,c`; const output = [[`a""b`, `c`]]; assertEquals(parse(input, { lazyQuotes: true }), output); }, }); await t.step({ name: "BadDoubleQuotes", fn() { const input = `a""b,c`; assertThrows( () => parse(input), ParseError, 'parse error on line 1, column 1: bare " in non-quoted-field', ); }, }); await t.step({ name: "TrimQuote", fn() { const input = ` "a"," b",c`; const output = [["a", " b", "c"]]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ name: "BadBareQuote", fn() { const input = `a "word","b"`; assertThrows( () => parse(input), ParseError, 'parse error on line 1, column 2: bare " in non-quoted-field', ); }, }); await t.step({ name: "BadTrailingQuote", fn() { const input = `"a word",b"`; assertThrows( () => parse(input), ParseError, 'parse error on line 1, column 10: bare " in non-quoted-field', ); }, }); await t.step({ name: "ExtraneousQuote", fn() { const input = `"a "word","b"`; assertThrows( () => parse(input), ParseError, `parse error on line 1, column 3: extraneous or missing " in quoted-field`, ); }, }); await t.step({ name: "BadFieldCount", fn() { const input = "a,b,c\nd,e"; assertThrows( () => parse(input, { fieldsPerRecord: 0 }), ParseError, "record on line 2: wrong number of fields", ); }, }); await t.step({ name: "BadFieldCount1", fn() { const input = `a,b,c`; assertThrows( () => parse(input, { fieldsPerRecord: 2 }), ParseError, "record on line 1: wrong number of fields", ); }, }); await t.step({ name: "FieldCount", fn() { const input = "a,b,c\nd,e"; const output = [ ["a", "b", "c"], ["d", "e"], ]; assertEquals(parse(input), output); }, }); await t.step({ name: "TrailingCommaEOF", fn() { const input = "a,b,c,"; const output = [["a", "b", "c", ""]]; assertEquals(parse(input), output); }, }); await t.step({ name: "TrailingCommaEOL", fn() { const input = "a,b,c,\n"; const output = [["a", "b", "c", ""]]; assertEquals(parse(input), output); }, }); await t.step({ name: "TrailingCommaSpaceEOF", fn() { const input = "a,b,c, "; const output = [["a", "b", "c", ""]]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ name: "TrailingCommaSpaceEOL", fn() { const input = "a,b,c, \n"; const output = [["a", "b", "c", ""]]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ name: "TrailingCommaLine3", fn() { const input = "a,b,c\nd,e,f\ng,hi,"; const output = [ ["a", "b", "c"], ["d", "e", "f"], ["g", "hi", ""], ]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ name: "NotTrailingComma3", fn() { const input = "a,b,c, \n"; const output = [["a", "b", "c", " "]]; assertEquals(parse(input), output); }, }); await t.step({ name: "CommaFieldTest", fn() { const input = `x,y,z,w\nx,y,z,\nx,y,,\nx,,,\n,,,\n"x","y","z","w"\n"x","y","z",""\n"x","y","",""\n"x","","",""\n"","","",""\n`; const output = [ ["x", "y", "z", "w"], ["x", "y", "z", ""], ["x", "y", "", ""], ["x", "", "", ""], ["", "", "", ""], ["x", "y", "z", "w"], ["x", "y", "z", ""], ["x", "y", "", ""], ["x", "", "", ""], ["", "", "", ""], ]; assertEquals(parse(input), output); }, }); await t.step({ name: "TrailingCommaIneffective1", fn() { const input = "a,b,\nc,d,e"; const output = [ ["a", "b", ""], ["c", "d", "e"], ]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ name: "ReadAllReuseRecord", fn() { const input = "a,b\nc,d"; const output = [ ["a", "b"], ["c", "d"], ]; assertEquals(parse(input), output); // ReuseRecord: true, }, }); await t.step({ name: "StartLine1", // Issue 19019 fn() { const input = 'a,"b\nc"d,e'; assertThrows( () => parse(input, { fieldsPerRecord: 2 }), ParseError, 'record on line 1; parse error on line 2, column 1: extraneous or missing " in quoted-field', ); }, }); await t.step({ name: "StartLine2", fn() { const input = 'a,b\n"d\n\n,e'; assertThrows( () => parse(input, { fieldsPerRecord: 2 }), ParseError, 'record on line 2; parse error on line 5, column 0: extraneous or missing " in quoted-field', ); }, }); await t.step({ name: "CRLFInQuotedField", // Issue 21201 fn() { const input = 'A,"Hello\r\nHi",B\r\n'; const output = [["A", "Hello\nHi", "B"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "BinaryBlobField", // Issue 19410 fn() { const input = "x09\x41\xb4\x1c,aktau"; const output = [["x09A\xb4\x1c", "aktau"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "TrailingCR", fn() { const input = "field1,field2\r"; const output = [["field1", "field2"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "QuotedTrailingCR", fn() { const input = '"field"\r'; const output = [["field"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "QuotedTrailingCRCR", fn() { const input = '"field"\r\r'; assertThrows( () => parse(input, { fieldsPerRecord: 2 }), ParseError, 'parse error on line 1, column 6: extraneous or missing " in quoted-field', ); }, }); await t.step({ name: "FieldCR", fn() { const input = "field\rfield\r"; const output = [["field\rfield"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "FieldCRCR", fn() { const input = "field\r\rfield\r\r"; const output = [["field\r\rfield\r"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "FieldCRCRLF", fn() { const input = "field\r\r\nfield\r\r\n"; const output = [["field\r"], ["field\r"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "FieldCRCRLFCR", fn() { const input = "field\r\r\n\rfield\r\r\n\r"; const output = [["field\r"], ["\rfield\r"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "FieldCRCRLFCRCR", fn() { const input = "field\r\r\n\r\rfield\r\r\n\r\r"; const output = [["field\r"], ["\r\rfield\r"], ["\r"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "MultiFieldCRCRLFCRCR", fn() { const input = "field1,field2\r\r\n\r\rfield1,field2\r\r\n\r\r,"; const output = [ ["field1", "field2\r"], ["\r\rfield1", "field2\r"], ["\r\r", ""], ]; assertEquals(parse(input), output); }, }); await t.step({ name: "NonASCIICommaAndComment", fn() { const input = "a£b,c£ \td,e\n€ comment\n"; const output = [["a", "b,c", "d,e"]]; assertEquals( parse(input, { trimLeadingSpace: true, separator: "£", comment: "€", }), output, ); }, }); await t.step({ name: "NonASCIICommaAndCommentWithQuotes", fn() { const input = 'a€" b,"€ c\nλ comment\n'; const output = [["a", " b,", " c"]]; assertEquals( parse(input, { separator: "€", comment: "λ" }), output, ); }, }); await t.step( { // λ and θ start with the same byte. // This tests that the parser doesn't confuse such characters. name: "NonASCIICommaConfusion", fn() { const input = '"abθcd"λefθgh'; const output = [["abθcd", "efθgh"]]; assertEquals( parse(input, { separator: "λ", comment: "€" }), output, ); }, }, ); await t.step({ name: "NonASCIICommentConfusion", fn() { const input = "λ\nλ\nθ\nλ\n"; const output = [["λ"], ["λ"], ["λ"]]; assertEquals(parse(input, { comment: "θ" }), output); }, }); await t.step({ name: "QuotedFieldMultipleLF", fn() { const input = '"\n\n\n\n"'; const output = [["\n\n\n\n"]]; assertEquals(parse(input), output); }, }); await t.step({ name: "MultipleCRLF", fn() { const input = "\r\n\r\n\r\n\r\n"; const output: string[][] = []; assertEquals(parse(input), output); }, /** * The implementation may read each line in several chunks if * it doesn't fit entirely. * in the read buffer, so we should test the code to handle that condition. */ } /* TODO(kt3k): Enable this test case) await t.step({ name: "HugeLines", fn() { const input = "#ignore\n".repeat(10000) + "@".repeat(5000) + "," "*".repeat(5000), const output = [["@".repeat(5000), "*".repeat(5000)]] assertEquals(parse(input), output) Comment: "#", }, }*/); await t.step({ name: "QuoteWithTrailingCRLF", fn() { const input = '"foo"bar"\r\n'; assertThrows( () => parse(input), ParseError, `parse error on line 1, column 4: extraneous or missing " in quoted-field`, ); }, }); await t.step({ name: "LazyQuoteWithTrailingCRLF", fn() { const input = '"foo"bar"\r\n'; const output = [[`foo"bar`]]; assertEquals(parse(input, { lazyQuotes: true }), output); }, }); await t.step({ name: "DoubleQuoteWithTrailingCRLF", fn() { const input = '"foo""bar"\r\n'; const output = [[`foo"bar`]]; assertEquals(parse(input), output); }, }); await t.step({ name: "EvenQuotes", fn() { const input = `""""""""`; const output = [[`"""`]]; assertEquals(parse(input), output); }, }); await t.step({ name: "OddQuotes", fn() { const input = `"""""""`; assertThrows( () => parse(input), ParseError, `parse error on line 1, column 7: extraneous or missing " in quoted-field`, ); }, }); await t.step({ name: "LazyOddQuotes", fn() { const input = `"""""""`; const output = [[`"""`]]; assertEquals(parse(input, { lazyQuotes: true }), output); }, }); await t.step({ name: "BadComma1", fn() { const input = ""; assertThrows( () => parse(input, { separator: "\n" }), Error, "Invalid Delimiter", ); }, }); await t.step({ name: "BadComma2", fn() { const input = ""; assertThrows( () => parse(input, { separator: "\r" }), Error, "Invalid Delimiter", ); }, }); await t.step({ name: "BadComma3", fn() { const input = ""; assertThrows( () => parse(input, { separator: '"' }), Error, "Invalid Delimiter", ); }, }); await t.step({ name: "BadComment1", fn() { const input = ""; assertThrows( () => parse(input, { comment: "\n" }), Error, "Invalid Delimiter", ); }, }); await t.step({ name: "BadComment2", fn() { const input = ""; assertThrows( () => parse(input, { comment: "\r" }), Error, "Invalid Delimiter", ); }, }); await t.step({ name: "BadCommaComment", fn() { const input = ""; assertThrows( () => parse(input, { separator: "X", comment: "X" }), Error, "Invalid Delimiter", ); }, });
await t.step({ name: "simple", fn() { const input = "a,b,c"; const output = [["a", "b", "c"]]; assertEquals(parse(input, { skipFirstRow: false }), output); }, }); await t.step({ name: "simple Bufreader", fn() { const input = "a,b,c"; const output = [["a", "b", "c"]]; assertEquals(parse(input, { skipFirstRow: false }), output); }, }); await t.step({ name: "multiline", fn() { const input = "a,b,c\ne,f,g\n"; const output = [ ["a", "b", "c"], ["e", "f", "g"], ]; assertEquals(parse(input, { skipFirstRow: false }), output); }, }); await t.step({ name: "header mapping boolean", fn() { const input = "a,b,c\ne,f,g\n"; const output = [{ a: "e", b: "f", c: "g" }]; assertEquals(parse(input, { skipFirstRow: true }), output); }, }); await t.step({ name: "header mapping array", fn() { const input = "a,b,c\ne,f,g\n"; const output = [ { this: "a", is: "b", sparta: "c" }, { this: "e", is: "f", sparta: "g" }, ]; assertEquals( parse(input, { columns: ["this", "is", "sparta"] }), output, ); }, });
await t.step({ name: "provides both opts.skipFirstRow and opts.columns", fn() { const input = "a,b,1\nc,d,2\ne,f,3"; const output = [ { foo: "c", bar: "d", baz: "2" }, { foo: "e", bar: "f", baz: "3" }, ]; assertEquals( parse(input, { skipFirstRow: true, columns: ["foo", "bar", "baz"], }), output, ); }, }); await t.step({ name: "mismatching number of headers and fields", fn() { const input = "a,b,c\nd,e"; assertThrows( () => parse(input, { skipFirstRow: true, columns: ["foo", "bar", "baz"], }), Error, "Error number of fields line: 1\nNumber of fields found: 3\nExpected number of fields: 2", ); }, }); await t.step({ name: "Strips leading byte-order mark with bare cell", fn() { const input = `${BYTE_ORDER_MARK}abc`; const output = [["abc"]]; assert(!JSON.stringify(output).includes(BYTE_ORDER_MARK)); assertEquals(parse(input), output); }, }); await t.step({ name: "Strips leading byte-order mark with quoted cell", fn() { const input = `${BYTE_ORDER_MARK}"a""b"`; const output = [['a"b']]; assert(!JSON.stringify(output).includes(BYTE_ORDER_MARK)); assertEquals(parse(input), output); }, }); await t.step({ name: "Does not strip byte-order mark after position [0]", fn() { const input = `a${BYTE_ORDER_MARK}bc`; const output = [[`a${BYTE_ORDER_MARK}bc`]]; assertEquals(parse(input), output); }, }); await t.step({ name: "trimLeadingSpace strips leading byte-order mark followed by whitespace", fn() { const input = `${BYTE_ORDER_MARK} abc`; const output = [["abc"]]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); await t.step({ // This behavior is due to String#trimStart including U+FEFF in the set of // characters to be trimmed name: "trimLeadingSpace strips leading whitespace followed by byte-order mark", fn() { const input = ` ${BYTE_ORDER_MARK}abc`; const output = [["abc"]]; assertEquals(parse(input, { trimLeadingSpace: true }), output); }, }); },});
Deno.test({ name: "parse() is correctly typed", fn() { // If no option is passed, defaults to string[][] { const parsed = parse("a\nb"); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { const parsed = parse("a\nb"); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { // `skipFirstRow` may be `true` or `false`. // `columns` may be `undefined` or `string[]`. // If you don't know exactly what the value of the option is, // the return type is string[][] | Record<string, string|undefined>[] const options: ParseOptions = {}; const parsed = parse("a\nb", options); type _ = AssertTrue< IsExact< typeof parsed, string[][] | Record<string, string | undefined>[] > >; } { const parsed = parse("a\nb", {}); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; }
// skipFirstRow option { const parsed = parse("a\nb", { skipFirstRow: undefined }); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { const parsed = parse("a\nb", { skipFirstRow: false }); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { const parsed = parse("a\nb", { skipFirstRow: true }); type _ = AssertTrue< IsExact<typeof parsed, Record<string, string | undefined>[]> >; }
// columns option { const parsed = parse("a\nb", { columns: undefined }); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { const parsed = parse("a,b\nc,d", { columns: ["aaa", "bbb"] }); type _ = AssertTrue< IsExact<typeof parsed, Record<"aaa" | "bbb", string>[]> >; } { const parsed = parse("a\nb", { columns: ["aaa"] as string[] }); type _ = AssertTrue< IsExact<typeof parsed, Record<string, string | undefined>[]> >; }
// skipFirstRow option + columns option { const parsed = parse("a\nb", { skipFirstRow: false, columns: undefined }); type _ = AssertTrue<IsExact<typeof parsed, string[][]>>; } { const parsed = parse("a\nb", { skipFirstRow: true, columns: undefined }); type _ = AssertTrue< IsExact<typeof parsed, Record<string, string | undefined>[]> >; } { const parsed = parse("a\nb", { skipFirstRow: false, columns: ["aaa"] }); type _ = AssertTrue< IsExact<typeof parsed, Record<"aaa", string>[]> >; } { const parsed = parse("a\nb", { skipFirstRow: true, columns: ["aaa"] }); type _ = AssertTrue< IsExact<typeof parsed, Record<"aaa", string>[]> >; } },});
std
Version Info
6 months ago
