blob: ceb813ae8d5938cf05f4dc638c74beb0c0b41019 [file] [log] [blame]
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package bufio_test
import (
. "bufio"
"bytes"
"errors"
"io"
"strings"
"testing"
"unicode"
"unicode/utf8"
)
const smallMaxTokenSize = 256 // Much smaller for more efficient testing.
// Test white space table matches the Unicode definition.
func TestSpace(t *testing.T) {
for r := rune(0); r <= utf8.MaxRune; r++ {
if IsSpace(r) != unicode.IsSpace(r) {
t.Fatalf("white space property disagrees: %#U should be %t", r, unicode.IsSpace(r))
}
}
}
var scanTests = []string{
"",
"a",
"¼",
"☹",
"\x81", // UTF-8 error
"\uFFFD", // correctly encoded RuneError
"abcdefgh",
"abc def\n\t\tgh ",
"abc¼☹\x81\uFFFD日本語\x82abc",
}
func TestScanByte(t *testing.T) {
for n, test := range scanTests {
buf := strings.NewReader(test)
s := NewScanner(buf)
s.Split(ScanBytes)
var i int
for i = 0; s.Scan(); i++ {
if b := s.Bytes(); len(b) != 1 || b[0] != test[i] {
t.Errorf("#%d: %d: expected %q got %q", n, i, test, b)
}
}
if i != len(test) {
t.Errorf("#%d: termination expected at %d; got %d", n, len(test), i)
}
err := s.Err()
if err != nil {
t.Errorf("#%d: %v", n, err)
}
}
}
// Test that the rune splitter returns same sequence of runes (not bytes) as for range string.
func TestScanRune(t *testing.T) {
for n, test := range scanTests {
buf := strings.NewReader(test)
s := NewScanner(buf)
s.Split(ScanRunes)
var i, runeCount int
var expect rune
// Use a string range loop to validate the sequence of runes.
for i, expect = range string(test) {
if !s.Scan() {
break
}
runeCount++
got, _ := utf8.DecodeRune(s.Bytes())
if got != expect {
t.Errorf("#%d: %d: expected %q got %q", n, i, expect, got)
}
}
if s.Scan() {
t.Errorf("#%d: scan ran too long, got %q", n, s.Text())
}
testRuneCount := utf8.RuneCountInString(test)
if runeCount != testRuneCount {
t.Errorf("#%d: termination expected at %d; got %d", n, testRuneCount, runeCount)
}
err := s.Err()
if err != nil {
t.Errorf("#%d: %v", n, err)
}
}
}
var wordScanTests = []string{
"",
" ",
"\n",
"a",
" a ",
"abc def",
" abc def ",
" abc\tdef\nghi\rjkl\fmno\vpqr\u0085stu\u00a0\n",
}
// Test that the word splitter returns the same data as strings.Fields.
func TestScanWords(t *testing.T) {
for n, test := range wordScanTests {
buf := strings.NewReader(test)
s := NewScanner(buf)
s.Split(ScanWords)
words := strings.Fields(test)
var wordCount int
for wordCount = 0; wordCount < len(words); wordCount++ {
if !s.Scan() {
break
}
got := s.Text()
if got != words[wordCount] {
t.Errorf("#%d: %d: expected %q got %q", n, wordCount, words[wordCount], got)
}
}
if s.Scan() {
t.Errorf("#%d: scan ran too long, got %q", n, s.Text())
}
if wordCount != len(words) {
t.Errorf("#%d: termination expected at %d; got %d", n, len(words), wordCount)
}
err := s.Err()
if err != nil {
t.Errorf("#%d: %v", n, err)
}
}
}
// slowReader is a reader that returns only a few bytes at a time, to test the incremental
// reads in Scanner.Scan.
type slowReader struct {
max int
buf io.Reader
}
func (sr *slowReader) Read(p []byte) (n int, err error) {
if len(p) > sr.max {
p = p[0:sr.max]
}
return sr.buf.Read(p)
}
// genLine writes to buf a predictable but non-trivial line of text of length
// n, including the terminal newline and an occasional carriage return.
// If addNewline is false, the \r and \n are not emitted.
func genLine(buf *bytes.Buffer, lineNum, n int, addNewline bool) {
buf.Reset()
doCR := lineNum%5 == 0
if doCR {
n--
}
for i := 0; i < n-1; i++ { // Stop early for \n.
c := 'a' + byte(lineNum+i)
if c == '\n' || c == '\r' { // Don't confuse us.
c = 'N'
}
buf.WriteByte(c)
}
if addNewline {
if doCR {
buf.WriteByte('\r')
}
buf.WriteByte('\n')
}
}
// Test the line splitter, including some carriage returns but no long lines.
func TestScanLongLines(t *testing.T) {
// Build a buffer of lots of line lengths up to but not exceeding smallMaxTokenSize.
tmp := new(bytes.Buffer)
buf := new(bytes.Buffer)
lineNum := 0
j := 0
for i := 0; i < 2*smallMaxTokenSize; i++ {
genLine(tmp, lineNum, j, true)
if j < smallMaxTokenSize {
j++
} else {
j--
}
buf.Write(tmp.Bytes())
lineNum++
}
s := NewScanner(&slowReader{1, buf})
s.Split(ScanLines)
s.MaxTokenSize(smallMaxTokenSize)
j = 0
for lineNum := 0; s.Scan(); lineNum++ {
genLine(tmp, lineNum, j, false)
if j < smallMaxTokenSize {
j++
} else {
j--
}
line := tmp.String() // We use the string-valued token here, for variety.
if s.Text() != line {
t.Errorf("%d: bad line: %d %d\n%.100q\n%.100q\n", lineNum, len(s.Bytes()), len(line), s.Text(), line)
}
}
err := s.Err()
if err != nil {
t.Fatal(err)
}
}
// Test that the line splitter errors out on a long line.
func TestScanLineTooLong(t *testing.T) {
const smallMaxTokenSize = 256 // Much smaller for more efficient testing.
// Build a buffer of lots of line lengths up to but not exceeding smallMaxTokenSize.
tmp := new(bytes.Buffer)
buf := new(bytes.Buffer)
lineNum := 0
j := 0
for i := 0; i < 2*smallMaxTokenSize; i++ {
genLine(tmp, lineNum, j, true)
j++
buf.Write(tmp.Bytes())
lineNum++
}
s := NewScanner(&slowReader{3, buf})
s.Split(ScanLines)
s.MaxTokenSize(smallMaxTokenSize)
j = 0
for lineNum := 0; s.Scan(); lineNum++ {
genLine(tmp, lineNum, j, false)
if j < smallMaxTokenSize {
j++
} else {
j--
}
line := tmp.Bytes()
if !bytes.Equal(s.Bytes(), line) {
t.Errorf("%d: bad line: %d %d\n%.100q\n%.100q\n", lineNum, len(s.Bytes()), len(line), s.Bytes(), line)
}
}
err := s.Err()
if err != ErrTooLong {
t.Fatalf("expected ErrTooLong; got %s", err)
}
}
// Test that the line splitter handles a final line without a newline.
func testNoNewline(text string, lines []string, t *testing.T) {
buf := strings.NewReader(text)
s := NewScanner(&slowReader{7, buf})
s.Split(ScanLines)
for lineNum := 0; s.Scan(); lineNum++ {
line := lines[lineNum]
if s.Text() != line {
t.Errorf("%d: bad line: %d %d\n%.100q\n%.100q\n", lineNum, len(s.Bytes()), len(line), s.Bytes(), line)
}
}
err := s.Err()
if err != nil {
t.Fatal(err)
}
}
// Test that the line splitter handles a final line without a newline.
func TestScanLineNoNewline(t *testing.T) {
const text = "abcdefghijklmn\nopqrstuvwxyz"
lines := []string{
"abcdefghijklmn",
"opqrstuvwxyz",
}
testNoNewline(text, lines, t)
}
// Test that the line splitter handles a final line with a carriage return but no newline.
func TestScanLineReturnButNoNewline(t *testing.T) {
const text = "abcdefghijklmn\nopqrstuvwxyz\r"
lines := []string{
"abcdefghijklmn",
"opqrstuvwxyz",
}
testNoNewline(text, lines, t)
}
// Test that the line splitter handles a final empty line.
func TestScanLineEmptyFinalLine(t *testing.T) {
const text = "abcdefghijklmn\nopqrstuvwxyz\n\n"
lines := []string{
"abcdefghijklmn",
"opqrstuvwxyz",
"",
}
testNoNewline(text, lines, t)
}
// Test that the line splitter handles a final empty line with a carriage return but no newline.
func TestScanLineEmptyFinalLineWithCR(t *testing.T) {
const text = "abcdefghijklmn\nopqrstuvwxyz\n\r"
lines := []string{
"abcdefghijklmn",
"opqrstuvwxyz",
"",
}
testNoNewline(text, lines, t)
}
var testError = errors.New("testError")
// Test the correct error is returned when the split function errors out.
func TestSplitError(t *testing.T) {
// Create a split function that delivers a little data, then a predictable error.
numSplits := 0
const okCount = 7
errorSplit := func(data []byte, atEOF bool) (advance int, token []byte, err error) {
if atEOF {
panic("didn't get enough data")
}
if numSplits >= okCount {
return 0, nil, testError
}
numSplits++
return 1, data[0:1], nil
}
// Read the data.
const text = "abcdefghijklmnopqrstuvwxyz"
buf := strings.NewReader(text)
s := NewScanner(&slowReader{1, buf})
s.Split(errorSplit)
var i int
for i = 0; s.Scan(); i++ {
if len(s.Bytes()) != 1 || text[i] != s.Bytes()[0] {
t.Errorf("#%d: expected %q got %q", i, text[i], s.Bytes()[0])
}
}
// Check correct termination location and error.
if i != okCount {
t.Errorf("unexpected termination; expected %d tokens got %d", okCount, i)
}
err := s.Err()
if err != testError {
t.Fatalf("expected %q got %v", testError, err)
}
}
// Test that an EOF is overridden by a user-generated scan error.
func TestErrAtEOF(t *testing.T) {
s := NewScanner(strings.NewReader("1 2 33"))
// This splitter will fail on last entry, after s.err==EOF.
split := func(data []byte, atEOF bool) (advance int, token []byte, err error) {
advance, token, err = ScanWords(data, atEOF)
if len(token) > 1 {
if s.ErrOrEOF() != io.EOF {
t.Fatal("not testing EOF")
}
err = testError
}
return
}
s.Split(split)
for s.Scan() {
}
if s.Err() != testError {
t.Fatal("wrong error:", s.Err())
}
}
// Test for issue 5268.
type alwaysError struct{}
func (alwaysError) Read(p []byte) (int, error) {
return 0, io.ErrUnexpectedEOF
}
func TestNonEOFWithEmptyRead(t *testing.T) {
scanner := NewScanner(alwaysError{})
for scanner.Scan() {
t.Fatal("read should fail")
}
err := scanner.Err()
if err != io.ErrUnexpectedEOF {
t.Errorf("unexpected error: %v", err)
}
}
// Test that Scan finishes if we have endless empty reads.
type endlessZeros struct{}
func (endlessZeros) Read(p []byte) (int, error) {
return 0, nil
}
func TestBadReader(t *testing.T) {
scanner := NewScanner(endlessZeros{})
for scanner.Scan() {
t.Fatal("read should fail")
}
err := scanner.Err()
if err != io.ErrNoProgress {
t.Errorf("unexpected error: %v", err)
}
}
func TestScanWordsExcessiveWhiteSpace(t *testing.T) {
const word = "ipsum"
s := strings.Repeat(" ", 4*smallMaxTokenSize) + word
scanner := NewScanner(strings.NewReader(s))
scanner.MaxTokenSize(smallMaxTokenSize)
scanner.Split(ScanWords)
if !scanner.Scan() {
t.Fatalf("scan failed: %v", scanner.Err())
}
if token := scanner.Text(); token != word {
t.Fatalf("unexpected token: %v", token)
}
}
// Test that empty tokens, including at end of line or end of file, are found by the scanner.
// Issue 8672: Could miss final empty token.
func commaSplit(data []byte, atEOF bool) (advance int, token []byte, err error) {
for i := 0; i < len(data); i++ {
if data[i] == ',' {
return i + 1, data[:i], nil
}
}
return 0, data, ErrFinalToken
}
func testEmptyTokens(t *testing.T, text string, values []string) {
s := NewScanner(strings.NewReader(text))
s.Split(commaSplit)
var i int
for i = 0; s.Scan(); i++ {
if i >= len(values) {
t.Fatalf("got %d fields, expected %d", i+1, len(values))
}
if s.Text() != values[i] {
t.Errorf("%d: expected %q got %q", i, values[i], s.Text())
}
}
if i != len(values) {
t.Fatalf("got %d fields, expected %d", i, len(values))
}
if err := s.Err(); err != nil {
t.Fatal(err)
}
}
func TestEmptyTokens(t *testing.T) {
testEmptyTokens(t, "1,2,3,", []string{"1", "2", "3", ""})
}
func TestWithNoEmptyTokens(t *testing.T) {
testEmptyTokens(t, "1,2,3", []string{"1", "2", "3"})
}
func loopAtEOFSplit(data []byte, atEOF bool) (advance int, token []byte, err error) {
if len(data) > 0 {
return 1, data[:1], nil
}
return 0, data, nil
}
func TestDontLoopForever(t *testing.T) {
s := NewScanner(strings.NewReader("abc"))
s.Split(loopAtEOFSplit)
// Expect a panic
defer func() {
err := recover()
if err == nil {
t.Fatal("should have panicked")
}
if msg, ok := err.(string); !ok || !strings.Contains(msg, "empty tokens") {
panic(err)
}
}()
for count := 0; s.Scan(); count++ {
if count > 1000 {
t.Fatal("looping")
}
}
if s.Err() != nil {
t.Fatal("after scan:", s.Err())
}
}
func TestBlankLines(t *testing.T) {
s := NewScanner(strings.NewReader(strings.Repeat("\n", 1000)))
for count := 0; s.Scan(); count++ {
if count > 2000 {
t.Fatal("looping")
}
}
if s.Err() != nil {
t.Fatal("after scan:", s.Err())
}
}
type countdown int
func (c *countdown) split(data []byte, atEOF bool) (advance int, token []byte, err error) {
if *c > 0 {
*c--
return 1, data[:1], nil
}
return 0, nil, nil
}
// Check that the looping-at-EOF check doesn't trigger for merely empty tokens.
func TestEmptyLinesOK(t *testing.T) {
c := countdown(10000)
s := NewScanner(strings.NewReader(strings.Repeat("\n", 10000)))
s.Split(c.split)
for s.Scan() {
}
if s.Err() != nil {
t.Fatal("after scan:", s.Err())
}
if c != 0 {
t.Fatalf("stopped with %d left to process", c)
}
}
// Make sure we can read a huge token if a big enough buffer is provided.
func TestHugeBuffer(t *testing.T) {
text := strings.Repeat("x", 2*MaxScanTokenSize)
s := NewScanner(strings.NewReader(text + "\n"))
s.Buffer(make([]byte, 100), 3*MaxScanTokenSize)
for s.Scan() {
token := s.Text()
if token != text {
t.Errorf("scan got incorrect token of length %d", len(token))
}
}
if s.Err() != nil {
t.Fatal("after scan:", s.Err())
}
}
// negativeEOFReader returns an invalid -1 at the end, as though it
// were wrapping the read system call.
type negativeEOFReader int
func (r *negativeEOFReader) Read(p []byte) (int, error) {
if *r > 0 {
c := int(*r)
if c > len(p) {
c = len(p)
}
for i := 0; i < c; i++ {
p[i] = 'a'
}
p[c-1] = '\n'
*r -= negativeEOFReader(c)
return c, nil
}
return -1, io.EOF
}
// Test that the scanner doesn't panic on a reader that returns a
// negative count of bytes read (issue 38053).
func TestNegativeEOFReader(t *testing.T) {
r := negativeEOFReader(10)
scanner := NewScanner(&r)
c := 0
for scanner.Scan() {
c++
if c > 1 {
t.Error("read too many lines")
break
}
}
if scanner.Err() == nil {
t.Error("scanner.Err returned nil, expected an error")
}
}
// largeReader returns an invalid count that is larger than the number
// of bytes requested.
type largeReader struct{}
func (largeReader) Read(p []byte) (int, error) {
return len(p) + 1, nil
}
func TestLargeReader(t *testing.T) {
scanner := NewScanner(largeReader{})
for scanner.Scan() {
}
if scanner.Err() == nil {
t.Error("scanner.Err returned nil, expected an error")
}
}