forks/forgejo
1
0
Fork 0
Code Releases Activity
forgejo/modules/charset/charset_test.go
guillep2k 5a44be627c Convert files to utf-8 for indexing (#7814) 
* Convert files to utf-8 for indexing

* Move utf8 functions to modules/base

* Bump repoIndexerLatestVersion to 3

* Add tests for base/encoding.go

* Changes to pass gosimple

* Move UTF8 funcs into new modules/charset package
2019-08-15 20:07:28 +08:00

191 lines
7.5 KiB
Go
Raw Blame History

// Copyright 2019 The Gitea Authors. All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package charset
import (
"testing"
"code.gitea.io/gitea/modules/setting"
"github.com/stretchr/testify/assert"
)
func TestRemoveBOMIfPresent(t *testing.T) {
res := RemoveBOMIfPresent([]byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba}, res)
res = RemoveBOMIfPresent([]byte{0xef, 0xbb, 0xbf, 0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba}, res)
}
func TestToUTF8WithErr(t *testing.T) {
var res string
var err error
res, err = ToUTF8WithErr([]byte{0x41, 0x42, 0x43})
assert.Equal(t, "ABC", res)
assert.NoError(t, err)
res, err = ToUTF8WithErr([]byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, "áéíóú", res)
assert.NoError(t, err)
res, err = ToUTF8WithErr([]byte{0xef, 0xbb, 0xbf, 0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, "áéíóú", res)
assert.NoError(t, err)
// This test FAILS
res, err = ToUTF8WithErr([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0xF1, 0x6F, 0x73})
assert.Equal(t, "Hola, así cómo ños", res)
assert.NoError(t, err)
res, err = ToUTF8WithErr([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x07, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Regexp(t, "^Hola, así cómo", res)
assert.NoError(t, err)
res, err = ToUTF8WithErr([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x81, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Regexp(t, "^Hola, así cómo", res)
assert.NoError(t, err)
// Japanese (Shift-JIS)
res, err = ToUTF8WithErr([]byte{0x93, 0xFA, 0x91, 0xAE, 0x94, 0xE9, 0x82, 0xBC, 0x82, 0xB5, 0x82, 0xBF, 0x82, 0xE3, 0x81, 0x42})
assert.Equal(t, "日属秘ぞしちゅ。", res)
assert.NoError(t, err)
res, err = ToUTF8WithErr([]byte{0x00, 0x00, 0x00, 0x00})
assert.Equal(t, "\x00\x00\x00\x00", res)
assert.NoError(t, err)
}
func TestToUTF8WithFallback(t *testing.T) {
res := ToUTF8WithFallback([]byte{0x41, 0x42, 0x43})
assert.Equal(t, []byte("ABC"), res)
res = ToUTF8WithFallback([]byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte("áéíóú"), res)
res = ToUTF8WithFallback([]byte{0xef, 0xbb, 0xbf, 0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte("áéíóú"), res)
res = ToUTF8WithFallback([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0xF1, 0x6F, 0x73})
assert.Equal(t, []byte("Hola, así cómo ños"), res)
minmatch := []byte("Hola, así cómo ")
res = ToUTF8WithFallback([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x07, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Equal(t, minmatch, res[0:len(minmatch)])
res = ToUTF8WithFallback([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x81, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Equal(t, minmatch, res[0:len(minmatch)])
// Japanese (Shift-JIS)
res = ToUTF8WithFallback([]byte{0x93, 0xFA, 0x91, 0xAE, 0x94, 0xE9, 0x82, 0xBC, 0x82, 0xB5, 0x82, 0xBF, 0x82, 0xE3, 0x81, 0x42})
assert.Equal(t, []byte("日属秘ぞしちゅ。"), res)
res = ToUTF8WithFallback([]byte{0x00, 0x00, 0x00, 0x00})
assert.Equal(t, []byte{0x00, 0x00, 0x00, 0x00}, res)
}
func TestToUTF8(t *testing.T) {
res := ToUTF8("ABC")
assert.Equal(t, "ABC", res)
res = ToUTF8("áéíóú")
assert.Equal(t, "áéíóú", res)
// With utf-8 BOM
res = ToUTF8("\ufeffáéíóú")
assert.Equal(t, "áéíóú", res)
res = ToUTF8("Hola, así cómo ños")
assert.Equal(t, "Hola, así cómo ños", res)
res = ToUTF8("Hola, así cómo \x07ños")
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Regexp(t, "^Hola, así cómo", res)
// This test FAILS
// res = ToUTF8("Hola, así cómo \x81ños")
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
// assert.Regexp(t, "^Hola, así cómo", res)
// Japanese (Shift-JIS)
res = ToUTF8("\x93\xFA\x91\xAE\x94\xE9\x82\xBC\x82\xB5\x82\xBF\x82\xE3\x81\x42")
assert.Equal(t, "日属秘ぞしちゅ。", res)
res = ToUTF8("\x00\x00\x00\x00")
assert.Equal(t, "\x00\x00\x00\x00", res)
}
func TestToUTF8DropErrors(t *testing.T) {
res := ToUTF8DropErrors([]byte{0x41, 0x42, 0x43})
assert.Equal(t, []byte("ABC"), res)
res = ToUTF8DropErrors([]byte{0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte("áéíóú"), res)
res = ToUTF8DropErrors([]byte{0xef, 0xbb, 0xbf, 0xc3, 0xa1, 0xc3, 0xa9, 0xc3, 0xad, 0xc3, 0xb3, 0xc3, 0xba})
assert.Equal(t, []byte("áéíóú"), res)
res = ToUTF8DropErrors([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0xF1, 0x6F, 0x73})
assert.Equal(t, []byte("Hola, así cómo ños"), res)
minmatch := []byte("Hola, así cómo ")
res = ToUTF8DropErrors([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x07, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Equal(t, minmatch, res[0:len(minmatch)])
res = ToUTF8DropErrors([]byte{0x48, 0x6F, 0x6C, 0x61, 0x2C, 0x20, 0x61, 0x73, 0xED, 0x20, 0x63, 0xF3, 0x6D, 0x6F, 0x20, 0x81, 0xA4, 0x6F, 0x73})
// Do not fail for differences in invalid cases, as the library might change the conversion criteria for those
assert.Equal(t, minmatch, res[0:len(minmatch)])
// Japanese (Shift-JIS)
res = ToUTF8DropErrors([]byte{0x93, 0xFA, 0x91, 0xAE, 0x94, 0xE9, 0x82, 0xBC, 0x82, 0xB5, 0x82, 0xBF, 0x82, 0xE3, 0x81, 0x42})
assert.Equal(t, []byte("日属秘ぞしちゅ。"), res)
res = ToUTF8DropErrors([]byte{0x00, 0x00, 0x00, 0x00})
assert.Equal(t, []byte{0x00, 0x00, 0x00, 0x00}, res)
}
func TestDetectEncoding(t *testing.T) {
testSuccess := func(b []byte, expected string) {
encoding, err := DetectEncoding(b)
assert.NoError(t, err)
assert.Equal(t, expected, encoding)
}
// utf-8
b := []byte("just some ascii")
testSuccess(b, "UTF-8")
// utf-8-sig: "hey" (with BOM)
b = []byte{0xef, 0xbb, 0xbf, 0x68, 0x65, 0x79}
testSuccess(b, "UTF-8")
// utf-16: "hey<accented G>"
b = []byte{0xff, 0xfe, 0x68, 0x00, 0x65, 0x00, 0x79, 0x00, 0xf4, 0x01}
testSuccess(b, "UTF-16LE")
// iso-8859-1: d<accented e>cor<newline>
b = []byte{0x44, 0xe9, 0x63, 0x6f, 0x72, 0x0a}
encoding, err := DetectEncoding(b)
assert.NoError(t, err)
// due to a race condition in `chardet` library, it could either detect
// "ISO-8859-1" or "IS0-8859-2" here. Technically either is correct, so
// we accept either.
assert.Contains(t, encoding, "ISO-8859")
setting.Repository.AnsiCharset = "placeholder"
testSuccess(b, "placeholder")
// invalid bytes
b = []byte{0xfa}
_, err = DetectEncoding(b)
assert.Error(t, err)
}
View git blame Copy permalink