1// Copyright 2015 The Go Authors. All rights reserved.
2// Use of this source code is governed by a BSD-style
3// license that can be found in the LICENSE file.
4
5package mime
6
7import (
8	"errors"
9	"io"
10	"strings"
11	"testing"
12)
13
14func TestEncodeWord(t *testing.T) {
15	utf8, iso88591 := "utf-8", "iso-8859-1"
16	tests := []struct {
17		enc      WordEncoder
18		charset  string
19		src, exp string
20	}{
21		{QEncoding, utf8, "François-Jérôme", "=?utf-8?q?Fran=C3=A7ois-J=C3=A9r=C3=B4me?="},
22		{BEncoding, utf8, "Café", "=?utf-8?b?Q2Fmw6k=?="},
23		{QEncoding, iso88591, "La Seleção", "=?iso-8859-1?q?La_Sele=C3=A7=C3=A3o?="},
24		{QEncoding, utf8, "", ""},
25		{QEncoding, utf8, "A", "A"},
26		{QEncoding, iso88591, "a", "a"},
27		{QEncoding, utf8, "123 456", "123 456"},
28		{QEncoding, utf8, "\t !\"#$%&'()*+,-./ :;<>?@[\\]^_`{|}~", "\t !\"#$%&'()*+,-./ :;<>?@[\\]^_`{|}~"},
29		{QEncoding, utf8, strings.Repeat("é", 10), "=?utf-8?q?" + strings.Repeat("=C3=A9", 10) + "?="},
30		{QEncoding, utf8, strings.Repeat("é", 11), "=?utf-8?q?" + strings.Repeat("=C3=A9", 10) + "?= =?utf-8?q?=C3=A9?="},
31		{QEncoding, iso88591, strings.Repeat("\xe9", 22), "=?iso-8859-1?q?" + strings.Repeat("=E9", 22) + "?="},
32		{QEncoding, utf8, strings.Repeat("\x80", 22), "=?utf-8?q?" + strings.Repeat("=80", 21) + "?= =?utf-8?q?=80?="},
33		{BEncoding, iso88591, strings.Repeat("\xe9", 45), "=?iso-8859-1?b?" + strings.Repeat("6enp", 15) + "?="},
34		{BEncoding, utf8, strings.Repeat("\x80", 48), "=?utf-8?b?" + strings.Repeat("gICA", 15) + "?= =?utf-8?b?gICA?="},
35	}
36
37	for _, test := range tests {
38		if s := test.enc.Encode(test.charset, test.src); s != test.exp {
39			t.Errorf("Encode(%q) = %q, want %q", test.src, s, test.exp)
40		}
41	}
42}
43
44func TestEncodedWordLength(t *testing.T) {
45	tests := []struct {
46		enc WordEncoder
47		src string
48	}{
49		{QEncoding, strings.Repeat("à", 30)},
50		{QEncoding, strings.Repeat("é", 60)},
51		{BEncoding, strings.Repeat("ï", 25)},
52		{BEncoding, strings.Repeat("ô", 37)},
53		{BEncoding, strings.Repeat("\x80", 50)},
54		{QEncoding, "{$firstname} Bienvendio a Apostolica, aquà inicia el camino de tu"},
55	}
56
57	for _, test := range tests {
58		s := test.enc.Encode("utf-8", test.src)
59		wordLen := 0
60		for i := 0; i < len(s); i++ {
61			if s[i] == ' ' {
62				wordLen = 0
63				continue
64			}
65
66			wordLen++
67			if wordLen > maxEncodedWordLen {
68				t.Errorf("Encode(%q) has more than %d characters: %q",
69					test.src, maxEncodedWordLen, s)
70			}
71		}
72	}
73}
74
75func TestDecodeWord(t *testing.T) {
76	tests := []struct {
77		src, exp string
78		hasErr   bool
79	}{
80		{"=?UTF-8?Q?=C2=A1Hola,_se=C3=B1or!?=", "¡Hola, señor!", false},
81		{"=?UTF-8?Q?Fran=C3=A7ois-J=C3=A9r=C3=B4me?=", "François-Jérôme", false},
82		{"=?UTF-8?q?ascii?=", "ascii", false},
83		{"=?utf-8?B?QW5kcsOp?=", "André", false},
84		{"=?ISO-8859-1?Q?Rapha=EBl_Dupont?=", "Raphaël Dupont", false},
85		{"=?utf-8?b?IkFudG9uaW8gSm9zw6kiIDxqb3NlQGV4YW1wbGUub3JnPg==?=", `"Antonio José" <jose@example.org>`, false},
86		{"=?UTF-8?A?Test?=", "", true},
87		{"=?UTF-8?Q?A=B?=", "", true},
88		{"=?UTF-8?Q?=A?=", "", true},
89		{"=?UTF-8?A?A?=", "", true},
90		{"=????=", "", true},
91		{"=?UTF-8???=", "", true},
92		{"=?UTF-8?Q??=", "", false},
93	}
94
95	for _, test := range tests {
96		dec := new(WordDecoder)
97		s, err := dec.Decode(test.src)
98		if test.hasErr && err == nil {
99			t.Errorf("Decode(%q) should return an error", test.src)
100			continue
101		}
102		if !test.hasErr && err != nil {
103			t.Errorf("Decode(%q): %v", test.src, err)
104			continue
105		}
106		if s != test.exp {
107			t.Errorf("Decode(%q) = %q, want %q", test.src, s, test.exp)
108		}
109	}
110}
111
112func TestDecodeHeader(t *testing.T) {
113	tests := []struct {
114		src, exp string
115	}{
116		{"=?UTF-8?Q?=C2=A1Hola,_se=C3=B1or!?=", "¡Hola, señor!"},
117		{"=?UTF-8?Q?Fran=C3=A7ois-J=C3=A9r=C3=B4me?=", "François-Jérôme"},
118		{"=?UTF-8?q?ascii?=", "ascii"},
119		{"=?utf-8?B?QW5kcsOp?=", "André"},
120		{"=?ISO-8859-1?Q?Rapha=EBl_Dupont?=", "Raphaël Dupont"},
121		{"Jean", "Jean"},
122		{"=?utf-8?b?IkFudG9uaW8gSm9zw6kiIDxqb3NlQGV4YW1wbGUub3JnPg==?=", `"Antonio José" <jose@example.org>`},
123		{"=?UTF-8?A?Test?=", "=?UTF-8?A?Test?="},
124		{"=?UTF-8?Q?A=B?=", "=?UTF-8?Q?A=B?="},
125		{"=?UTF-8?Q?=A?=", "=?UTF-8?Q?=A?="},
126		{"=?UTF-8?A?A?=", "=?UTF-8?A?A?="},
127		// Incomplete words
128		{"=?", "=?"},
129		{"=?UTF-8?", "=?UTF-8?"},
130		{"=?UTF-8?=", "=?UTF-8?="},
131		{"=?UTF-8?Q", "=?UTF-8?Q"},
132		{"=?UTF-8?Q?", "=?UTF-8?Q?"},
133		{"=?UTF-8?Q?=", "=?UTF-8?Q?="},
134		{"=?UTF-8?Q?A", "=?UTF-8?Q?A"},
135		{"=?UTF-8?Q?A?", "=?UTF-8?Q?A?"},
136		// Tests from RFC 2047
137		{"=?ISO-8859-1?Q?a?=", "a"},
138		{"=?ISO-8859-1?Q?a?= b", "a b"},
139		{"=?ISO-8859-1?Q?a?= =?ISO-8859-1?Q?b?=", "ab"},
140		{"=?ISO-8859-1?Q?a?=  =?ISO-8859-1?Q?b?=", "ab"},
141		{"=?ISO-8859-1?Q?a?= \r\n\t =?ISO-8859-1?Q?b?=", "ab"},
142		{"=?ISO-8859-1?Q?a_b?=", "a b"},
143	}
144
145	for _, test := range tests {
146		dec := new(WordDecoder)
147		s, err := dec.DecodeHeader(test.src)
148		if err != nil {
149			t.Errorf("DecodeHeader(%q): %v", test.src, err)
150		}
151		if s != test.exp {
152			t.Errorf("DecodeHeader(%q) = %q, want %q", test.src, s, test.exp)
153		}
154	}
155}
156
157func TestCharsetDecoder(t *testing.T) {
158	tests := []struct {
159		src      string
160		want     string
161		charsets []string
162		content  []string
163	}{
164		{"=?utf-8?b?Q2Fmw6k=?=", "Café", nil, nil},
165		{"=?ISO-8859-1?Q?caf=E9?=", "café", nil, nil},
166		{"=?US-ASCII?Q?foo_bar?=", "foo bar", nil, nil},
167		{"=?utf-8?Q?=?=", "=?utf-8?Q?=?=", nil, nil},
168		{"=?utf-8?Q?=A?=", "=?utf-8?Q?=A?=", nil, nil},
169		{
170			"=?ISO-8859-15?Q?f=F5=F6?=  =?windows-1252?Q?b=E0r?=",
171			"f\xf5\xf6b\xe0r",
172			[]string{"iso-8859-15", "windows-1252"},
173			[]string{"f\xf5\xf6", "b\xe0r"},
174		},
175	}
176
177	for _, test := range tests {
178		i := 0
179		dec := &WordDecoder{
180			CharsetReader: func(charset string, input io.Reader) (io.Reader, error) {
181				if charset != test.charsets[i] {
182					t.Errorf("DecodeHeader(%q), got charset %q, want %q", test.src, charset, test.charsets[i])
183				}
184				content, err := io.ReadAll(input)
185				if err != nil {
186					t.Errorf("DecodeHeader(%q), error in reader: %v", test.src, err)
187				}
188				got := string(content)
189				if got != test.content[i] {
190					t.Errorf("DecodeHeader(%q), got content %q, want %q", test.src, got, test.content[i])
191				}
192				i++
193
194				return strings.NewReader(got), nil
195			},
196		}
197		got, err := dec.DecodeHeader(test.src)
198		if err != nil {
199			t.Errorf("DecodeHeader(%q): %v", test.src, err)
200		}
201		if got != test.want {
202			t.Errorf("DecodeHeader(%q) = %q, want %q", test.src, got, test.want)
203		}
204	}
205}
206
207func TestCharsetDecoderError(t *testing.T) {
208	dec := &WordDecoder{
209		CharsetReader: func(charset string, input io.Reader) (io.Reader, error) {
210			return nil, errors.New("Test error")
211		},
212	}
213
214	if _, err := dec.DecodeHeader("=?charset?Q?foo?="); err == nil {
215		t.Error("DecodeHeader should return an error")
216	}
217}
218
219func BenchmarkQEncodeWord(b *testing.B) {
220	for i := 0; i < b.N; i++ {
221		QEncoding.Encode("UTF-8", "¡Hola, señor!")
222	}
223}
224
225func BenchmarkQDecodeWord(b *testing.B) {
226	dec := new(WordDecoder)
227
228	for i := 0; i < b.N; i++ {
229		dec.Decode("=?utf-8?q?=C2=A1Hola,_se=C3=B1or!?=")
230	}
231}
232
233func BenchmarkQDecodeHeader(b *testing.B) {
234	dec := new(WordDecoder)
235
236	for i := 0; i < b.N; i++ {
237		dec.DecodeHeader("=?utf-8?q?=C2=A1Hola,_se=C3=B1or!?=")
238	}
239}
240