1 /*
2 * Copyright (c) 2002-2026 Gargoyle Software Inc.
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 * https://www.apache.org/licenses/LICENSE-2.0
8 *
9 * Unless required by applicable law or agreed to in writing, software
10 * distributed under the License is distributed on an "AS IS" BASIS,
11 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 * See the License for the specific language governing permissions and
13 * limitations under the License.
14 */
15 package org.htmlunit.http;
16
17 import java.io.ByteArrayOutputStream;
18 import java.nio.charset.Charset;
19 import java.nio.charset.StandardCharsets;
20 import java.util.LinkedHashMap;
21 import java.util.Locale;
22 import java.util.Map;
23
24 /**
25 * Parser for the {@code Content-Disposition} header.
26 * <ul>
27 * <li>RFC 6266 (HTTP) and RFC 2183 (disposition type and parameters)</li>
28 * <li>RFC 8187 extended parameters ({@code filename*=UTF-8''...}); supported charsets are
29 * UTF-8 and ISO-8859-1</li>
30 * </ul>
31 *
32 * <p>The parser is lenient like the browsers are: parameter names are case-insensitive,
33 * white space around names and values is ignored, an unterminated quoted string
34 * ends at the end of the header. An extended value with a broken percent encoding is ignored
35 * (like Firefox does), so the plain parameter can be used as fallback.</p>
36 *
37 * @author Ronald Brill
38 */
39 public final class ContentDispositionParser {
40
41 /**
42 * Utility class, not intended to be instantiated.
43 */
44 private ContentDispositionParser() {
45 // Utility class
46 }
47
48 /**
49 * Parses the value of a {@code Content-Disposition} header.
50 *
51 * @param header the header value
52 * @return the parsed header or {@code null} if the given value is {@code null}
53 */
54 public static ContentDisposition parse(final String header) {
55 if (header == null) {
56 return null;
57 }
58
59 final Map<String, String> parameters = new LinkedHashMap<>();
60 final Map<String, String> extendedParameters = new LinkedHashMap<>();
61
62 final int length = header.length();
63 int pos = header.indexOf(';');
64 final String type = (pos < 0 ? header : header.substring(0, pos)).trim().toLowerCase(Locale.ROOT);
65
66 while (pos >= 0 && pos < length) {
67 pos++; // skip ';'
68
69 // name
70 while (pos < length && Character.isWhitespace(header.charAt(pos))) {
71 pos++;
72 }
73 final int nameStart = pos;
74 while (pos < length && header.charAt(pos) != '=' && header.charAt(pos) != ';') {
75 pos++;
76 }
77 String name = header.substring(nameStart, pos).trim().toLowerCase(Locale.ROOT);
78
79 // value
80 String value = "";
81 if (pos < length && header.charAt(pos) == '=') {
82 pos++;
83 while (pos < length && Character.isWhitespace(header.charAt(pos))) {
84 pos++;
85 }
86 if (pos < length && header.charAt(pos) == '"') {
87 // quoted string; a backslash escapes the next character
88 final StringBuilder sb = new StringBuilder();
89 pos++;
90 while (pos < length && header.charAt(pos) != '"') {
91 if (header.charAt(pos) == '\\' && pos + 1 < length) {
92 pos++;
93 }
94 sb.append(header.charAt(pos));
95 pos++;
96 }
97 if (pos < length) {
98 pos++; // closing quote
99 }
100 // ignore everything up to the next ';'
101 while (pos < length && header.charAt(pos) != ';') {
102 pos++;
103 }
104 value = sb.toString();
105 }
106 else {
107 final int valueStart = pos;
108 while (pos < length && header.charAt(pos) != ';') {
109 pos++;
110 }
111 value = header.substring(valueStart, pos).trim();
112 }
113 }
114
115 if (!name.isEmpty()) {
116 if (name.endsWith("*")) {
117 name = name.substring(0, name.length() - 1);
118 if (!name.isEmpty() && !extendedParameters.containsKey(name)) {
119 final String decoded = decodeExtendedValue(value);
120 if (decoded != null) {
121 extendedParameters.put(name, decoded);
122 }
123 }
124 }
125 else if (!parameters.containsKey(name)) {
126 parameters.put(name, value);
127 }
128 }
129
130 pos = header.indexOf(';', pos);
131 }
132
133 return new ContentDisposition(type, parameters, extendedParameters);
134 }
135
136 /**
137 * Decodes a RFC 8187 value (charset'language'percent-encoded).
138 *
139 * @param value the raw value of the extended parameter
140 * @return the decoded value or {@code null} if the format is wrong, the charset is not supported
141 * or the percent encoding is invalid
142 */
143 private static String decodeExtendedValue(final String value) {
144 final int first = value.indexOf('\'');
145 final int second = first < 0 ? -1 : value.indexOf('\'', first + 1);
146 if (second < 0) {
147 return null;
148 }
149
150 final String charsetName = value.substring(0, first).trim();
151 final Charset charset;
152 if ("UTF-8".equalsIgnoreCase(charsetName)) {
153 charset = StandardCharsets.UTF_8;
154 }
155 else if ("ISO-8859-1".equalsIgnoreCase(charsetName)) {
156 charset = StandardCharsets.ISO_8859_1;
157 }
158 else {
159 return null;
160 }
161
162 final String encoded = value.substring(second + 1);
163 final ByteArrayOutputStream out = new ByteArrayOutputStream();
164 for (int i = 0; i < encoded.length(); i++) {
165 final char c = encoded.charAt(i);
166 if (c == '%' && i + 2 < encoded.length() && isHex(encoded.charAt(i + 1))
167 && isHex(encoded.charAt(i + 2))) {
168 out.write((Character.digit(encoded.charAt(i + 1), 16) << 4)
169 | Character.digit(encoded.charAt(i + 2), 16));
170 i += 2;
171 }
172 else if (c == '%') {
173 // invalid percent encoding; ignore the whole value
174 return null;
175 }
176 else {
177 final byte[] bytes = String.valueOf(c).getBytes(charset);
178 out.write(bytes, 0, bytes.length);
179 }
180 }
181 return new String(out.toByteArray(), charset);
182 }
183
184 /**
185 * Checks for a hexadecimal digit (ASCII only).
186 *
187 * @param c the character to check
188 * @return {@code true} if the character is a hexadecimal digit
189 */
190 private static boolean isHex(final char c) {
191 return c < 128 && Character.digit(c, 16) >= 0;
192 }
193 }