1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * http://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17 package org.apache.commons.jexl3.parser;
18
19 import java.util.Objects;
20
21 /**
22 * Common constant strings utilities.
23 * <p>
24 * This package methods read JEXL string literals and handle escaping through the
25 * 'backslash' (ie: \) character. Escaping is used to neutralize string delimiters (the single
26 * and double quotes) and read Unicode hexadecimal encoded characters.
27 * </p>
28 * <p>
29 * The only escapable characters are the single and double quotes - ''' and '"' -,
30 * a Unicode sequence starting with 'u' followed by 4 hexadecimals and
31 * the backslash character - '\' - itself.
32 * </p>
33 * <p>
34 * A sequence where '\' occurs before any non-escapable character or sequence has no effect, the
35 * sequence output being the same as the input.
36 * </p>
37 */
38 public class StringParser {
39 /** The length of an escaped unicode sequence. */
40 private static final int UCHAR_LEN = 4;
41
42 /** Initial shift value for composing a Unicode char from 4 nibbles (16 - 4). */
43 private static final int SHIFT = 12;
44
45 /** The base 10 offset used to convert hexa characters to decimal. */
46 private static final int BASE10 = 10;
47
48 /** The last 7bits ASCII character. */
49 private static final char LAST_ASCII = 127;
50
51 /** The first printable 7bits ASCII character. */
52 private static final char FIRST_ASCII = 32;
53
54 /**
55 * Builds a regex pattern string, handles escaping '/' through '\/' syntax.
56 * @param str the string to build from
57 * @return the built string
58 */
59 public static String buildRegex(final CharSequence str) {
60 return buildString(str.subSequence(1, str.length()), true);
61 }
62 /**
63 * Builds a string, handles escaping through '\' syntax.
64 * @param str the string to build from
65 * @param eatsep whether the separator, the first character, should be considered
66 * @return the built string
67 */
68 public static String buildString(final CharSequence str, final boolean eatsep) {
69 return buildString(str, eatsep, true);
70 }
71
72 /**
73 * Builds a string, handles escaping through '\' syntax.
74 * @param str the string to build from
75 * @param eatsep whether the separator, the first character, should be considered
76 * @param esc whether escape characters are interpreted or escaped
77 * @return the built string
78 */
79 private static String buildString(final CharSequence str, final boolean eatsep, final boolean esc) {
80 final StringBuilder strb = new StringBuilder(str.length());
81 final char sep = eatsep ? str.charAt(0) : 0;
82 final int end = str.length() - (eatsep ? 1 : 0);
83 final int begin = eatsep ? 1 : 0;
84 read(strb, str, begin, end, sep, esc);
85 return strb.toString();
86 }
87 /**
88 * Builds a template, does not escape characters.
89 * @param str the string to build from
90 * @param eatsep whether the separator, the first character, should be considered
91 * @return the built string
92 */
93 public static String buildTemplate(final CharSequence str, final boolean eatsep) {
94 return buildString(str, eatsep, false);
95 }
96 /**
97 * Adds a escape char ('\') where needed in a string form of an ide
98 * @param str the identifier un-escaped string
99 * @return the string with added backslash character before space, quote, double-quote and backslash
100 */
101 public static String escapeIdentifier(final String str) {
102 StringBuilder strb = null;
103 if (str != null) {
104 int n = 0;
105 final int last = str.length();
106 while (n < last) {
107 final char c = str.charAt(n);
108 switch (c) {
109 case ' ':
110 case '\'':
111 case '"':
112 case '\\': {
113 if (strb == null) {
114 strb = new StringBuilder(last);
115 strb.append(str, 0, n);
116 }
117 strb.append('\\');
118 strb.append(c);
119 break;
120 }
121 default:
122 if (strb != null) {
123 strb.append(c);
124 }
125 }
126 n += 1;
127 }
128 }
129 return Objects.toString(strb, str);
130 }
131
132 /**
133 * Escapes a String representation, expand non-ASCII characters as Unicode escape sequence.
134 * @param delim the delimiter character
135 * @param str the string to escape
136 * @return the escaped representation
137 */
138 public static String escapeString(final String str, final char delim) {
139 if (str == null) {
140 return null;
141 }
142 final int length = str.length();
143 final StringBuilder strb = new StringBuilder(length + 2);
144 strb.append(delim);
145 for (int i = 0; i < length; ++i) {
146 final char c = str.charAt(i);
147 switch (c) {
148 case 0:
149 continue;
150 case '\b':
151 strb.append('\\');
152 strb.append('b');
153 break;
154 case '\t':
155 strb.append('\\');
156 strb.append('t');
157 break;
158 case '\n':
159 strb.append('\\');
160 strb.append('n');
161 break;
162 case '\f':
163 strb.append('\\');
164 strb.append('f');
165 break;
166 case '\r':
167 strb.append('\\');
168 strb.append('r');
169 break;
170 case '\\':
171 strb.append('\\');
172 strb.append('\\');
173 break;
174 default:
175 if (c == delim) {
176 strb.append('\\');
177 strb.append(delim);
178 } else if (c >= FIRST_ASCII && c <= LAST_ASCII) {
179 strb.append(c);
180 } else {
181 // convert to Unicode escape sequence
182 strb.append('\\');
183 strb.append('u');
184 final String hex = Integer.toHexString(c);
185 for (int h = hex.length(); h < UCHAR_LEN; ++h) {
186 strb.append('0');
187 }
188 strb.append(hex);
189 }
190 }
191 }
192 strb.append(delim);
193 return strb.toString();
194 }
195 /**
196 * Read the remainder of a string till a given separator,
197 * handles escaping through '\' syntax.
198 * @param strb the destination buffer to copy characters into
199 * @param str the origin
200 * @param begin the relative offset in str to begin reading
201 * @param end the relative offset in str to end reading
202 * @param sep the separator, single or double quote, marking end of string
203 * @param esc whether escape characters are interpreted or escaped
204 * @return the last character offset handled in origin
205 */
206 private static int read(final StringBuilder strb, final CharSequence str, final int begin, final int end, final char sep, final boolean esc) {
207 boolean escape = false;
208 int index = begin;
209 for (; index < end; ++index) {
210 final char c = str.charAt(index);
211 if (escape) {
212 if (c == 'u' && index + UCHAR_LEN < end && readUnicodeChar(strb, str, index + 1) > 0) {
213 index += UCHAR_LEN;
214 } else {
215 // if c is not an escapable character, re-emmit the backslash before it
216 final boolean notSeparator = sep == 0 ? c != '\'' && c != '"' : c != sep;
217 if (notSeparator && c != '\\') {
218 if (!esc) {
219 strb.append('\\').append(c);
220 } else {
221 switch (c) {
222 // https://es5.github.io/x7.html#x7.8.4
223 case 'b':
224 strb.append('\b');
225 break; // backspace \u0008
226 case 't':
227 strb.append('\t');
228 break; // horizontal tab \u0009
229 case 'n':
230 strb.append('\n');
231 break; // line feed \u000A
232 // We don't support vertical tab. If needed, the unicode (\u000B) should be used instead
233 case 'f':
234 strb.append('\f');
235 break; // form feed \u000C
236 case 'r':
237 strb.append('\r');
238 break; // carriage return \u000D
239 default:
240 strb.append('\\').append(c);
241 }
242 }
243 } else {
244 strb.append(c);
245 }
246 }
247 escape = false;
248 continue;
249 }
250 if (c == '\\') {
251 escape = true;
252 continue;
253 }
254 strb.append(c);
255 if (c == sep) {
256 break;
257 }
258 }
259 return index;
260 }
261 /**
262 * Read the remainder of a string till a given separator,
263 * handles escaping through '\' syntax.
264 * @param strb the destination buffer to copy characters into
265 * @param str the origin
266 * @param index the offset into the origin
267 * @param sep the separator, single or double quote, marking end of string
268 * @return the offset in origin
269 */
270 public static int readString(final StringBuilder strb, final CharSequence str, final int index, final char sep) {
271 return read(strb, str, index, str.length(), sep, true);
272 }
273
274 /**
275 * Reads a Unicode escape character.
276 * @param strb the builder to write the character to
277 * @param str the sequence
278 * @param begin the begin offset in sequence (after the '\\u')
279 * @return 0 if char could not be read, 4 otherwise
280 */
281 private static int readUnicodeChar(final StringBuilder strb, final CharSequence str, final int begin) {
282 char xc = 0;
283 int bits = SHIFT;
284 int value;
285 for (int offset = 0; offset < UCHAR_LEN; ++offset) {
286 final char c = str.charAt(begin + offset);
287 if (c >= '0' && c <= '9') {
288 value = c - '0';
289 } else if (c >= 'a' && c <= 'h') {
290 value = c - 'a' + BASE10;
291 } else if (c >= 'A' && c <= 'H') {
292 value = c - 'A' + BASE10;
293 } else {
294 return 0;
295 }
296 xc |= value << bits;
297 bits -= UCHAR_LEN;
298 }
299 strb.append(xc);
300 return UCHAR_LEN;
301 }
302
303 /**
304 * Remove escape char ('\') from an identifier.
305 * @param str the identifier escaped string, ie with a backslash before space, quote, double-quote and backslash
306 * @return the string with no '\\' character
307 */
308 public static String unescapeIdentifier(final String str) {
309 StringBuilder strb = null;
310 if (str != null) {
311 int n = 0;
312 final int last = str.length();
313 while (n < last) {
314 final char c = str.charAt(n);
315 if (c == '\\') {
316 if (strb == null) {
317 strb = new StringBuilder(last);
318 strb.append(str, 0, n);
319 }
320 } else if (strb != null) {
321 strb.append(c);
322 }
323 n += 1;
324 }
325 }
326 return Objects.toString(strb, str);
327 }
328
329 /** Default constructor. */
330 protected StringParser() {
331 // nothing to initialize
332 }
333 }