1 | // License: GPL. For details, see LICENSE file.
|
---|
2 | package org.openstreetmap.josm.data.osm.search;
|
---|
3 |
|
---|
4 | import static org.openstreetmap.josm.tools.I18n.marktr;
|
---|
5 | import static org.openstreetmap.josm.tools.I18n.tr;
|
---|
6 |
|
---|
7 | import java.io.IOException;
|
---|
8 | import java.io.Reader;
|
---|
9 | import java.util.Arrays;
|
---|
10 | import java.util.List;
|
---|
11 | import java.util.Objects;
|
---|
12 |
|
---|
13 | import org.openstreetmap.josm.tools.JosmRuntimeException;
|
---|
14 |
|
---|
15 | /**
|
---|
16 | * This class is used to parse a search string and split it into tokens.
|
---|
17 | * It provides methods to parse numbers and extract strings.
|
---|
18 | * @since 12656 (moved from actions.search package)
|
---|
19 | */
|
---|
20 | public class PushbackTokenizer {
|
---|
21 |
|
---|
22 | /**
|
---|
23 | * A range of long numbers. Immutable
|
---|
24 | */
|
---|
25 | public static class Range {
|
---|
26 | private final long start;
|
---|
27 | private final long end;
|
---|
28 |
|
---|
29 | /**
|
---|
30 | * Create a new range
|
---|
31 | * @param start The start
|
---|
32 | * @param end The end (inclusive)
|
---|
33 | */
|
---|
34 | public Range(long start, long end) {
|
---|
35 | this.start = start;
|
---|
36 | this.end = end;
|
---|
37 | }
|
---|
38 |
|
---|
39 | /**
|
---|
40 | * @return The start
|
---|
41 | */
|
---|
42 | public long getStart() {
|
---|
43 | return start;
|
---|
44 | }
|
---|
45 |
|
---|
46 | /**
|
---|
47 | * @return The end (inclusive)
|
---|
48 | */
|
---|
49 | public long getEnd() {
|
---|
50 | return end;
|
---|
51 | }
|
---|
52 |
|
---|
53 | @Override
|
---|
54 | public String toString() {
|
---|
55 | return "Range [start=" + start + ", end=" + end + ']';
|
---|
56 | }
|
---|
57 | }
|
---|
58 |
|
---|
59 | private final Reader search;
|
---|
60 |
|
---|
61 | private Token currentToken;
|
---|
62 | private String currentText;
|
---|
63 | private Long currentNumber;
|
---|
64 | private Long currentRange;
|
---|
65 | private int c;
|
---|
66 | private boolean isRange;
|
---|
67 |
|
---|
68 | /**
|
---|
69 | * Creates a new {@link PushbackTokenizer}
|
---|
70 | * @param search The search string reader to read the tokens from
|
---|
71 | */
|
---|
72 | public PushbackTokenizer(Reader search) {
|
---|
73 | this.search = search;
|
---|
74 | getChar();
|
---|
75 | }
|
---|
76 |
|
---|
77 | /**
|
---|
78 | * The token types that may be read
|
---|
79 | */
|
---|
80 | public enum Token {
|
---|
81 | /**
|
---|
82 | * Not token (-)
|
---|
83 | */
|
---|
84 | NOT(marktr("<not>")),
|
---|
85 | /**
|
---|
86 | * Or token (or) (|)
|
---|
87 | */
|
---|
88 | OR(marktr("<or>")),
|
---|
89 | /**
|
---|
90 | * Xor token (xor) (^)
|
---|
91 | */
|
---|
92 | XOR(marktr("<xor>")),
|
---|
93 | /**
|
---|
94 | * opening parentheses token (
|
---|
95 | */
|
---|
96 | LEFT_PARENT(marktr("<left parent>")),
|
---|
97 | /**
|
---|
98 | * closing parentheses token )
|
---|
99 | */
|
---|
100 | RIGHT_PARENT(marktr("<right parent>")),
|
---|
101 | /**
|
---|
102 | * Colon :
|
---|
103 | */
|
---|
104 | COLON(marktr("<colon>")),
|
---|
105 | /**
|
---|
106 | * The equals sign (=)
|
---|
107 | */
|
---|
108 | EQUALS(marktr("<equals>")),
|
---|
109 | /**
|
---|
110 | * The tilde sign (~)
|
---|
111 | */
|
---|
112 | TILDE(marktr("<tilde>")),
|
---|
113 | /**
|
---|
114 | * A text
|
---|
115 | */
|
---|
116 | KEY(marktr("<key>")),
|
---|
117 | /**
|
---|
118 | * A question mark (?)
|
---|
119 | */
|
---|
120 | QUESTION_MARK(marktr("<question mark>")),
|
---|
121 | /**
|
---|
122 | * Marks the end of the input
|
---|
123 | */
|
---|
124 | EOF(marktr("<end-of-file>")),
|
---|
125 | /**
|
---|
126 | * Less than sign (<)
|
---|
127 | */
|
---|
128 | LESS_THAN("<less-than>"),
|
---|
129 | /**
|
---|
130 | * Greater than sign (>)
|
---|
131 | */
|
---|
132 | GREATER_THAN("<greater-than>");
|
---|
133 |
|
---|
134 | Token(String name) {
|
---|
135 | this.name = name;
|
---|
136 | }
|
---|
137 |
|
---|
138 | private final String name;
|
---|
139 |
|
---|
140 | @Override
|
---|
141 | public String toString() {
|
---|
142 | return tr(name);
|
---|
143 | }
|
---|
144 | }
|
---|
145 |
|
---|
146 | private void getChar() {
|
---|
147 | try {
|
---|
148 | c = search.read();
|
---|
149 | } catch (IOException e) {
|
---|
150 | throw new JosmRuntimeException(e.getMessage(), e);
|
---|
151 | }
|
---|
152 | }
|
---|
153 |
|
---|
154 | private static final List<Character> SPECIAL_CHARS = Arrays.asList('"', ':', '(', ')', '|', '^', '=', '~', '?', '<', '>');
|
---|
155 | private static final List<Character> SPECIAL_CHARS_QUOTED = Arrays.asList('"');
|
---|
156 |
|
---|
157 | private String getString(boolean quoted) {
|
---|
158 | List<Character> sChars = quoted ? SPECIAL_CHARS_QUOTED : SPECIAL_CHARS;
|
---|
159 | StringBuilder s = new StringBuilder();
|
---|
160 | boolean escape = false;
|
---|
161 | while (c != -1 && (escape || (!sChars.contains((char) c) && (quoted || !Character.isWhitespace(c))))) {
|
---|
162 | if (c == '\\' && !escape) {
|
---|
163 | escape = true;
|
---|
164 | } else {
|
---|
165 | s.append((char) c);
|
---|
166 | escape = false;
|
---|
167 | }
|
---|
168 | getChar();
|
---|
169 | }
|
---|
170 | return s.toString();
|
---|
171 | }
|
---|
172 |
|
---|
173 | private String getString() {
|
---|
174 | return getString(false);
|
---|
175 | }
|
---|
176 |
|
---|
177 | /**
|
---|
178 | * The token returned is <code>null</code> or starts with an identifier character:
|
---|
179 | * - for an '-'. This will be the only character
|
---|
180 | * : for an key. The value is the next token
|
---|
181 | * | for "OR"
|
---|
182 | * ^ for "XOR"
|
---|
183 | * ' ' for anything else.
|
---|
184 | * @return The next token in the stream.
|
---|
185 | */
|
---|
186 | public Token nextToken() {
|
---|
187 | if (currentToken != null) {
|
---|
188 | Token result = currentToken;
|
---|
189 | currentToken = null;
|
---|
190 | return result;
|
---|
191 | }
|
---|
192 |
|
---|
193 | while (Character.isWhitespace(c)) {
|
---|
194 | getChar();
|
---|
195 | }
|
---|
196 | switch (c) {
|
---|
197 | case -1:
|
---|
198 | getChar();
|
---|
199 | return Token.EOF;
|
---|
200 | case ':':
|
---|
201 | getChar();
|
---|
202 | return Token.COLON;
|
---|
203 | case '=':
|
---|
204 | getChar();
|
---|
205 | return Token.EQUALS;
|
---|
206 | case '~':
|
---|
207 | getChar();
|
---|
208 | return Token.TILDE;
|
---|
209 | case '<':
|
---|
210 | getChar();
|
---|
211 | return Token.LESS_THAN;
|
---|
212 | case '>':
|
---|
213 | getChar();
|
---|
214 | return Token.GREATER_THAN;
|
---|
215 | case '(':
|
---|
216 | getChar();
|
---|
217 | return Token.LEFT_PARENT;
|
---|
218 | case ')':
|
---|
219 | getChar();
|
---|
220 | return Token.RIGHT_PARENT;
|
---|
221 | case '|':
|
---|
222 | getChar();
|
---|
223 | return Token.OR;
|
---|
224 | case '^':
|
---|
225 | getChar();
|
---|
226 | return Token.XOR;
|
---|
227 | case '&':
|
---|
228 | getChar();
|
---|
229 | return nextToken();
|
---|
230 | case '?':
|
---|
231 | getChar();
|
---|
232 | return Token.QUESTION_MARK;
|
---|
233 | case '"':
|
---|
234 | getChar();
|
---|
235 | currentText = getString(true);
|
---|
236 | getChar();
|
---|
237 | return Token.KEY;
|
---|
238 | default:
|
---|
239 | String prefix = "";
|
---|
240 | if (c == '-') {
|
---|
241 | getChar();
|
---|
242 | if (!Character.isDigit(c))
|
---|
243 | return Token.NOT;
|
---|
244 | prefix = "-";
|
---|
245 | }
|
---|
246 | currentText = prefix + getString();
|
---|
247 | if ("or".equalsIgnoreCase(currentText))
|
---|
248 | return Token.OR;
|
---|
249 | else if ("xor".equalsIgnoreCase(currentText))
|
---|
250 | return Token.XOR;
|
---|
251 | else if ("and".equalsIgnoreCase(currentText))
|
---|
252 | return nextToken();
|
---|
253 | // try parsing number
|
---|
254 | try {
|
---|
255 | currentNumber = Long.valueOf(currentText);
|
---|
256 | } catch (NumberFormatException e) {
|
---|
257 | currentNumber = null;
|
---|
258 | }
|
---|
259 | // if text contains "-", try parsing a range
|
---|
260 | int pos = currentText.indexOf('-', 1);
|
---|
261 | isRange = pos > 0;
|
---|
262 | if (isRange) {
|
---|
263 | try {
|
---|
264 | currentNumber = Long.valueOf(currentText.substring(0, pos));
|
---|
265 | } catch (NumberFormatException e) {
|
---|
266 | currentNumber = null;
|
---|
267 | }
|
---|
268 | try {
|
---|
269 | currentRange = Long.valueOf(currentText.substring(pos + 1));
|
---|
270 | } catch (NumberFormatException e) {
|
---|
271 | currentRange = null;
|
---|
272 | }
|
---|
273 | } else {
|
---|
274 | currentRange = null;
|
---|
275 | }
|
---|
276 | return Token.KEY;
|
---|
277 | }
|
---|
278 | }
|
---|
279 |
|
---|
280 | /**
|
---|
281 | * Reads the next token if it is equal to the given, suggested token
|
---|
282 | * @param token The token the next one should be equal to
|
---|
283 | * @return <code>true</code> if it has been read
|
---|
284 | */
|
---|
285 | public boolean readIfEqual(Token token) {
|
---|
286 | Token nextTok = nextToken();
|
---|
287 | if (Objects.equals(nextTok, token))
|
---|
288 | return true;
|
---|
289 | currentToken = nextTok;
|
---|
290 | return false;
|
---|
291 | }
|
---|
292 |
|
---|
293 | /**
|
---|
294 | * Reads the next token. If it is a text, return that text. If not, advance
|
---|
295 | * @return the text or <code>null</code> if the reader was advanced
|
---|
296 | */
|
---|
297 | public String readTextOrNumber() {
|
---|
298 | Token nextTok = nextToken();
|
---|
299 | if (nextTok == Token.KEY)
|
---|
300 | return currentText;
|
---|
301 | currentToken = nextTok;
|
---|
302 | return null;
|
---|
303 | }
|
---|
304 |
|
---|
305 | /**
|
---|
306 | * Reads a number
|
---|
307 | * @param errorMessage The error if the number cannot be read
|
---|
308 | * @return The number that was found
|
---|
309 | * @throws SearchParseError if there is no number
|
---|
310 | */
|
---|
311 | public long readNumber(String errorMessage) throws SearchParseError {
|
---|
312 | if ((nextToken() == Token.KEY) && (currentNumber != null))
|
---|
313 | return currentNumber;
|
---|
314 | else
|
---|
315 | throw new SearchParseError(errorMessage);
|
---|
316 | }
|
---|
317 |
|
---|
318 | /**
|
---|
319 | * Gets the last number that was read
|
---|
320 | * @return The last number
|
---|
321 | */
|
---|
322 | public long getReadNumber() {
|
---|
323 | return (currentNumber != null) ? currentNumber : 0;
|
---|
324 | }
|
---|
325 |
|
---|
326 | /**
|
---|
327 | * Reads a range of numbers
|
---|
328 | * @param errorMessage The error if the input is malformed
|
---|
329 | * @return The range that was found
|
---|
330 | * @throws SearchParseError If the input is not as expected for a range
|
---|
331 | */
|
---|
332 | public Range readRange(String errorMessage) throws SearchParseError {
|
---|
333 | if (nextToken() != Token.KEY || (currentNumber == null && currentRange == null)) {
|
---|
334 | throw new SearchParseError(errorMessage);
|
---|
335 | } else if (!isRange && currentNumber != null) {
|
---|
336 | if (currentNumber >= 0) {
|
---|
337 | return new Range(currentNumber, currentNumber);
|
---|
338 | } else {
|
---|
339 | return new Range(0, Math.abs(currentNumber));
|
---|
340 | }
|
---|
341 | } else if (isRange && currentRange == null) {
|
---|
342 | return new Range(currentNumber, Long.MAX_VALUE);
|
---|
343 | } else if (currentNumber != null && currentRange != null) {
|
---|
344 | return new Range(currentNumber, currentRange);
|
---|
345 | } else {
|
---|
346 | throw new SearchParseError(errorMessage);
|
---|
347 | }
|
---|
348 | }
|
---|
349 |
|
---|
350 | /**
|
---|
351 | * Gets the last text that was found
|
---|
352 | * @return The text
|
---|
353 | */
|
---|
354 | public String getText() {
|
---|
355 | return currentText;
|
---|
356 | }
|
---|
357 | }
|
---|