1 | /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
|
---|
2 | * jflex 1.4 *
|
---|
3 | * Copyright (C) 1998-2009 Gerwin Klein <[email protected]> *
|
---|
4 | * All rights reserved. *
|
---|
5 | * *
|
---|
6 | * This program is free software; you can redistribute it and/or modify *
|
---|
7 | * it under the terms of the GNU General Public License. See the file *
|
---|
8 | * COPYRIGHT for more information. *
|
---|
9 | * *
|
---|
10 | * This program is distributed in the hope that it will be useful, *
|
---|
11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of *
|
---|
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the *
|
---|
13 | * GNU General Public License for more details. *
|
---|
14 | * *
|
---|
15 | * You should have received a copy of the GNU General Public License along *
|
---|
16 | * with this program; if not, write to the Free Software Foundation, Inc., *
|
---|
17 | * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA *
|
---|
18 | * *
|
---|
19 | * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
|
---|
20 |
|
---|
21 | package JFlex;
|
---|
22 |
|
---|
23 |
|
---|
24 | /**
|
---|
25 | * Encodes <code>int</code> arrays as strings.
|
---|
26 | *
|
---|
27 | * Also splits up strings when longer than 64K in UTF8 encoding.
|
---|
28 | * Subclasses emit unpacking code.
|
---|
29 | *
|
---|
30 | * Usage protocol:
|
---|
31 | * <code>p.emitInit();</code><br>
|
---|
32 | * <code>for each data: p.emitData(data);</code><br>
|
---|
33 | * <code>p.emitUnpack();</code>
|
---|
34 | *
|
---|
35 | * @author Gerwin Klein
|
---|
36 | * @version $Revision: 433 $, $Date: 2009-01-31 19:52:34 +1100 (Sat, 31 Jan 2009) $
|
---|
37 | */
|
---|
38 | public abstract class PackEmitter {
|
---|
39 |
|
---|
40 | /** name of the generated array (mixed case, no yy prefix) */
|
---|
41 | protected String name;
|
---|
42 |
|
---|
43 | /** current UTF8 length of generated string in current chunk */
|
---|
44 | private int UTF8Length;
|
---|
45 |
|
---|
46 | /** position in the current line */
|
---|
47 | private int linepos;
|
---|
48 |
|
---|
49 | /** max number of entries per line */
|
---|
50 | private static final int maxEntries = 16;
|
---|
51 |
|
---|
52 | /** output buffer */
|
---|
53 | protected StringBuffer out = new StringBuffer();
|
---|
54 |
|
---|
55 | /** number of existing string chunks */
|
---|
56 | protected int chunks;
|
---|
57 |
|
---|
58 | /** maximum size of chunks */
|
---|
59 | // String constants are stored as UTF8 with 2 bytes length
|
---|
60 | // field in class files. One Unicode char can be up to 3
|
---|
61 | // UTF8 bytes. 64K max and two chars safety.
|
---|
62 | private static final int maxSize = 0xFFFF-6;
|
---|
63 |
|
---|
64 | /** indent for string lines */
|
---|
65 | private static final String indent = " ";
|
---|
66 |
|
---|
67 | /**
|
---|
68 | * Create new emitter for an array.
|
---|
69 | *
|
---|
70 | * @param name the name of the generated array
|
---|
71 | */
|
---|
72 | public PackEmitter(String name) {
|
---|
73 | this.name = name;
|
---|
74 | }
|
---|
75 |
|
---|
76 | /**
|
---|
77 | * Convert array name into all uppercase internal scanner
|
---|
78 | * constant name.
|
---|
79 | *
|
---|
80 | * @return <code>name</code> as a internal constant name.
|
---|
81 | * @see PackEmitter#name
|
---|
82 | */
|
---|
83 | protected String constName() {
|
---|
84 | return "ZZ_" + name.toUpperCase();
|
---|
85 | }
|
---|
86 |
|
---|
87 | /**
|
---|
88 | * Return current output buffer.
|
---|
89 | */
|
---|
90 | public String toString() {
|
---|
91 | return out.toString();
|
---|
92 | }
|
---|
93 |
|
---|
94 | /**
|
---|
95 | * Emit declaration of decoded member and open first chunk.
|
---|
96 | */
|
---|
97 | public void emitInit() {
|
---|
98 | out.append(" private static final int [] ");
|
---|
99 | out.append(constName());
|
---|
100 | out.append(" = zzUnpack");
|
---|
101 | out.append(name);
|
---|
102 | out.append("();");
|
---|
103 | nl();
|
---|
104 | nextChunk();
|
---|
105 | }
|
---|
106 |
|
---|
107 | /**
|
---|
108 | * Emit single unicode character.
|
---|
109 | *
|
---|
110 | * Updates length, position, etc.
|
---|
111 | *
|
---|
112 | * @param i the character to emit.
|
---|
113 | * @prec 0 <= i <= 0xFFFF
|
---|
114 | */
|
---|
115 | public void emitUC(int i) {
|
---|
116 | if (i < 0 || i > 0xFFFF)
|
---|
117 | throw new IllegalArgumentException("character value expected");
|
---|
118 |
|
---|
119 | // cast ok because of prec
|
---|
120 | char c = (char) i;
|
---|
121 |
|
---|
122 | printUC(c);
|
---|
123 | UTF8Length += UTF8Length(c);
|
---|
124 | linepos++;
|
---|
125 | }
|
---|
126 |
|
---|
127 | /**
|
---|
128 | * Execute line/chunk break if necessary.
|
---|
129 | * Leave space for at least two chars.
|
---|
130 | */
|
---|
131 | public void breaks() {
|
---|
132 | if (UTF8Length >= maxSize) {
|
---|
133 | // close current chunk
|
---|
134 | out.append("\";");
|
---|
135 | nl();
|
---|
136 |
|
---|
137 | nextChunk();
|
---|
138 | }
|
---|
139 | else {
|
---|
140 | if (linepos >= maxEntries) {
|
---|
141 | // line break
|
---|
142 | out.append("\"+");
|
---|
143 | nl();
|
---|
144 | out.append(indent);
|
---|
145 | out.append("\"");
|
---|
146 | linepos = 0;
|
---|
147 | }
|
---|
148 | }
|
---|
149 | }
|
---|
150 |
|
---|
151 | /**
|
---|
152 | * Emit the unpacking code.
|
---|
153 | */
|
---|
154 | public abstract void emitUnpack();
|
---|
155 |
|
---|
156 | /**
|
---|
157 | * emit next chunk
|
---|
158 | */
|
---|
159 | private void nextChunk() {
|
---|
160 | nl();
|
---|
161 | out.append(" private static final String ");
|
---|
162 | out.append(constName());
|
---|
163 | out.append("_PACKED_");
|
---|
164 | out.append(chunks);
|
---|
165 | out.append(" =");
|
---|
166 | nl();
|
---|
167 | out.append(indent);
|
---|
168 | out.append("\"");
|
---|
169 |
|
---|
170 | UTF8Length = 0;
|
---|
171 | linepos = 0;
|
---|
172 | chunks++;
|
---|
173 | }
|
---|
174 |
|
---|
175 | /**
|
---|
176 | * emit newline
|
---|
177 | */
|
---|
178 | protected void nl() {
|
---|
179 | out.append(Out.NL);
|
---|
180 | }
|
---|
181 |
|
---|
182 | /**
|
---|
183 | * Append a unicode/octal escaped character
|
---|
184 | * to <code>out</code> buffer.
|
---|
185 | *
|
---|
186 | * @param c the character to append
|
---|
187 | */
|
---|
188 | private void printUC(char c) {
|
---|
189 | if (c > 255) {
|
---|
190 | out.append("\\u");
|
---|
191 | if (c < 0x1000) out.append("0");
|
---|
192 | out.append(Integer.toHexString(c));
|
---|
193 | }
|
---|
194 | else {
|
---|
195 | out.append("\\");
|
---|
196 | out.append(Integer.toOctalString(c));
|
---|
197 | }
|
---|
198 | }
|
---|
199 |
|
---|
200 | /**
|
---|
201 | * Calculates the number of bytes a Unicode character
|
---|
202 | * would have in UTF8 representation in a class file.
|
---|
203 | *
|
---|
204 | * @param value the char code of the Unicode character
|
---|
205 | * @prec 0 <= value <= 0xFFFF
|
---|
206 | *
|
---|
207 | * @return length of UTF8 representation.
|
---|
208 | */
|
---|
209 | private int UTF8Length(char value) {
|
---|
210 | // if (value < 0 || value > 0xFFFF) throw new Error("not a char value ("+value+")");
|
---|
211 |
|
---|
212 | // see JVM spec section 4.4.7, p 111
|
---|
213 | if (value == 0) return 2;
|
---|
214 | if (value <= 0x7F) return 1;
|
---|
215 |
|
---|
216 | // workaround for javac bug (up to jdk 1.3):
|
---|
217 | if (value < 0x0400) return 2;
|
---|
218 | if (value <= 0x07FF) return 3;
|
---|
219 |
|
---|
220 | // correct would be:
|
---|
221 | // if (value <= 0x7FF) return 2;
|
---|
222 | return 3;
|
---|
223 | }
|
---|
224 |
|
---|
225 | // convenience
|
---|
226 | protected void println(String s) {
|
---|
227 | out.append(s);
|
---|
228 | nl();
|
---|
229 | }
|
---|
230 | }
|
---|