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.pdfbox.encoding;
18
19 import java.io.BufferedReader;
20 import java.io.IOException;
21 import java.io.InputStream;
22 import java.io.InputStreamReader;
23 import java.util.Collections;
24 import java.util.HashMap;
25 import java.util.Map;
26 import java.util.StringTokenizer;
27 import java.util.logging.Level;
28 import java.util.logging.Logger;
29 import java.io.File;
30
31 import org.apache.pdfbox.cos.COSName;
32 import org.apache.pdfbox.pdmodel.common.COSObjectable;
33 import org.apache.pdfbox.util.ResourceLoader;
34
35 /**
36 * This is an interface to a text encoder.
37 *
38 * @author <a href="mailto:ben@benlitchfield.com">Ben Litchfield</a>
39 * @version $Revision: 1.15 $
40 */
41 public abstract class Encoding implements COSObjectable
42 {
43 /**
44 * This is a mapping from a character code to a character name.
45 */
46 protected final Map<Integer, COSName> codeToName =
47 new HashMap<Integer, COSName>();
48
49 /**
50 * This is a mapping from a character name to a character code.
51 */
52 protected final Map<COSName, Integer> nameToCode =
53 new HashMap<COSName, Integer>();
54
55 private static final Map<COSName, String> NAME_TO_CHARACTER =
56 new HashMap<COSName, String>();
57
58 private static final Map<String, COSName> CHARACTER_TO_NAME =
59 new HashMap<String, COSName>();
60
61 static
62 {
63 //Loads the official Adobe Glyph List
64 loadGlyphList("Resources/glyphlist.txt");
65
66 // Load an external glyph list file that user can give as JVM property
67 String location = System.getProperty("glyphlist_ext");
68 if(location != null)
69 {
70 File external = new File(location);
71 if(external.exists())
72 {
73 loadGlyphList(location);
74 }
75 }
76
77 NAME_TO_CHARACTER.put( COSName.getPDFName( ".notdef" ), "" );
78 NAME_TO_CHARACTER.put( COSName.getPDFName( "fi" ), "fi" );
79 NAME_TO_CHARACTER.put( COSName.getPDFName( "fl" ), "fl" );
80 NAME_TO_CHARACTER.put( COSName.getPDFName( "ffi" ), "ffi" );
81 NAME_TO_CHARACTER.put( COSName.getPDFName( "ff" ), "ff" );
82 NAME_TO_CHARACTER.put( COSName.getPDFName( "pi" ), "pi" );
83
84 // add some (alternative) glyph mappings. These are missing in
85 // the original copy of the adobe glyphlist.txt
86 // also mapped as anglebracketleft
87 NAME_TO_CHARACTER.put(COSName.getPDFName("angbracketleft"), "\u3008");
88 // also mapped as anglebracketright
89 NAME_TO_CHARACTER.put(COSName.getPDFName("angbracketright"), "\u3009");
90 // also mapped as copyright
91 NAME_TO_CHARACTER.put(COSName.getPDFName("circlecopyrt"), "\u00A9");
92 NAME_TO_CHARACTER.put(COSName.getPDFName("controlNULL"), "\u0000");
93
94 for( Map.Entry<COSName, String> entry : NAME_TO_CHARACTER.entrySet() )
95 {
96 CHARACTER_TO_NAME.put( entry.getValue(), entry.getKey() );
97 }
98 }
99
100 /**
101 * Loads a glyph list from a given location and populates the NAME_TO_CHARACTER hashmap
102 * for character lookups.
103 * @param location - The string location of the glyphlist file
104 */
105 private static void loadGlyphList(String location)
106 {
107 BufferedReader glyphStream = null;
108 try
109 {
110 InputStream resource = ResourceLoader.loadResource( location );
111 glyphStream = new BufferedReader( new InputStreamReader( resource ) );
112 String line = null;
113 while( (line = glyphStream.readLine()) != null )
114 {
115 line = line.trim();
116 //lines starting with # are comments which we can ignore.
117 if( !line.startsWith("#" ) )
118 {
119 int semicolonIndex = line.indexOf( ';' );
120 if( semicolonIndex >= 0 )
121 {
122 try
123 {
124 String characterName = line.substring( 0, semicolonIndex );
125 String unicodeValue = line.substring( semicolonIndex+1, line.length() );
126 StringTokenizer tokenizer = new StringTokenizer( unicodeValue, " ", false );
127 String value = "";
128 while(tokenizer.hasMoreTokens())
129 {
130 int characterCode = Integer.parseInt( tokenizer.nextToken(), 16 );
131 value += (char)characterCode;
132 }
133 NAME_TO_CHARACTER.put( COSName.getPDFName( characterName ), value );
134 }
135 catch( NumberFormatException nfe )
136 {
137 nfe.printStackTrace();
138 }
139 }
140 }
141 }
142 }
143 catch( IOException io )
144 {
145 io.printStackTrace();
146 }
147 finally
148 {
149 if( glyphStream != null )
150 {
151 try
152 {
153 glyphStream.close();
154 }
155 catch( IOException e )
156 {
157 e.printStackTrace();
158 }
159
160 }
161 }
162 }
163
164 /**
165 * Returns an unmodifiable view of the Code2Name mapping.
166 * @return the Code2Name map
167 */
168 public Map<Integer, COSName> getCodeToNameMap()
169 {
170 return Collections.unmodifiableMap(codeToName);
171 }
172
173 /**
174 * Returns an unmodifiable view of the Name2Code mapping.
175 * @return the Name2Code map
176 */
177 public Map<COSName, Integer> getNameToCodeMap()
178 {
179 return Collections.unmodifiableMap(nameToCode);
180 }
181
182 /**
183 * This will add a character encoding.
184 *
185 * @param code The character code that matches the character.
186 * @param name The name of the character.
187 */
188 protected void addCharacterEncoding( int code, COSName name )
189 {
190 codeToName.put( code, name );
191 nameToCode.put( name, code );
192 }
193
194 /**
195 * This will get the character code for the name.
196 *
197 * @param name The name of the character.
198 *
199 * @return The code for the character.
200 *
201 * @throws IOException If there is no character code for the name.
202 */
203 public int getCode( COSName name ) throws IOException
204 {
205 Integer code = nameToCode.get( name );
206 if( code == null )
207 {
208 throw new IOException( "No character code for character name '" + name.getName() + "'" );
209 }
210 return code;
211 }
212
213 /**
214 * This will take a character code and get the name from the code.
215 *
216 * @param code The character code.
217 *
218 * @return The name of the character.
219 *
220 * @throws IOException If there is no name for the code.
221 */
222 public COSName getName( int code ) throws IOException
223 {
224 COSName name = codeToName.get( code );
225 if( name == null )
226 {
227 //lets be forgiving for now
228 name = COSName.getPDFName( "space" );
229 //throw new IOException( getClass().getName() +
230 // ": No name for character code '" + code + "'" );
231 }
232 return name;
233 }
234
235 /**
236 * This will take a character code and get the name from the code.
237 *
238 * @param c The character.
239 *
240 * @return The name of the character.
241 *
242 * @throws IOException If there is no name for the character.
243 */
244 public COSName getNameFromCharacter( char c ) throws IOException
245 {
246 COSName name = CHARACTER_TO_NAME.get( Character.toString(c) );
247 if( name == null )
248 {
249 throw new IOException( "No name for character '" + c + "'" );
250 }
251 return name;
252 }
253
254 /**
255 * This will get the character from the code.
256 *
257 * @param code The character code.
258 *
259 * @return The printable character for the code.
260 *
261 * @throws IOException If there is not name for the character.
262 */
263 public String getCharacter( int code ) throws IOException
264 {
265 String character = getCharacter( getName( code ) );
266 return character;
267 }
268
269 /**
270 * This will get the character from the name.
271 *
272 * @param name The name of the character.
273 *
274 * @return The printable character for the code.
275 */
276 public static String getCharacter( COSName name )
277 {
278 COSName baseName = name;
279
280 String character = NAME_TO_CHARACTER.get( baseName );
281 if( character == null )
282 {
283 String nameStr = baseName.getName();
284 // test for Unicode name
285 // (uniXXXX - XXXX must be a multiple of four;
286 // each representing a hexadecimal Unicode code point)
287 if ( nameStr.startsWith( "uni" ) )
288 {
289 StringBuilder uniStr = new StringBuilder();
290
291 for ( int chPos = 3; chPos + 4 <= nameStr.length(); chPos += 4 )
292 {
293 try
294 {
295 int characterCode = Integer.parseInt( nameStr.substring( chPos, chPos + 4), 16 );
296
297 if ( ( characterCode > 0xD7FF ) && ( characterCode < 0xE000 ) )
298 {
299 Logger.getLogger(Encoding.class.getName()).log( Level.WARNING,
300 "Unicode character name with not allowed code area: " +
301 nameStr );
302 }
303 else
304 {
305 uniStr.append( (char) characterCode );
306 }
307 }
308 catch (NumberFormatException nfe)
309 {
310 Logger.getLogger(Encoding.class.getName()).log( Level.WARNING,
311 "Not a number in Unicode character name: " +
312 nameStr );
313 }
314 }
315 character = uniStr.toString();
316 }
317 else
318 {
319 // test if we have a suffix and if so remove it
320 if ( nameStr.indexOf('.') > 0 )
321 {
322 nameStr = nameStr.substring( 0, nameStr.indexOf('.') );
323 baseName = COSName.getPDFName( nameStr );
324 getCharacter(baseName);
325 }
326
327 character = nameStr;
328 }
329 }
330 return character;
331 }
332
333 }