1 /**************************************************************
2  *
3  * Licensed to the Apache Software Foundation (ASF) under one
4  * or more contributor license agreements.  See the NOTICE file
5  * distributed with this work for additional information
6  * regarding copyright ownership.  The ASF licenses this file
7  * to you under the Apache License, Version 2.0 (the
8  * "License"); you may not use this file except in compliance
9  * with the License.  You may obtain a copy of the License at
10  *
11  *   http://www.apache.org/licenses/LICENSE-2.0
12  *
13  * Unless required by applicable law or agreed to in writing,
14  * software distributed under the License is distributed on an
15  * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
16  * KIND, either express or implied.  See the License for the
17  * specific language governing permissions and limitations
18  * under the License.
19  *
20  *************************************************************/
21 
22 
23 
24 // MARKER(update_precomp.py): autogen include statement, do not remove
25 #include "precompiled_i18npool.hxx"
26 
27 #include <stdio.h>
28 #include <string.h>
29 #include <stdlib.h>
30 #include <sal/main.h>
31 #include <sal/types.h>
32 #include <rtl/strbuf.hxx>
33 #include <rtl/ustring.hxx>
34 
35 using namespace ::rtl;
36 
37 /* Main Procedure */
38 
39 SAL_IMPLEMENT_MAIN_WITH_ARGS(argc, argv)
40 {
41 	FILE *sfp, *cfp;
42 
43 	if (argc < 3) exit(-1);
44 
45 	sfp = fopen(argv[1], "rb");	// open the source file for read;
46 	if (sfp == NULL)
47 	{
48 	    printf("Open the dictionary source file failed.");
49 	    return -1;
50 	}
51 
52 	// create the C source file to write
53 	cfp = fopen(argv[2], "wb");
54 	if (cfp == NULL) {
55 	    fclose(sfp);
56 	    printf("Can't create the C source file.");
57 	    return -1;
58 	}
59 
60 	fprintf(cfp, "/*\n");
61 	fprintf(cfp, " * Copyright(c) 1999 - 2000, Sun Microsystems, Inc.\n");
62 	fprintf(cfp, " * All Rights Reserved.\n");
63 	fprintf(cfp, " */\n\n");
64 	fprintf(cfp, "/* !!!The file is generated automatically. DONOT edit the file manually!!! */\n\n");
65 	fprintf(cfp, "#include <sal/types.h>\n\n");
66 	fprintf(cfp, "extern \"C\" {\n");
67 
68 	sal_Int32 count, i, j;
69 	sal_Int32 lenArrayCurr = 0, lenArrayCount = 0, lenArrayLen = 0, *lenArray = NULL, charArray[0x10000];
70 	sal_Bool exist[0x10000];
71 	for (i = 0; i < 0x10000; i++) {
72 	    exist[i] = sal_False;
73 	    charArray[i] = 0;
74 	}
75 
76 	// generate main dict. data array
77 	fprintf(cfp, "static const sal_Unicode dataArea[] = {");
78 	sal_Char str[1024];
79 	sal_Unicode current = 0;
80 	count = 0;
81 	while (fgets(str, 1024, sfp)) {
82 	    // input file is in UTF-8 encoding
83 	    // don't convert last new line character to Ostr.
84 	    OUString Ostr((const sal_Char *)str, strlen(str) - 1, RTL_TEXTENCODING_UTF8);
85 	    const sal_Unicode *u = Ostr.getStr();
86 
87 	    sal_Int32 len = Ostr.getLength();
88 
89         i=0;
90         Ostr.iterateCodePoints(&i, 1);
91 	    if (len == i) continue;	// skip one character word
92 
93 	    if (*u != current) {
94 		if (*u < current)
95 		printf("u %x, current %x, count %d, lenArrayCount %d\n", *u, current,
96 		            sal::static_int_cast<int>(count), sal::static_int_cast<int>(lenArrayCount));
97 		current = *u;
98 		charArray[current] = lenArrayCount;
99 	    }
100 
101 	    if (lenArrayLen <= lenArrayCount+1)
102 		lenArray = (sal_Int32*) realloc(lenArray, (lenArrayLen += 1000) * sizeof(sal_Int32));
103 	    lenArray[lenArrayCount++] = lenArrayCurr;
104 
105 	    exist[u[0]] = sal_True;
106 	    for (i = 1; i < len; i++) {		// start from second character,
107 		exist[u[i]] = sal_True; 	// since the first character is captured in charArray.
108 		lenArrayCurr++;
109 		if ((count++) % 0x10 == 0)
110 		    fprintf(cfp, "\n\t");
111 		fprintf(cfp, "0x%04x, ", u[i]);
112 	    }
113 	}
114 	lenArray[lenArrayCount++] = lenArrayCurr; // store last ending pointer
115 	charArray[current+1] = lenArrayCount;
116 	fprintf(cfp, "\n};\n");
117 
118 	// generate lenArray
119 	fprintf(cfp, "static const sal_Int32 lenArray[] = {\n\t");
120 	count = 1;
121 	fprintf(cfp, "0x%x, ", 0); // insert one slat for skipping 0 in index2 array.
122 	for (i = 0; i < lenArrayCount; i++) {
123 	    fprintf(cfp, "0x%lx, ", static_cast<long unsigned int>(lenArray[i]));
124 	    if (count == 0xf) {
125 		count = 0;
126 		fprintf(cfp, "\n\t");
127 	    } else count++;
128 	}
129 	fprintf(cfp, "\n};\n");
130 
131 	free(lenArray);
132 
133 	// generate index1 array
134 	fprintf (cfp, "static const sal_Int16 index1[] = {\n\t");
135 	sal_Int16 set[0x100];
136 	count = 0;
137 	for (i = 0; i < 0x100; i++) {
138 	    for (j = 0; j < 0x100; j++)
139 		if (charArray[(i*0x100) + j] != 0)
140 		    break;
141 
142 	    fprintf(cfp, "0x%02x, ", set[i] = (j < 0x100 ? sal::static_int_cast<sal_Int16>(count++) : 0xff));
143 	    if ((i+1) % 0x10 == 0)
144 		fprintf (cfp, "\n\t");
145 	}
146 	fprintf (cfp, "};\n");
147 
148 	// generate index2 array
149 	fprintf (cfp, "static const sal_Int32 index2[] = {\n\t");
150 	sal_Int32 prev = 0;
151 	for (i = 0; i < 0x100; i++) {
152 	    if (set[i] != 0xff) {
153 		for (j = 0; j < 0x100; j++) {
154 		    sal_Int32 k = (i*0x100) + j;
155 		    if (prev != 0 && charArray[k] == 0) {
156 			for (k++; k < 0x10000; k++)
157 			    if (charArray[k] != 0)
158 				break;
159 		    }
160 		    prev = charArray[(i*0x100) + j];
161 		    fprintf(
162                 cfp, "0x%lx, ",
163                 sal::static_int_cast< unsigned long >(
164                     k < 0x10000 ? charArray[k] + 1 : 0));
165 		    if ((j+1) % 0x10 == 0)
166 			fprintf (cfp, "\n\t");
167 		}
168 		fprintf (cfp, "\n\t");
169 	    }
170 	}
171 	fprintf (cfp, "\n};\n");
172 
173 	// generate existMark array
174 	count = 0;
175 	fprintf (cfp, "static const sal_uInt8 existMark[] = {\n\t");
176 	for (i = 0; i < 0x1FFF; i++) {
177 	    sal_uInt8 bit = 0;
178 	    for (j = 0; j < 8; j++)
179 		if (exist[i * 8 + j])
180 		    bit |= 1 << j;
181 	    fprintf(cfp, "0x%02x, ", bit);
182 	    if (count == 0xf) {
183 		count = 0;
184 		fprintf(cfp, "\n\t");
185 	    } else count++;
186 	}
187 	fprintf (cfp, "\n};\n");
188 
189 	// create function to return arrays
190 	fprintf (cfp, "\tconst sal_uInt8* getExistMark() { return existMark; }\n");
191 	fprintf (cfp, "\tconst sal_Int16* getIndex1() { return index1; }\n");
192 	fprintf (cfp, "\tconst sal_Int32* getIndex2() { return index2; }\n");
193 	fprintf (cfp, "\tconst sal_Int32* getLenArray() { return lenArray; }\n");
194 	fprintf (cfp, "\tconst sal_Unicode* getDataArea() { return dataArea; }\n");
195 	fprintf (cfp, "}\n");
196 
197 	fclose(sfp);
198 	fclose(cfp);
199 
200 	return 0;
201 }	// End of main
202