1 | /**************************************************************************
|
---|
2 | *
|
---|
3 | * FText.cpp -- File structures for text compression
|
---|
4 | * Copyright (C) 1999 Rodger McNab
|
---|
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 as published by
|
---|
8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
9 | * (at your option) any later version.
|
---|
10 | *
|
---|
11 | * This program is distributed in the hope that it will be useful,
|
---|
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
14 | * GNU General Public License for more details.
|
---|
15 | *
|
---|
16 | * You should have received a copy of the GNU General Public License
|
---|
17 | * along with this program; if not, write to the Free Software
|
---|
18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
19 | *
|
---|
20 | * $Id: FText.cpp 860 2000-01-18 03:53:24Z rjmcnab $
|
---|
21 | *
|
---|
22 | **************************************************************************/
|
---|
23 |
|
---|
24 | #include "FText.h"
|
---|
25 | #include "netorder.h" /* [RPAP - Jan 97: Endian Ordering] */
|
---|
26 |
|
---|
27 |
|
---|
28 | BitAddr::BitAddr () {
|
---|
29 | Clear ();
|
---|
30 | }
|
---|
31 |
|
---|
32 | void BitAddr::Clear () {
|
---|
33 | byte = 0;
|
---|
34 | bit = 0;
|
---|
35 | }
|
---|
36 |
|
---|
37 | bool BitAddr::Read (FILE *f) {
|
---|
38 | return (ReadUL (f, byte) && ReadUC (f, bit));
|
---|
39 | }
|
---|
40 |
|
---|
41 | bool BitAddr::Write (FILE *f) const {
|
---|
42 | return (WriteUL (f, byte) && WriteUC (f, bit));
|
---|
43 | }
|
---|
44 |
|
---|
45 | #define BITADDRLEN (sizeof(unsigned long) + sizeof(unsigned char))
|
---|
46 |
|
---|
47 |
|
---|
48 | TextLevelInfo::TextLevelInfo () {
|
---|
49 | Clear();
|
---|
50 | }
|
---|
51 |
|
---|
52 | void TextLevelInfo::Clear () {
|
---|
53 | levelTag.erase (levelTag.begin(), levelTag.end());
|
---|
54 | textIdxPtr = 0;
|
---|
55 | numEntries = 0;
|
---|
56 | }
|
---|
57 |
|
---|
58 | bool TextLevelInfo::Read (FILE *f) {
|
---|
59 | return (ReadUCArray (f, levelTag) &&
|
---|
60 | ReadUL (f, textIdxPtr) &&
|
---|
61 | ReadUL (f, numEntries));
|
---|
62 | }
|
---|
63 |
|
---|
64 | bool TextLevelInfo::Write (FILE *f) const {
|
---|
65 | return (WriteUCArray (f, levelTag) &&
|
---|
66 | WriteUL (f, textIdxPtr) &&
|
---|
67 | WriteUL (f, numEntries));
|
---|
68 | }
|
---|
69 |
|
---|
70 | ostream &operator<<(ostream &s, const TextLevelInfo &l) {
|
---|
71 | s << " Tag: \"" << l.levelTag << "\"\n"
|
---|
72 | << " textIdxPtr: " << l.textIdxPtr << "\n"
|
---|
73 | << " numEntries: " << l.numEntries << "\n\n";
|
---|
74 |
|
---|
75 | return s;
|
---|
76 | }
|
---|
77 |
|
---|
78 |
|
---|
79 | FTextLevel::FTextLevel () {
|
---|
80 | Clear();
|
---|
81 | }
|
---|
82 |
|
---|
83 | void FTextLevel::Clear () {
|
---|
84 | levelInfo.erase (levelInfo.begin(), levelInfo.end());
|
---|
85 | }
|
---|
86 |
|
---|
87 | bool FTextLevel::Read (FILE *f) {
|
---|
88 | // read in the array size
|
---|
89 | unsigned long arrSize = 0;
|
---|
90 | if (!ReadVarLenUL (f, arrSize)) return false;
|
---|
91 |
|
---|
92 | TextLevelInfo thisLevel;
|
---|
93 | while (arrSize > 0) {
|
---|
94 | // read and insert the next level information
|
---|
95 | if (!thisLevel.Read (f)) return false;
|
---|
96 | levelInfo[thisLevel.levelTag] = thisLevel;
|
---|
97 |
|
---|
98 | arrSize--;
|
---|
99 | }
|
---|
100 |
|
---|
101 | return true;
|
---|
102 | }
|
---|
103 |
|
---|
104 | bool FTextLevel::Write (FILE *f) const {
|
---|
105 | // write out the array size
|
---|
106 | if (!WriteVarLenUL (f, levelInfo.size())) return false;
|
---|
107 |
|
---|
108 | // write out each level info
|
---|
109 | TextLevelInfoMap::const_iterator here = levelInfo.begin();
|
---|
110 | TextLevelInfoMap::const_iterator end = levelInfo.end();
|
---|
111 | while (here != end) {
|
---|
112 | if (!(*here).second.Write (f)) return false;
|
---|
113 |
|
---|
114 | here++;
|
---|
115 | }
|
---|
116 |
|
---|
117 | return true;
|
---|
118 | }
|
---|
119 |
|
---|
120 | ostream &operator<<(ostream &s, const FTextLevel &l) {
|
---|
121 | TextLevelInfoMap::const_iterator here = l.levelInfo.begin ();
|
---|
122 | TextLevelInfoMap::const_iterator end = l.levelInfo.end ();
|
---|
123 | while (here != end) {
|
---|
124 | s << (*here).second;
|
---|
125 | here++;
|
---|
126 | }
|
---|
127 |
|
---|
128 | return s;
|
---|
129 | }
|
---|
130 |
|
---|
131 |
|
---|
132 | TextIdx::TextIdx () {
|
---|
133 | Clear ();
|
---|
134 | }
|
---|
135 |
|
---|
136 | void TextIdx::Clear () {
|
---|
137 | start.Clear();
|
---|
138 | end.Clear();
|
---|
139 | which = 0;
|
---|
140 | }
|
---|
141 |
|
---|
142 | #define TEXTIDXLEN (BITADDRLEN*2 + sizeof(char))
|
---|
143 |
|
---|
144 | bool TextIdx::Read (FILE *f, const TextLevelInfo &levelInfo,
|
---|
145 | unsigned long docNum) {
|
---|
146 | if (!SeekTextIdx (f, levelInfo, docNum)) return false;
|
---|
147 | return Read (f);
|
---|
148 | }
|
---|
149 |
|
---|
150 | bool TextIdx::Read (FILE *f) {
|
---|
151 | return (start.Read (f) && end.Read (f) && ReadUC (f, which));
|
---|
152 | }
|
---|
153 |
|
---|
154 | bool TextIdx::Write (FILE *f) const {
|
---|
155 | return (start.Write (f) && end.Write (f) && WriteUC (f, which));
|
---|
156 | }
|
---|
157 |
|
---|
158 | ostream &operator<<(ostream &s, const TextIdx &t) {
|
---|
159 | s << "start byte: " << t.start.byte << "\n";
|
---|
160 | s << "start bit: " << (int)t.start.bit << "\n";
|
---|
161 | s << "end byte: " << t.end.byte << "\n";
|
---|
162 | s << "end bit: " << (int)t.end.bit << "\n";
|
---|
163 | s << "which: " << (int)t.which << "\n\n";
|
---|
164 |
|
---|
165 | return s;
|
---|
166 | }
|
---|
167 |
|
---|
168 |
|
---|
169 | bool SeekTextIdx (FILE *f, const TextLevelInfo &levelInfo,
|
---|
170 | unsigned long docNum) {
|
---|
171 | if (docNum == 0 || docNum > levelInfo.numEntries) return false;
|
---|
172 |
|
---|
173 | unsigned long seekPos = levelInfo.textIdxPtr + (docNum-1) * TEXTIDXLEN;
|
---|
174 | if (fseek (f, seekPos, SEEK_SET) != 0) return false;
|
---|
175 |
|
---|
176 | return true;
|
---|
177 | }
|
---|
178 |
|
---|
179 |
|
---|
180 | bool ReadTextIdxArray (FILE *f, TextIdxArray &a, unsigned long arrSize) {
|
---|
181 | // clear the array
|
---|
182 | a.erase (a.begin(), a.end());
|
---|
183 |
|
---|
184 | // read in each element
|
---|
185 | TextIdx idx;
|
---|
186 | while (arrSize > 0) {
|
---|
187 | if (!idx.Read(f)) return false;
|
---|
188 | a.push_back (idx);
|
---|
189 | arrSize --;
|
---|
190 | }
|
---|
191 |
|
---|
192 | return true;
|
---|
193 | }
|
---|
194 |
|
---|
195 | bool WriteTextIdxArray (FILE *f, const TextIdxArray &a) {
|
---|
196 | TextIdxArray::const_iterator here = a.begin();
|
---|
197 | TextIdxArray::const_iterator end = a.end();
|
---|
198 | while (here != end) {
|
---|
199 | if (!(*here).Write(f)) return false;
|
---|
200 | here++;
|
---|
201 | }
|
---|
202 |
|
---|
203 | return true;
|
---|
204 | }
|
---|
205 |
|
---|
206 |
|
---|
207 | CompressTextInfo::CompressTextInfo () {
|
---|
208 | Clear();
|
---|
209 | }
|
---|
210 |
|
---|
211 | void CompressTextInfo::Clear () {
|
---|
212 | ResetStart();
|
---|
213 | docPtrs.erase (docPtrs.begin(), docPtrs.end());
|
---|
214 | }
|
---|
215 |
|
---|
216 | void CompressTextInfo::ResetStart () {
|
---|
217 | inDoc = false;
|
---|
218 | start.Clear();
|
---|
219 | which = 0;
|
---|
220 | }
|
---|
221 |
|
---|
222 | void CompressTextInfo::SetStart (unsigned long startPos,
|
---|
223 | unsigned char startBit,
|
---|
224 | unsigned char startWhich) {
|
---|
225 | // place an imaginary end tag if needed
|
---|
226 | // note: the document tag always needs a closing tag
|
---|
227 | if (inDoc) SetEnd (startPos, startBit);
|
---|
228 |
|
---|
229 | // remember start of level
|
---|
230 | inDoc = true;
|
---|
231 | start.byte = startPos;
|
---|
232 | start.bit = startBit;
|
---|
233 | which = startWhich;
|
---|
234 | }
|
---|
235 |
|
---|
236 | void CompressTextInfo::SetEnd (unsigned long endPos,
|
---|
237 | unsigned char endBit) {
|
---|
238 | if (inDoc) {
|
---|
239 | // add completed entry to list of ptrs
|
---|
240 | TextIdx textIdx;
|
---|
241 | textIdx.start = start;
|
---|
242 | textIdx.end.byte = endPos;
|
---|
243 | textIdx.end.bit = endBit;
|
---|
244 | textIdx.which = which;
|
---|
245 | docPtrs.push_back (textIdx);
|
---|
246 |
|
---|
247 | // reset this entry
|
---|
248 | ResetStart();
|
---|
249 | }
|
---|
250 | }
|
---|
251 |
|
---|