[3365] | 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 | **************************************************************************/
|
---|
| 21 |
|
---|
| 22 | #include "FText.h"
|
---|
| 23 | #include "netorder.h" /* [RPAP - Jan 97: Endian Ordering] */
|
---|
| 24 |
|
---|
| 25 |
|
---|
| 26 | BitAddr::BitAddr () {
|
---|
| 27 | Clear ();
|
---|
| 28 | }
|
---|
| 29 |
|
---|
| 30 | void BitAddr::Clear () {
|
---|
| 31 | byte = 0;
|
---|
| 32 | bit = 0;
|
---|
| 33 | }
|
---|
| 34 |
|
---|
| 35 | bool BitAddr::Read (FILE *f) {
|
---|
| 36 | return (ReadUL (f, byte) && ReadUC (f, bit));
|
---|
| 37 | }
|
---|
| 38 |
|
---|
| 39 | bool BitAddr::Write (FILE *f) const {
|
---|
| 40 | return (WriteUL (f, byte) && WriteUC (f, bit));
|
---|
| 41 | }
|
---|
| 42 |
|
---|
| 43 | #define BITADDRLEN (sizeof(unsigned long) + sizeof(unsigned char))
|
---|
| 44 |
|
---|
| 45 |
|
---|
| 46 | TextLevelInfo::TextLevelInfo () {
|
---|
| 47 | Clear();
|
---|
| 48 | }
|
---|
| 49 |
|
---|
| 50 | void TextLevelInfo::Clear () {
|
---|
| 51 | levelTag.erase (levelTag.begin(), levelTag.end());
|
---|
| 52 | textIdxPtr = 0;
|
---|
| 53 | numEntries = 0;
|
---|
| 54 | }
|
---|
| 55 |
|
---|
| 56 | bool TextLevelInfo::Read (FILE *f) {
|
---|
| 57 | return (ReadUCArray (f, levelTag) &&
|
---|
| 58 | ReadUL (f, textIdxPtr) &&
|
---|
| 59 | ReadUL (f, numEntries));
|
---|
| 60 | }
|
---|
| 61 |
|
---|
| 62 | bool TextLevelInfo::Write (FILE *f) const {
|
---|
| 63 | return (WriteUCArray (f, levelTag) &&
|
---|
| 64 | WriteUL (f, textIdxPtr) &&
|
---|
| 65 | WriteUL (f, numEntries));
|
---|
| 66 | }
|
---|
| 67 |
|
---|
| 68 | ostream &operator<<(ostream &s, const TextLevelInfo &l) {
|
---|
| 69 | s << " Tag: \"" << l.levelTag << "\"\n"
|
---|
| 70 | << " textIdxPtr: " << l.textIdxPtr << "\n"
|
---|
| 71 | << " numEntries: " << l.numEntries << "\n\n";
|
---|
| 72 |
|
---|
| 73 | return s;
|
---|
| 74 | }
|
---|
| 75 |
|
---|
| 76 |
|
---|
| 77 | FTextLevel::FTextLevel () {
|
---|
| 78 | Clear();
|
---|
| 79 | }
|
---|
| 80 |
|
---|
| 81 | void FTextLevel::Clear () {
|
---|
| 82 | levelInfo.erase (levelInfo.begin(), levelInfo.end());
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | bool FTextLevel::Read (FILE *f) {
|
---|
| 86 | // read in the array size
|
---|
| 87 | unsigned long arrSize = 0;
|
---|
| 88 | if (!ReadVarLenUL (f, arrSize)) return false;
|
---|
| 89 |
|
---|
| 90 | TextLevelInfo thisLevel;
|
---|
| 91 | while (arrSize > 0) {
|
---|
| 92 | // read and insert the next level information
|
---|
| 93 | if (!thisLevel.Read (f)) return false;
|
---|
| 94 | levelInfo[thisLevel.levelTag] = thisLevel;
|
---|
| 95 |
|
---|
[9613] | 96 | --arrSize;
|
---|
[3365] | 97 | }
|
---|
| 98 |
|
---|
| 99 | return true;
|
---|
| 100 | }
|
---|
| 101 |
|
---|
| 102 | bool FTextLevel::Write (FILE *f) const {
|
---|
| 103 | // write out the array size
|
---|
| 104 | if (!WriteVarLenUL (f, levelInfo.size())) return false;
|
---|
| 105 |
|
---|
| 106 | // write out each level info
|
---|
| 107 | TextLevelInfoMap::const_iterator here = levelInfo.begin();
|
---|
| 108 | TextLevelInfoMap::const_iterator end = levelInfo.end();
|
---|
| 109 | while (here != end) {
|
---|
| 110 | if (!(*here).second.Write (f)) return false;
|
---|
| 111 |
|
---|
[9613] | 112 | ++here;
|
---|
[3365] | 113 | }
|
---|
| 114 |
|
---|
| 115 | return true;
|
---|
| 116 | }
|
---|
| 117 |
|
---|
| 118 | ostream &operator<<(ostream &s, const FTextLevel &l) {
|
---|
| 119 | TextLevelInfoMap::const_iterator here = l.levelInfo.begin ();
|
---|
| 120 | TextLevelInfoMap::const_iterator end = l.levelInfo.end ();
|
---|
| 121 | while (here != end) {
|
---|
| 122 | s << (*here).second;
|
---|
[9613] | 123 | ++here;
|
---|
[3365] | 124 | }
|
---|
| 125 |
|
---|
| 126 | return s;
|
---|
| 127 | }
|
---|
| 128 |
|
---|
| 129 |
|
---|
| 130 | TextIdx::TextIdx () {
|
---|
| 131 | Clear ();
|
---|
| 132 | }
|
---|
| 133 |
|
---|
| 134 | void TextIdx::Clear () {
|
---|
| 135 | start.Clear();
|
---|
| 136 | end.Clear();
|
---|
| 137 | which = 0;
|
---|
| 138 | }
|
---|
| 139 |
|
---|
| 140 | #define TEXTIDXLEN (BITADDRLEN*2 + sizeof(char))
|
---|
| 141 |
|
---|
| 142 | bool TextIdx::Read (FILE *f, const TextLevelInfo &levelInfo,
|
---|
| 143 | unsigned long docNum) {
|
---|
| 144 | if (!SeekTextIdx (f, levelInfo, docNum)) return false;
|
---|
| 145 | return Read (f);
|
---|
| 146 | }
|
---|
| 147 |
|
---|
| 148 | bool TextIdx::Read (FILE *f) {
|
---|
| 149 | return (start.Read (f) && end.Read (f) && ReadUC (f, which));
|
---|
| 150 | }
|
---|
| 151 |
|
---|
| 152 | bool TextIdx::Write (FILE *f) const {
|
---|
| 153 | return (start.Write (f) && end.Write (f) && WriteUC (f, which));
|
---|
| 154 | }
|
---|
| 155 |
|
---|
| 156 | ostream &operator<<(ostream &s, const TextIdx &t) {
|
---|
| 157 | s << "start byte: " << t.start.byte << "\n";
|
---|
| 158 | s << "start bit: " << (int)t.start.bit << "\n";
|
---|
| 159 | s << "end byte: " << t.end.byte << "\n";
|
---|
| 160 | s << "end bit: " << (int)t.end.bit << "\n";
|
---|
| 161 | s << "which: " << (int)t.which << "\n\n";
|
---|
| 162 |
|
---|
| 163 | return s;
|
---|
| 164 | }
|
---|
| 165 |
|
---|
| 166 |
|
---|
| 167 | bool SeekTextIdx (FILE *f, const TextLevelInfo &levelInfo,
|
---|
| 168 | unsigned long docNum) {
|
---|
| 169 | if (docNum == 0 || docNum > levelInfo.numEntries) return false;
|
---|
| 170 |
|
---|
| 171 | unsigned long seekPos = levelInfo.textIdxPtr + (docNum-1) * TEXTIDXLEN;
|
---|
| 172 | if (fseek (f, seekPos, SEEK_SET) != 0) return false;
|
---|
| 173 |
|
---|
| 174 | return true;
|
---|
| 175 | }
|
---|
| 176 |
|
---|
| 177 |
|
---|
| 178 | bool ReadTextIdxArray (FILE *f, TextIdxArray &a, unsigned long arrSize) {
|
---|
| 179 | // clear the array
|
---|
| 180 | a.erase (a.begin(), a.end());
|
---|
| 181 |
|
---|
| 182 | // read in each element
|
---|
| 183 | TextIdx idx;
|
---|
| 184 | while (arrSize > 0) {
|
---|
| 185 | if (!idx.Read(f)) return false;
|
---|
| 186 | a.push_back (idx);
|
---|
[9613] | 187 | --arrSize;
|
---|
[3365] | 188 | }
|
---|
| 189 |
|
---|
| 190 | return true;
|
---|
| 191 | }
|
---|
| 192 |
|
---|
| 193 | bool WriteTextIdxArray (FILE *f, const TextIdxArray &a) {
|
---|
| 194 | TextIdxArray::const_iterator here = a.begin();
|
---|
| 195 | TextIdxArray::const_iterator end = a.end();
|
---|
| 196 | while (here != end) {
|
---|
| 197 | if (!(*here).Write(f)) return false;
|
---|
[9613] | 198 | ++here;
|
---|
[3365] | 199 | }
|
---|
| 200 |
|
---|
| 201 | return true;
|
---|
| 202 | }
|
---|
| 203 |
|
---|
| 204 |
|
---|
| 205 | CompressTextInfo::CompressTextInfo () {
|
---|
| 206 | Clear();
|
---|
| 207 | }
|
---|
| 208 |
|
---|
| 209 | void CompressTextInfo::Clear () {
|
---|
| 210 | ResetStart();
|
---|
| 211 | docPtrs.erase (docPtrs.begin(), docPtrs.end());
|
---|
| 212 | }
|
---|
| 213 |
|
---|
| 214 | void CompressTextInfo::ResetStart () {
|
---|
| 215 | inDoc = false;
|
---|
| 216 | start.Clear();
|
---|
| 217 | which = 0;
|
---|
| 218 | }
|
---|
| 219 |
|
---|
| 220 | void CompressTextInfo::SetStart (unsigned long startPos,
|
---|
| 221 | unsigned char startBit,
|
---|
| 222 | unsigned char startWhich) {
|
---|
| 223 | // place an imaginary end tag if needed
|
---|
| 224 | // note: the document tag always needs a closing tag
|
---|
| 225 | if (inDoc) SetEnd (startPos, startBit);
|
---|
| 226 |
|
---|
| 227 | // remember start of level
|
---|
| 228 | inDoc = true;
|
---|
| 229 | start.byte = startPos;
|
---|
| 230 | start.bit = startBit;
|
---|
| 231 | which = startWhich;
|
---|
| 232 | }
|
---|
| 233 |
|
---|
| 234 | void CompressTextInfo::SetEnd (unsigned long endPos,
|
---|
| 235 | unsigned char endBit) {
|
---|
| 236 | if (inDoc) {
|
---|
| 237 | // add completed entry to list of ptrs
|
---|
| 238 | TextIdx textIdx;
|
---|
| 239 | textIdx.start = start;
|
---|
| 240 | textIdx.end.byte = endPos;
|
---|
| 241 | textIdx.end.bit = endBit;
|
---|
| 242 | textIdx.which = which;
|
---|
| 243 | docPtrs.push_back (textIdx);
|
---|
| 244 |
|
---|
| 245 | // reset this entry
|
---|
| 246 | ResetStart();
|
---|
| 247 | }
|
---|
| 248 | }
|
---|
| 249 |
|
---|