| 2 |
- |
1 |
/*
|
|
|
2 |
* findtext.c
|
|
|
3 |
* Copyright (C) 1998-2004 A.J. van Os; Released under GNU GPL
|
|
|
4 |
*
|
|
|
5 |
* Description:
|
|
|
6 |
* Find the blocks that contain the text of MS Word files
|
|
|
7 |
*/
|
|
|
8 |
|
|
|
9 |
#include <stdio.h>
|
|
|
10 |
#include <stdlib.h>
|
|
|
11 |
#include "antiword.h"
|
|
|
12 |
|
|
|
13 |
|
|
|
14 |
/*
|
|
|
15 |
* bAddTextBlocks - Add the blocks to the text block list
|
|
|
16 |
*
|
|
|
17 |
* Returns TRUE when successful, FALSE if not
|
|
|
18 |
*/
|
|
|
19 |
BOOL
|
|
|
20 |
bAddTextBlocks(ULONG ulCharPosFirst, ULONG ulTotalLength,
|
|
|
21 |
BOOL bUsesUnicode, USHORT usPropMod,
|
|
|
22 |
ULONG ulStartBlock, const ULONG *aulBBD, size_t tBBDLen)
|
|
|
23 |
{
|
|
|
24 |
text_block_type tTextBlock;
|
|
|
25 |
ULONG ulCharPos, ulOffset, ulIndex;
|
|
|
26 |
long lToGo;
|
|
|
27 |
|
|
|
28 |
fail(ulTotalLength > (ULONG)LONG_MAX / 2);
|
|
|
29 |
fail(ulStartBlock > MAX_BLOCKNUMBER && ulStartBlock != END_OF_CHAIN);
|
|
|
30 |
fail(aulBBD == NULL);
|
|
|
31 |
|
|
|
32 |
NO_DBG_HEX(ulCharPosFirst);
|
|
|
33 |
NO_DBG_DEC(ulTotalLength);
|
|
|
34 |
|
|
|
35 |
if (bUsesUnicode) {
|
|
|
36 |
/* One character equals two bytes */
|
|
|
37 |
NO_DBG_MSG("Uses Unicode");
|
|
|
38 |
lToGo = (long)ulTotalLength * 2;
|
|
|
39 |
} else {
|
|
|
40 |
/* One character equals one byte */
|
|
|
41 |
NO_DBG_MSG("Uses ASCII");
|
|
|
42 |
lToGo = (long)ulTotalLength;
|
|
|
43 |
}
|
|
|
44 |
|
|
|
45 |
ulCharPos = ulCharPosFirst;
|
|
|
46 |
ulOffset = ulCharPosFirst;
|
|
|
47 |
for (ulIndex = ulStartBlock;
|
|
|
48 |
ulIndex != END_OF_CHAIN && lToGo > 0;
|
|
|
49 |
ulIndex = aulBBD[ulIndex]) {
|
|
|
50 |
if (ulIndex >= (ULONG)tBBDLen) {
|
|
|
51 |
DBG_DEC(ulIndex);
|
|
|
52 |
DBG_DEC(tBBDLen);
|
|
|
53 |
werr(1, "The Big Block Depot is damaged");
|
|
|
54 |
}
|
|
|
55 |
if (ulOffset >= BIG_BLOCK_SIZE) {
|
|
|
56 |
ulOffset -= BIG_BLOCK_SIZE;
|
|
|
57 |
continue;
|
|
|
58 |
}
|
|
|
59 |
tTextBlock.ulFileOffset =
|
|
|
60 |
(ulIndex + 1) * BIG_BLOCK_SIZE + ulOffset;
|
|
|
61 |
tTextBlock.ulCharPos = ulCharPos;
|
|
|
62 |
tTextBlock.ulLength = min(BIG_BLOCK_SIZE - ulOffset,
|
|
|
63 |
(ULONG)lToGo);
|
|
|
64 |
tTextBlock.bUsesUnicode = bUsesUnicode;
|
|
|
65 |
tTextBlock.usPropMod = usPropMod;
|
|
|
66 |
ulOffset = 0;
|
|
|
67 |
if (!bAdd2TextBlockList(&tTextBlock)) {
|
|
|
68 |
DBG_HEX(tTextBlock.ulFileOffset);
|
|
|
69 |
DBG_HEX(tTextBlock.ulCharPos);
|
|
|
70 |
DBG_DEC(tTextBlock.ulLength);
|
|
|
71 |
DBG_DEC(tTextBlock.bUsesUnicode);
|
|
|
72 |
DBG_DEC(tTextBlock.usPropMod);
|
|
|
73 |
return FALSE;
|
|
|
74 |
}
|
|
|
75 |
ulCharPos += tTextBlock.ulLength;
|
|
|
76 |
lToGo -= (long)tTextBlock.ulLength;
|
|
|
77 |
}
|
|
|
78 |
DBG_DEC_C(lToGo != 0, lToGo);
|
|
|
79 |
return lToGo == 0;
|
|
|
80 |
} /* end of bAddTextBlocks */
|
|
|
81 |
|
|
|
82 |
/*
|
|
|
83 |
* bGet6DocumentText - make a list of the text blocks of Word 6/7 files
|
|
|
84 |
*
|
|
|
85 |
* Code for "fast saved" files.
|
|
|
86 |
*
|
|
|
87 |
* Returns TRUE when successful, FALSE if not
|
|
|
88 |
*/
|
|
|
89 |
BOOL
|
|
|
90 |
bGet6DocumentText(FILE *pFile, BOOL bUsesUnicode, ULONG ulStartBlock,
|
|
|
91 |
const ULONG *aulBBD, size_t tBBDLen, const UCHAR *aucHeader)
|
|
|
92 |
{
|
|
|
93 |
UCHAR *aucBuffer;
|
|
|
94 |
ULONG ulBeginTextInfo, ulTextOffset, ulTotLength;
|
|
|
95 |
size_t tTextInfoLen;
|
|
|
96 |
int iIndex, iType, iOff, iLen, iPieces;
|
|
|
97 |
USHORT usPropMod;
|
|
|
98 |
|
|
|
99 |
DBG_MSG("bGet6DocumentText");
|
|
|
100 |
|
|
|
101 |
fail(pFile == NULL);
|
|
|
102 |
fail(aulBBD == NULL);
|
|
|
103 |
fail(aucHeader == NULL);
|
|
|
104 |
|
|
|
105 |
ulBeginTextInfo = ulGetLong(0x160, aucHeader); /* fcClx */
|
|
|
106 |
DBG_HEX(ulBeginTextInfo);
|
|
|
107 |
tTextInfoLen = (size_t)ulGetLong(0x164, aucHeader); /* lcbClx */
|
|
|
108 |
DBG_DEC(tTextInfoLen);
|
|
|
109 |
|
|
|
110 |
aucBuffer = xmalloc(tTextInfoLen);
|
|
|
111 |
if (!bReadBuffer(pFile, ulStartBlock,
|
|
|
112 |
aulBBD, tBBDLen, BIG_BLOCK_SIZE,
|
|
|
113 |
aucBuffer, ulBeginTextInfo, tTextInfoLen)) {
|
|
|
114 |
aucBuffer = xfree(aucBuffer);
|
|
|
115 |
return FALSE;
|
|
|
116 |
}
|
|
|
117 |
NO_DBG_PRINT_BLOCK(aucBuffer, tTextInfoLen);
|
|
|
118 |
|
|
|
119 |
iOff = 0;
|
|
|
120 |
while ((size_t)iOff < tTextInfoLen) {
|
|
|
121 |
iType = (int)ucGetByte(iOff, aucBuffer);
|
|
|
122 |
iOff++;
|
|
|
123 |
if (iType == 0) {
|
|
|
124 |
DBG_FIXME();
|
|
|
125 |
iOff++;
|
|
|
126 |
continue;
|
|
|
127 |
}
|
|
|
128 |
if (iType == 1) {
|
|
|
129 |
iLen = (int)usGetWord(iOff, aucBuffer);
|
|
|
130 |
vAdd2PropModList(aucBuffer + iOff);
|
|
|
131 |
iOff += iLen + 2;
|
|
|
132 |
continue;
|
|
|
133 |
}
|
|
|
134 |
if (iType != 2) {
|
|
|
135 |
werr(0, "Unknown type of 'fastsaved' format");
|
|
|
136 |
aucBuffer = xfree(aucBuffer);
|
|
|
137 |
return FALSE;
|
|
|
138 |
}
|
|
|
139 |
/* Type 2 */
|
|
|
140 |
iLen = (int)usGetWord(iOff, aucBuffer);
|
|
|
141 |
NO_DBG_DEC(iLen);
|
|
|
142 |
iOff += 4;
|
|
|
143 |
iPieces = (iLen - 4) / 12;
|
|
|
144 |
DBG_DEC(iPieces);
|
|
|
145 |
for (iIndex = 0; iIndex < iPieces; iIndex++) {
|
|
|
146 |
ulTextOffset = ulGetLong(
|
|
|
147 |
iOff + (iPieces + 1) * 4 + iIndex * 8 + 2,
|
|
|
148 |
aucBuffer);
|
|
|
149 |
usPropMod = usGetWord(
|
|
|
150 |
iOff + (iPieces + 1) * 4 + iIndex * 8 + 6,
|
|
|
151 |
aucBuffer);
|
|
|
152 |
ulTotLength = ulGetLong(iOff + (iIndex + 1) * 4,
|
|
|
153 |
aucBuffer) -
|
|
|
154 |
ulGetLong(iOff + iIndex * 4,
|
|
|
155 |
aucBuffer);
|
|
|
156 |
NO_DBG_HEX_C(usPropMod != 0, usPropMod);
|
|
|
157 |
if (!bAddTextBlocks(ulTextOffset, ulTotLength,
|
|
|
158 |
bUsesUnicode, usPropMod,
|
|
|
159 |
ulStartBlock,
|
|
|
160 |
aulBBD, tBBDLen)) {
|
|
|
161 |
aucBuffer = xfree(aucBuffer);
|
|
|
162 |
return FALSE;
|
|
|
163 |
}
|
|
|
164 |
}
|
|
|
165 |
break;
|
|
|
166 |
}
|
|
|
167 |
aucBuffer = xfree(aucBuffer);
|
|
|
168 |
return TRUE;
|
|
|
169 |
} /* end of bGet6DocumentText */
|
|
|
170 |
|
|
|
171 |
/*
|
|
|
172 |
* bGet8DocumentText - make a list of the text blocks of Word 8/97 files
|
|
|
173 |
*
|
|
|
174 |
* Returns TRUE when successful, FALSE if not
|
|
|
175 |
*/
|
|
|
176 |
BOOL
|
|
|
177 |
bGet8DocumentText(FILE *pFile, const pps_info_type *pPPS,
|
|
|
178 |
const ULONG *aulBBD, size_t tBBDLen,
|
|
|
179 |
const ULONG *aulSBD, size_t tSBDLen,
|
|
|
180 |
const UCHAR *aucHeader)
|
|
|
181 |
{
|
|
|
182 |
const ULONG *aulBlockDepot;
|
|
|
183 |
UCHAR *aucBuffer;
|
|
|
184 |
ULONG ulTextOffset, ulBeginTextInfo;
|
|
|
185 |
ULONG ulTotLength, ulLen;
|
|
|
186 |
long lIndex, lPieces, lOff;
|
|
|
187 |
size_t tTextInfoLen, tBlockDepotLen, tBlockSize;
|
|
|
188 |
int iType, iLen;
|
|
|
189 |
BOOL bUsesUnicode;
|
|
|
190 |
USHORT usPropMod;
|
|
|
191 |
|
|
|
192 |
DBG_MSG("bGet8DocumentText");
|
|
|
193 |
|
|
|
194 |
fail(pFile == NULL || pPPS == NULL);
|
|
|
195 |
fail(aulBBD == NULL || aulSBD == NULL);
|
|
|
196 |
fail(aucHeader == NULL);
|
|
|
197 |
|
|
|
198 |
ulBeginTextInfo = ulGetLong(0x1a2, aucHeader); /* fcClx */
|
|
|
199 |
DBG_HEX(ulBeginTextInfo);
|
|
|
200 |
tTextInfoLen = (size_t)ulGetLong(0x1a6, aucHeader); /* lcbClx */
|
|
|
201 |
DBG_DEC(tTextInfoLen);
|
|
|
202 |
|
|
|
203 |
DBG_DEC(pPPS->tTable.ulSB);
|
|
|
204 |
DBG_HEX(pPPS->tTable.ulSize);
|
|
|
205 |
if (pPPS->tTable.ulSize == 0) {
|
|
|
206 |
return FALSE;
|
|
|
207 |
}
|
|
|
208 |
|
|
|
209 |
if (pPPS->tTable.ulSize < MIN_SIZE_FOR_BBD_USE) {
|
|
|
210 |
/* Use the Small Block Depot */
|
|
|
211 |
aulBlockDepot = aulSBD;
|
|
|
212 |
tBlockDepotLen = tSBDLen;
|
|
|
213 |
tBlockSize = SMALL_BLOCK_SIZE;
|
|
|
214 |
} else {
|
|
|
215 |
/* Use the Big Block Depot */
|
|
|
216 |
aulBlockDepot = aulBBD;
|
|
|
217 |
tBlockDepotLen = tBBDLen;
|
|
|
218 |
tBlockSize = BIG_BLOCK_SIZE;
|
|
|
219 |
}
|
|
|
220 |
aucBuffer = xmalloc(tTextInfoLen);
|
|
|
221 |
if (!bReadBuffer(pFile, pPPS->tTable.ulSB,
|
|
|
222 |
aulBlockDepot, tBlockDepotLen, tBlockSize,
|
|
|
223 |
aucBuffer, ulBeginTextInfo, tTextInfoLen)) {
|
|
|
224 |
aucBuffer = xfree(aucBuffer);
|
|
|
225 |
return FALSE;
|
|
|
226 |
}
|
|
|
227 |
NO_DBG_PRINT_BLOCK(aucBuffer, tTextInfoLen);
|
|
|
228 |
|
|
|
229 |
lOff = 0;
|
|
|
230 |
while (lOff < (long)tTextInfoLen) {
|
|
|
231 |
iType = (int)ucGetByte(lOff, aucBuffer);
|
|
|
232 |
lOff++;
|
|
|
233 |
if (iType == 0) {
|
|
|
234 |
DBG_FIXME();
|
|
|
235 |
lOff++;
|
|
|
236 |
continue;
|
|
|
237 |
}
|
|
|
238 |
if (iType == 1) {
|
|
|
239 |
iLen = (int)usGetWord(lOff, aucBuffer);
|
|
|
240 |
vAdd2PropModList(aucBuffer + lOff);
|
|
|
241 |
lOff += (long)iLen + 2;
|
|
|
242 |
continue;
|
|
|
243 |
}
|
|
|
244 |
if (iType != 2) {
|
|
|
245 |
werr(0, "Unknown type of 'fastsaved' format");
|
|
|
246 |
aucBuffer = xfree(aucBuffer);
|
|
|
247 |
return FALSE;
|
|
|
248 |
}
|
|
|
249 |
/* Type 2 */
|
|
|
250 |
ulLen = ulGetLong(lOff, aucBuffer);
|
|
|
251 |
if (ulLen < 4) {
|
|
|
252 |
DBG_DEC(ulLen);
|
|
|
253 |
return FALSE;
|
|
|
254 |
}
|
|
|
255 |
lOff += 4;
|
|
|
256 |
lPieces = (long)((ulLen - 4) / 12);
|
|
|
257 |
DBG_DEC(lPieces);
|
|
|
258 |
for (lIndex = 0; lIndex < lPieces; lIndex++) {
|
|
|
259 |
ulTextOffset = ulGetLong(
|
|
|
260 |
lOff + (lPieces + 1) * 4 + lIndex * 8 + 2,
|
|
|
261 |
aucBuffer);
|
|
|
262 |
usPropMod = usGetWord(
|
|
|
263 |
lOff + (lPieces + 1) * 4 + lIndex * 8 + 6,
|
|
|
264 |
aucBuffer);
|
|
|
265 |
ulTotLength = ulGetLong(lOff + (lIndex + 1) * 4,
|
|
|
266 |
aucBuffer) -
|
|
|
267 |
ulGetLong(lOff + lIndex * 4,
|
|
|
268 |
aucBuffer);
|
|
|
269 |
if ((ulTextOffset & BIT(30)) == 0) {
|
|
|
270 |
bUsesUnicode = TRUE;
|
|
|
271 |
} else {
|
|
|
272 |
bUsesUnicode = FALSE;
|
|
|
273 |
ulTextOffset &= ~BIT(30);
|
|
|
274 |
ulTextOffset /= 2;
|
|
|
275 |
}
|
|
|
276 |
NO_DBG_HEX_C(usPropMod != 0, usPropMod);
|
|
|
277 |
if (!bAddTextBlocks(ulTextOffset, ulTotLength,
|
|
|
278 |
bUsesUnicode, usPropMod,
|
|
|
279 |
pPPS->tWordDocument.ulSB,
|
|
|
280 |
aulBBD, tBBDLen)) {
|
|
|
281 |
aucBuffer = xfree(aucBuffer);
|
|
|
282 |
return FALSE;
|
|
|
283 |
}
|
|
|
284 |
}
|
|
|
285 |
break;
|
|
|
286 |
}
|
|
|
287 |
aucBuffer = xfree(aucBuffer);
|
|
|
288 |
return TRUE;
|
|
|
289 |
} /* end of bGet8DocumentText */
|