//
// XRef.cc
//
-// Copyright 1996 Derek B. Noonburg
+// Copyright 1996-2003 Glyph & Cog, LLC
//
//========================================================================
-#ifdef __GNUC__
+#include <aconf.h>
+
+#ifdef USE_GCC_PRAGMAS
#pragma implementation
#endif
#include "Lexer.h"
#include "Parser.h"
#include "Dict.h"
-#ifndef NO_DECRYPTION
-#include "Decrypt.h"
-#endif
#include "Error.h"
+#include "ErrorCodes.h"
#include "XRef.h"
//------------------------------------------------------------------------
#define xrefSearchSize 1024 // read this many bytes at end of file
// to look for 'startxref'
-#ifndef NO_DECRYPTION
//------------------------------------------------------------------------
// Permission bits
//------------------------------------------------------------------------
#define permCopy (1<<4)
#define permNotes (1<<5)
#define defPermFlags 0xfffc
-#endif
//------------------------------------------------------------------------
-// The global xref table
+// ObjectStream
//------------------------------------------------------------------------
-XRef *xref = NULL;
+class ObjectStream {
+public:
+
+ // Create an object stream, using object number <objStrNum>,
+ // generation 0.
+ ObjectStream(XRef *xref, int objStrNumA);
+
+ ~ObjectStream();
+
+ // Return the object number of this object stream.
+ int getObjStrNum() { return objStrNum; }
+
+ // Get the <objIdx>th object from this stream, which should be
+ // object number <objNum>, generation 0.
+ Object *getObject(int objIdx, int objNum, Object *obj);
+
+private:
+
+ int objStrNum; // object number of the object stream
+ int nObjects; // number of objects in the stream
+ Object *objs; // the objects (length = nObjects)
+ int *objNums; // the object numbers (length = nObjects)
+};
+
+ObjectStream::ObjectStream(XRef *xref, int objStrNumA) {
+ Stream *str;
+ Parser *parser;
+ int *offsets;
+ Object objStr, obj1, obj2;
+ int first, i;
+
+ objStrNum = objStrNumA;
+ nObjects = 0;
+ objs = NULL;
+ objNums = NULL;
+
+ if (!xref->fetch(objStrNum, 0, &objStr)->isStream()) {
+ goto err1;
+ }
+
+ if (!objStr.streamGetDict()->lookup("N", &obj1)->isInt()) {
+ obj1.free();
+ goto err1;
+ }
+ nObjects = obj1.getInt();
+ obj1.free();
+ if (nObjects <= 0) {
+ goto err1;
+ }
+
+ if (!objStr.streamGetDict()->lookup("First", &obj1)->isInt()) {
+ obj1.free();
+ goto err1;
+ }
+ first = obj1.getInt();
+ obj1.free();
+ if (first < 0) {
+ goto err1;
+ }
+
+ objs = new Object[nObjects];
+ objNums = (int *)gmallocn(nObjects, sizeof(int));
+ offsets = (int *)gmallocn(nObjects, sizeof(int));
+
+ // parse the header: object numbers and offsets
+ objStr.streamReset();
+ obj1.initNull();
+ str = new EmbedStream(objStr.getStream(), &obj1, gTrue, first);
+ parser = new Parser(xref, new Lexer(xref, str));
+ for (i = 0; i < nObjects; ++i) {
+ parser->getObj(&obj1);
+ parser->getObj(&obj2);
+ if (!obj1.isInt() || !obj2.isInt()) {
+ obj1.free();
+ obj2.free();
+ delete parser;
+ gfree(offsets);
+ goto err1;
+ }
+ objNums[i] = obj1.getInt();
+ offsets[i] = obj2.getInt();
+ obj1.free();
+ obj2.free();
+ if (objNums[i] < 0 || offsets[i] < 0 ||
+ (i > 0 && offsets[i] < offsets[i-1])) {
+ delete parser;
+ gfree(offsets);
+ goto err1;
+ }
+ }
+ while (str->getChar() != EOF) ;
+ delete parser;
+
+ // skip to the first object - this shouldn't be necessary because
+ // the First key is supposed to be equal to offsets[0], but just in
+ // case...
+ for (i = first; i < offsets[0]; ++i) {
+ objStr.getStream()->getChar();
+ }
+
+ // parse the objects
+ for (i = 0; i < nObjects; ++i) {
+ obj1.initNull();
+ if (i == nObjects - 1) {
+ str = new EmbedStream(objStr.getStream(), &obj1, gFalse, 0);
+ } else {
+ str = new EmbedStream(objStr.getStream(), &obj1, gTrue,
+ offsets[i+1] - offsets[i]);
+ }
+ parser = new Parser(xref, new Lexer(xref, str));
+ parser->getObj(&objs[i]);
+ while (str->getChar() != EOF) ;
+ delete parser;
+ }
+
+ gfree(offsets);
+
+ err1:
+ objStr.free();
+ return;
+}
+
+ObjectStream::~ObjectStream() {
+ int i;
+
+ if (objs) {
+ for (i = 0; i < nObjects; ++i) {
+ objs[i].free();
+ }
+ delete[] objs;
+ }
+ gfree(objNums);
+}
+
+Object *ObjectStream::getObject(int objIdx, int objNum, Object *obj) {
+ if (objIdx < 0 || objIdx >= nObjects || objNum != objNums[objIdx]) {
+ return obj->initNull();
+ }
+ return objs[objIdx].copy(obj);
+}
//------------------------------------------------------------------------
// XRef
//------------------------------------------------------------------------
-XRef::XRef(BaseStream *str, GString *userPassword) {
- XRef *oldXref;
- int pos;
- int i;
+XRef::XRef(BaseStream *strA) {
+ Guint pos;
+ Object obj;
ok = gTrue;
+ errCode = errNone;
size = 0;
entries = NULL;
streamEnds = NULL;
streamEndsLen = 0;
+ objStr = NULL;
- // get rid of old xref (otherwise it will try to fetch the Root object
- // in the new document, using the old xref)
- oldXref = xref;
- xref = NULL;
+ encrypted = gFalse;
+ permFlags = defPermFlags;
+ ownerPasswordOk = gFalse;
// read the trailer
- this->str = str;
+ str = strA;
start = str->getStart();
- pos = readTrailer();
+ pos = getStartXref();
- // if there was a problem with the trailer,
- // try to reconstruct the xref table
+ // if there was a problem with the 'startxref' position, try to
+ // reconstruct the xref table
if (pos == 0) {
if (!(ok = constructXRef())) {
- xref = oldXref;
+ errCode = errDamaged;
return;
}
- // trailer is ok - read the xref table
+ // read the xref table
} else {
- entries = (XRefEntry *)gmalloc(size * sizeof(XRefEntry));
- for (i = 0; i < size; ++i) {
- entries[i].offset = -1;
- entries[i].used = gFalse;
- }
while (readXRef(&pos)) ;
// if there was a problem with the xref table,
// try to reconstruct it
if (!ok) {
- gfree(entries);
- size = 0;
- entries = NULL;
if (!(ok = constructXRef())) {
- xref = oldXref;
+ errCode = errDamaged;
return;
}
}
}
- // set up new xref table
- xref = this;
-
- // check for encryption
-#ifndef NO_DECRYPTION
- encrypted = gFalse;
-#endif
- if (checkEncrypted(userPassword)) {
- ok = gFalse;
- xref = oldXref;
- return;
+ // get the root dictionary (catalog) object
+ trailerDict.dictLookupNF("Root", &obj);
+ if (obj.isRef()) {
+ rootNum = obj.getRefNum();
+ rootGen = obj.getRefGen();
+ obj.free();
+ } else {
+ obj.free();
+ if (!(ok = constructXRef())) {
+ errCode = errDamaged;
+ return;
+ }
}
+
+ // now set the trailer dictionary's xref pointer so we can fetch
+ // indirect objects from it
+ trailerDict.getDict()->setXRef(this);
}
XRef::~XRef() {
if (streamEnds) {
gfree(streamEnds);
}
+ if (objStr) {
+ delete objStr;
+ }
}
-// Read startxref position, xref table size, and root. Returns
-// first xref position.
-int XRef::readTrailer() {
- Parser *parser;
- Object obj;
+// Read the 'startxref' position.
+Guint XRef::getStartXref() {
char buf[xrefSearchSize+1];
- int n, pos, pos1;
char *p;
- int c;
- int i;
+ int c, n, i;
// read last xrefSearchSize bytes
- str->setPos(-xrefSearchSize);
+ str->setPos(xrefSearchSize, -1);
for (n = 0; n < xrefSearchSize; ++n) {
- if ((c = str->getChar()) == EOF)
+ if ((c = str->getChar()) == EOF) {
break;
+ }
buf[n] = c;
}
buf[n] = '\0';
// find startxref
for (i = n - 9; i >= 0; --i) {
- if (!strncmp(&buf[i], "startxref", 9))
+ if (!strncmp(&buf[i], "startxref", 9)) {
break;
+ }
}
- if (i < 0)
+ if (i < 0) {
return 0;
+ }
for (p = &buf[i+9]; isspace(*p); ++p) ;
- pos = lastXRefPos = atoi(p);
-
- // find trailer dict by looking after first xref table
- // (NB: we can't just use the trailer dict at the end of the file --
- // this won't work for linearized files.)
- str->setPos(start + pos);
- for (i = 0; i < 4; ++i)
- buf[i] = str->getChar();
- if (strncmp(buf, "xref", 4))
- return 0;
- pos1 = pos + 4;
- while (1) {
- str->setPos(start + pos1);
- for (i = 0; i < 35; ++i) {
- if ((c = str->getChar()) == EOF)
- return 0;
- buf[i] = c;
- }
- if (!strncmp(buf, "trailer", 7))
- break;
- p = buf;
- while (isspace(*p)) ++p;
- while ('0' <= *p && *p <= '9') ++p;
- while (isspace(*p)) ++p;
- n = atoi(p);
- while ('0' <= *p && *p <= '9') ++p;
- while (isspace(*p)) ++p;
- if (p == buf)
- return 0;
- pos1 += (p - buf) + n * 20;
- }
- pos1 += 7;
-
- // read trailer dict
+ lastXRefPos = strToUnsigned(p);
+
+ return lastXRefPos;
+}
+
+// Read one xref table section. Also reads the associated trailer
+// dictionary, and returns the prev pointer (if any).
+GBool XRef::readXRef(Guint *pos) {
+ Parser *parser;
+ Object obj;
+ GBool more;
+
+ // start up a parser, parse one token
obj.initNull();
- parser = new Parser(new Lexer(str->makeSubStream(start + pos1, -1, &obj)));
- parser->getObj(&trailerDict);
- if (trailerDict.isDict()) {
- trailerDict.dictLookupNF("Size", &obj);
- if (obj.isInt())
- size = obj.getInt();
- else
- pos = 0;
+ parser = new Parser(NULL,
+ new Lexer(NULL,
+ str->makeSubStream(start + *pos, gFalse, 0, &obj)));
+ parser->getObj(&obj);
+
+ // parse an old-style xref table
+ if (obj.isCmd("xref")) {
obj.free();
- trailerDict.dictLookupNF("Root", &obj);
- if (obj.isRef()) {
- rootNum = obj.getRefNum();
- rootGen = obj.getRefGen();
- } else {
- pos = 0;
+ more = readXRefTable(parser, pos);
+
+ // parse an xref stream
+ } else if (obj.isInt()) {
+ obj.free();
+ if (!parser->getObj(&obj)->isInt()) {
+ goto err1;
+ }
+ obj.free();
+ if (!parser->getObj(&obj)->isCmd("obj")) {
+ goto err1;
+ }
+ obj.free();
+ if (!parser->getObj(&obj)->isStream()) {
+ goto err1;
}
+ more = readXRefStream(obj.getStream(), pos);
obj.free();
+
} else {
- pos = 0;
+ goto err1;
}
+
delete parser;
+ return more;
- // return first xref position
- return pos;
+ err1:
+ obj.free();
+ delete parser;
+ ok = gFalse;
+ return gFalse;
}
-// Read an xref table and the prev pointer from the trailer.
-GBool XRef::readXRef(int *pos) {
- Parser *parser;
- Object obj, obj2;
- char s[20];
+GBool XRef::readXRefTable(Parser *parser, Guint *pos) {
+ XRefEntry entry;
GBool more;
- int first, n, i, j;
- int c;
-
- // seek to xref in stream
- str->setPos(start + *pos);
-
- // make sure it's an xref table
- while ((c = str->getChar()) != EOF && isspace(c)) ;
- s[0] = (char)c;
- s[1] = (char)str->getChar();
- s[2] = (char)str->getChar();
- s[3] = (char)str->getChar();
- if (!(s[0] == 'x' && s[1] == 'r' && s[2] == 'e' && s[3] == 'f'))
- goto err2;
-
- // read xref
+ Object obj, obj2;
+ Guint pos2;
+ int first, n, newSize, i;
+
while (1) {
- while ((c = str->lookChar()) != EOF && isspace(c))
- str->getChar();
- if (c == 't')
+ parser->getObj(&obj);
+ if (obj.isCmd("trailer")) {
+ obj.free();
break;
- for (i = 0; (c = str->getChar()) != EOF && isdigit(c) && i < 20; ++i)
- s[i] = (char)c;
- if (i == 0)
- goto err2;
- s[i] = '\0';
- first = atoi(s);
- while ((c = str->lookChar()) != EOF && isspace(c))
- str->getChar();
- for (i = 0; (c = str->getChar()) != EOF && isdigit(c) && i < 20; ++i)
- s[i] = (char)c;
- if (i == 0)
- goto err2;
- s[i] = '\0';
- n = atoi(s);
- while ((c = str->lookChar()) != EOF && isspace(c))
- str->getChar();
+ }
+ if (!obj.isInt()) {
+ goto err1;
+ }
+ first = obj.getInt();
+ obj.free();
+ if (!parser->getObj(&obj)->isInt()) {
+ goto err1;
+ }
+ n = obj.getInt();
+ obj.free();
+ if (first < 0 || n < 0 || first + n < 0) {
+ goto err1;
+ }
+ if (first + n > size) {
+ for (newSize = size ? 2 * size : 1024;
+ first + n > newSize && newSize > 0;
+ newSize <<= 1) ;
+ if (newSize < 0) {
+ goto err1;
+ }
+ entries = (XRefEntry *)greallocn(entries, newSize, sizeof(XRefEntry));
+ for (i = size; i < newSize; ++i) {
+ entries[i].offset = 0xffffffff;
+ entries[i].type = xrefEntryFree;
+ }
+ size = newSize;
+ }
for (i = first; i < first + n; ++i) {
- for (j = 0; j < 20; ++j) {
- if ((c = str->getChar()) == EOF)
- goto err2;
- s[j] = (char)c;
+ if (!parser->getObj(&obj)->isInt()) {
+ goto err1;
}
- if (entries[i].offset < 0) {
- s[10] = '\0';
- entries[i].offset = atoi(s);
- s[16] = '\0';
- entries[i].gen = atoi(&s[11]);
- if (s[17] == 'n')
- entries[i].used = gTrue;
- else if (s[17] == 'f')
- entries[i].used = gFalse;
- else
- goto err2;
-#if 1 //~
- //~ PDF files of patents from the IBM Intellectual Property
- //~ Network have a bug: the xref table claims to start at 1
- //~ instead of 0.
+ entry.offset = (Guint)obj.getInt();
+ obj.free();
+ if (!parser->getObj(&obj)->isInt()) {
+ goto err1;
+ }
+ entry.gen = obj.getInt();
+ obj.free();
+ parser->getObj(&obj);
+ if (obj.isCmd("n")) {
+ entry.type = xrefEntryUncompressed;
+ } else if (obj.isCmd("f")) {
+ entry.type = xrefEntryFree;
+ } else {
+ goto err1;
+ }
+ obj.free();
+ if (entries[i].offset == 0xffffffff) {
+ entries[i] = entry;
+ // PDF files of patents from the IBM Intellectual Property
+ // Network have a bug: the xref table claims to start at 1
+ // instead of 0.
if (i == 1 && first == 1 &&
entries[1].offset == 0 && entries[1].gen == 65535 &&
- !entries[1].used) {
+ entries[1].type == xrefEntryFree) {
i = first = 0;
entries[0] = entries[1];
- entries[1].offset = -1;
+ entries[1].offset = 0xffffffff;
}
-#endif
}
}
}
- // read prev pointer from trailer dictionary
- obj.initNull();
- parser = new Parser(new Lexer(str->makeSubStream(str->getPos(), -1, &obj)));
- parser->getObj(&obj);
- if (!obj.isCmd("trailer"))
- goto err1;
- obj.free();
- parser->getObj(&obj);
- if (!obj.isDict())
+ // read the trailer dictionary
+ if (!parser->getObj(&obj)->isDict()) {
goto err1;
+ }
+
+ // get the 'Prev' pointer
obj.getDict()->lookupNF("Prev", &obj2);
if (obj2.isInt()) {
- *pos = obj2.getInt();
+ *pos = (Guint)obj2.getInt();
+ more = gTrue;
+ } else if (obj2.isRef()) {
+ // certain buggy PDF generators generate "/Prev NNN 0 R" instead
+ // of "/Prev NNN"
+ *pos = (Guint)obj2.getRefNum();
more = gTrue;
} else {
more = gFalse;
}
- obj.free();
obj2.free();
- delete parser;
+ // save the first trailer dictionary
+ if (trailerDict.isNone()) {
+ obj.copy(&trailerDict);
+ }
+
+ // check for an 'XRefStm' key
+ if (obj.getDict()->lookup("XRefStm", &obj2)->isInt()) {
+ pos2 = (Guint)obj2.getInt();
+ readXRef(&pos2);
+ if (!ok) {
+ obj2.free();
+ goto err1;
+ }
+ }
+ obj2.free();
+
+ obj.free();
return more;
err1:
obj.free();
- err2:
ok = gFalse;
return gFalse;
}
+GBool XRef::readXRefStream(Stream *xrefStr, Guint *pos) {
+ Dict *dict;
+ int w[3];
+ GBool more;
+ Object obj, obj2, idx;
+ int newSize, first, n, i;
+
+ dict = xrefStr->getDict();
+
+ if (!dict->lookupNF("Size", &obj)->isInt()) {
+ goto err1;
+ }
+ newSize = obj.getInt();
+ obj.free();
+ if (newSize < 0) {
+ goto err1;
+ }
+ if (newSize > size) {
+ entries = (XRefEntry *)greallocn(entries, newSize, sizeof(XRefEntry));
+ for (i = size; i < newSize; ++i) {
+ entries[i].offset = 0xffffffff;
+ entries[i].type = xrefEntryFree;
+ }
+ size = newSize;
+ }
+
+ if (!dict->lookupNF("W", &obj)->isArray() ||
+ obj.arrayGetLength() < 3) {
+ goto err1;
+ }
+ for (i = 0; i < 3; ++i) {
+ if (!obj.arrayGet(i, &obj2)->isInt()) {
+ obj2.free();
+ goto err1;
+ }
+ w[i] = obj2.getInt();
+ obj2.free();
+ if (w[i] < 0 || w[i] > 4) {
+ goto err1;
+ }
+ }
+ obj.free();
+
+ xrefStr->reset();
+ dict->lookupNF("Index", &idx);
+ if (idx.isArray()) {
+ for (i = 0; i+1 < idx.arrayGetLength(); i += 2) {
+ if (!idx.arrayGet(i, &obj)->isInt()) {
+ idx.free();
+ goto err1;
+ }
+ first = obj.getInt();
+ obj.free();
+ if (!idx.arrayGet(i+1, &obj)->isInt()) {
+ idx.free();
+ goto err1;
+ }
+ n = obj.getInt();
+ obj.free();
+ if (first < 0 || n < 0 ||
+ !readXRefStreamSection(xrefStr, w, first, n)) {
+ idx.free();
+ goto err0;
+ }
+ }
+ } else {
+ if (!readXRefStreamSection(xrefStr, w, 0, newSize)) {
+ idx.free();
+ goto err0;
+ }
+ }
+ idx.free();
+
+ dict->lookupNF("Prev", &obj);
+ if (obj.isInt()) {
+ *pos = (Guint)obj.getInt();
+ more = gTrue;
+ } else {
+ more = gFalse;
+ }
+ obj.free();
+ if (trailerDict.isNone()) {
+ trailerDict.initDict(dict);
+ }
+
+ return more;
+
+ err1:
+ obj.free();
+ err0:
+ ok = gFalse;
+ return gFalse;
+}
+
+GBool XRef::readXRefStreamSection(Stream *xrefStr, int *w, int first, int n) {
+ Guint offset;
+ int type, gen, c, newSize, i, j;
+
+ if (first + n < 0) {
+ return gFalse;
+ }
+ if (first + n > size) {
+ for (newSize = size ? 2 * size : 1024;
+ first + n > newSize && newSize > 0;
+ newSize <<= 1) ;
+ if (newSize < 0) {
+ return gFalse;
+ }
+ entries = (XRefEntry *)greallocn(entries, newSize, sizeof(XRefEntry));
+ for (i = size; i < newSize; ++i) {
+ entries[i].offset = 0xffffffff;
+ entries[i].type = xrefEntryFree;
+ }
+ size = newSize;
+ }
+ for (i = first; i < first + n; ++i) {
+ if (w[0] == 0) {
+ type = 1;
+ } else {
+ for (type = 0, j = 0; j < w[0]; ++j) {
+ if ((c = xrefStr->getChar()) == EOF) {
+ return gFalse;
+ }
+ type = (type << 8) + c;
+ }
+ }
+ for (offset = 0, j = 0; j < w[1]; ++j) {
+ if ((c = xrefStr->getChar()) == EOF) {
+ return gFalse;
+ }
+ offset = (offset << 8) + c;
+ }
+ for (gen = 0, j = 0; j < w[2]; ++j) {
+ if ((c = xrefStr->getChar()) == EOF) {
+ return gFalse;
+ }
+ gen = (gen << 8) + c;
+ }
+ if (entries[i].offset == 0xffffffff) {
+ switch (type) {
+ case 0:
+ entries[i].offset = offset;
+ entries[i].gen = gen;
+ entries[i].type = xrefEntryFree;
+ break;
+ case 1:
+ entries[i].offset = offset;
+ entries[i].gen = gen;
+ entries[i].type = xrefEntryUncompressed;
+ break;
+ case 2:
+ entries[i].offset = offset;
+ entries[i].gen = gen;
+ entries[i].type = xrefEntryCompressed;
+ break;
+ default:
+ return gFalse;
+ }
+ }
+ }
+
+ return gTrue;
+}
+
// Attempt to construct an xref table for a damaged file.
GBool XRef::constructXRef() {
Parser *parser;
- Object obj;
+ Object newTrailerDict, obj;
char buf[256];
- int pos;
+ Guint pos;
int num, gen;
int newSize;
int streamEndsSize;
int i;
GBool gotRoot;
+ gfree(entries);
+ size = 0;
+ entries = NULL;
+
error(0, "PDF file is damaged - attempting to reconstruct xref table...");
gotRoot = gFalse;
streamEndsLen = streamEndsSize = 0;
// got trailer dictionary
if (!strncmp(p, "trailer", 7)) {
obj.initNull();
- parser = new Parser(new Lexer(
- str->makeSubStream(start + pos + 7, -1, &obj)));
- if (!trailerDict.isNone())
- trailerDict.free();
- parser->getObj(&trailerDict);
- if (trailerDict.isDict()) {
- trailerDict.dictLookupNF("Root", &obj);
+ parser = new Parser(NULL,
+ new Lexer(NULL,
+ str->makeSubStream(pos + 7, gFalse, 0, &obj)));
+ parser->getObj(&newTrailerDict);
+ if (newTrailerDict.isDict()) {
+ newTrailerDict.dictLookupNF("Root", &obj);
if (obj.isRef()) {
rootNum = obj.getRefNum();
rootGen = obj.getRefGen();
+ if (!trailerDict.isNone()) {
+ trailerDict.free();
+ }
+ newTrailerDict.copy(&trailerDict);
gotRoot = gTrue;
}
obj.free();
- } else {
- pos = 0;
}
+ newTrailerDict.free();
delete parser;
// look for object
} else if (isdigit(*p)) {
num = atoi(p);
- do {
- ++p;
- } while (*p && isdigit(*p));
- if (isspace(*p)) {
+ if (num > 0) {
do {
++p;
- } while (*p && isspace(*p));
- if (isdigit(*p)) {
- gen = atoi(p);
+ } while (*p && isdigit(*p));
+ if (isspace(*p)) {
do {
++p;
- } while (*p && isdigit(*p));
- if (isspace(*p)) {
+ } while (*p && isspace(*p));
+ if (isdigit(*p)) {
+ gen = atoi(p);
do {
++p;
- } while (*p && isspace(*p));
- if (!strncmp(p, "obj", 3)) {
- if (num >= size) {
- newSize = (num + 1 + 255) & ~255;
- entries = (XRefEntry *)
- grealloc(entries, newSize * sizeof(XRefEntry));
- for (i = size; i < newSize; ++i) {
- entries[i].offset = -1;
- entries[i].used = gFalse;
+ } while (*p && isdigit(*p));
+ if (isspace(*p)) {
+ do {
+ ++p;
+ } while (*p && isspace(*p));
+ if (!strncmp(p, "obj", 3)) {
+ if (num >= size) {
+ newSize = (num + 1 + 255) & ~255;
+ if (newSize < 0) {
+ error(-1, "Bad object number");
+ return gFalse;
+ }
+ entries = (XRefEntry *)
+ greallocn(entries, newSize, sizeof(XRefEntry));
+ for (i = size; i < newSize; ++i) {
+ entries[i].offset = 0xffffffff;
+ entries[i].type = xrefEntryFree;
+ }
+ size = newSize;
+ }
+ if (entries[num].type == xrefEntryFree ||
+ gen >= entries[num].gen) {
+ entries[num].offset = pos - start;
+ entries[num].gen = gen;
+ entries[num].type = xrefEntryUncompressed;
}
- size = newSize;
- }
- if (!entries[num].used || gen >= entries[num].gen) {
- entries[num].offset = pos - start;
- entries[num].gen = gen;
- entries[num].used = gTrue;
}
}
}
} else if (!strncmp(p, "endstream", 9)) {
if (streamEndsLen == streamEndsSize) {
streamEndsSize += 64;
- streamEnds = (int *)grealloc(streamEnds, streamEndsSize * sizeof(int));
+ streamEnds = (Guint *)greallocn(streamEnds,
+ streamEndsSize, sizeof(int));
}
streamEnds[streamEndsLen++] = pos;
}
return gFalse;
}
-#ifndef NO_DECRYPTION
-GBool XRef::checkEncrypted(GString *userPassword) {
- Object encrypt, ownerKey, userKey, permissions, fileID, fileID1;
- GBool encrypted1;
- GBool ret;
-
- ret = gFalse;
+void XRef::setEncryption(int permFlagsA, GBool ownerPasswordOkA,
+ Guchar *fileKeyA, int keyLengthA, int encVersionA) {
+ int i;
- permFlags = defPermFlags;
- trailerDict.dictLookup("Encrypt", &encrypt);
- if ((encrypted1 = encrypt.isDict())) {
- ret = gTrue;
- encrypt.dictLookup("O", &ownerKey);
- encrypt.dictLookup("U", &userKey);
- encrypt.dictLookup("P", &permissions);
- trailerDict.dictLookup("ID", &fileID);
- if (ownerKey.isString() && ownerKey.getString()->getLength() == 32 &&
- userKey.isString() && userKey.getString()->getLength() == 32 &&
- permissions.isInt() &&
- fileID.isArray()) {
- permFlags = permissions.getInt();
- fileID.arrayGet(0, &fileID1);
- if (fileID1.isString()) {
- if (Decrypt::makeFileKey(ownerKey.getString(), userKey.getString(),
- permFlags, fileID1.getString(),
- userPassword, fileKey)) {
- ret = gFalse;
- } else {
- error(-1, "Incorrect user password");
- }
- } else {
- error(-1, "Weird encryption info");
- }
- fileID1.free();
- } else {
- error(-1, "Weird encryption info");
- }
- ownerKey.free();
- userKey.free();
- permissions.free();
- fileID.free();
+ encrypted = gTrue;
+ permFlags = permFlagsA;
+ ownerPasswordOk = ownerPasswordOkA;
+ if (keyLengthA <= 16) {
+ keyLength = keyLengthA;
+ } else {
+ keyLength = 16;
}
- encrypt.free();
-
- // this flag has to be set *after* we read the O/U/P strings
- encrypted = encrypted1;
-
- return ret;
-}
-#else
-GBool XRef::checkEncrypted(GString *userPassword) {
- Object obj;
- GBool encrypted;
-
- trailerDict.dictLookup("Encrypt", &obj);
- if ((encrypted = !obj.isNull())) {
- error(-1, "PDF file is encrypted and cannot be displayed");
- error(-1, "* Decryption support is currently not included in xpdf");
- error(-1, "* due to legal restrictions: the U.S.A. still has bogus");
- error(-1, "* export controls on cryptography software.");
+ for (i = 0; i < keyLength; ++i) {
+ fileKey[i] = fileKeyA[i];
}
- obj.free();
- return encrypted;
+ encVersion = encVersionA;
}
-#endif
-GBool XRef::okToPrint() {
-#ifndef NO_DECRYPTION
- if (!(permFlags & permPrint)) {
- return gFalse;
- }
-#endif
- return gTrue;
+GBool XRef::okToPrint(GBool ignoreOwnerPW) {
+ return (!ignoreOwnerPW && ownerPasswordOk) || (permFlags & permPrint);
}
-GBool XRef::okToChange() {
-#ifndef NO_DECRYPTION
- if (!(permFlags & permChange)) {
- return gFalse;
- }
-#endif
- return gTrue;
+GBool XRef::okToChange(GBool ignoreOwnerPW) {
+ return (!ignoreOwnerPW && ownerPasswordOk) || (permFlags & permChange);
}
-GBool XRef::okToCopy() {
-#ifndef NO_DECRYPTION
- if (!(permFlags & permCopy)) {
- return gFalse;
- }
-#endif
- return gTrue;
+GBool XRef::okToCopy(GBool ignoreOwnerPW) {
+ return (!ignoreOwnerPW && ownerPasswordOk) || (permFlags & permCopy);
}
-GBool XRef::okToAddNotes() {
-#ifndef NO_DECRYPTION
- if (!(permFlags & permNotes)) {
- return gFalse;
- }
-#endif
- return gTrue;
+GBool XRef::okToAddNotes(GBool ignoreOwnerPW) {
+ return (!ignoreOwnerPW && ownerPasswordOk) || (permFlags & permNotes);
}
Object *XRef::fetch(int num, int gen, Object *obj) {
// check for bogus ref - this can happen in corrupted PDF files
if (num < 0 || num >= size) {
- obj->initNull();
- return obj;
+ goto err;
}
e = &entries[num];
- if (e->gen == gen && e->offset >= 0) {
+ switch (e->type) {
+
+ case xrefEntryUncompressed:
+ if (e->gen != gen) {
+ goto err;
+ }
obj1.initNull();
- parser = new Parser(new Lexer(
- str->makeSubStream(start + e->offset, -1, &obj1)));
+ parser = new Parser(this,
+ new Lexer(this,
+ str->makeSubStream(start + e->offset, gFalse, 0, &obj1)));
parser->getObj(&obj1);
parser->getObj(&obj2);
parser->getObj(&obj3);
- if (obj1.isInt() && obj1.getInt() == num &&
- obj2.isInt() && obj2.getInt() == gen &&
- obj3.isCmd("obj")) {
-#ifndef NO_DECRYPTION
- parser->getObj(obj, encrypted ? fileKey : (Guchar *)NULL, num, gen);
-#else
- parser->getObj(obj);
-#endif
- } else {
- obj->initNull();
+ if (!obj1.isInt() || obj1.getInt() != num ||
+ !obj2.isInt() || obj2.getInt() != gen ||
+ !obj3.isCmd("obj")) {
+ obj1.free();
+ obj2.free();
+ obj3.free();
+ delete parser;
+ goto err;
}
+ parser->getObj(obj, encrypted ? fileKey : (Guchar *)NULL, keyLength,
+ num, gen);
obj1.free();
obj2.free();
obj3.free();
delete parser;
- } else {
- obj->initNull();
+ break;
+
+ case xrefEntryCompressed:
+ if (gen != 0) {
+ goto err;
+ }
+ if (!objStr || objStr->getObjStrNum() != (int)e->offset) {
+ if (objStr) {
+ delete objStr;
+ }
+ objStr = new ObjectStream(this, e->offset);
+ }
+ objStr->getObject(e->gen, num, obj);
+ break;
+
+ default:
+ goto err;
}
+
return obj;
+
+ err:
+ return obj->initNull();
}
Object *XRef::getDocInfo(Object *obj) {
return trailerDict.dictLookup("Info", obj);
}
-int XRef::getStreamEnd(int start) {
+// Added for the pdftex project.
+Object *XRef::getDocInfoNF(Object *obj) {
+ return trailerDict.dictLookupNF("Info", obj);
+}
+
+GBool XRef::getStreamEnd(Guint streamStart, Guint *streamEnd) {
int a, b, m;
if (streamEndsLen == 0 ||
- start > streamEnds[streamEndsLen - 1]) {
- return -1;
+ streamStart > streamEnds[streamEndsLen - 1]) {
+ return gFalse;
}
a = -1;
b = streamEndsLen - 1;
- // invariant: streamEnds[a] < start <= streamEnds[b]
+ // invariant: streamEnds[a] < streamStart <= streamEnds[b]
while (b - a > 1) {
m = (a + b) / 2;
- if (start <= streamEnds[m]) {
+ if (streamStart <= streamEnds[m]) {
b = m;
} else {
a = m;
}
}
- return streamEnds[b];
+ *streamEnd = streamEnds[b];
+ return gTrue;
+}
+
+Guint XRef::strToUnsigned(char *s) {
+ Guint x;
+ char *p;
+ int i;
+
+ x = 0;
+ for (p = s, i = 0; *p && isdigit(*p) && i < 10; ++p, ++i) {
+ x = 10 * x + (*p - '0');
+ }
+ return x;
}