Dur DurParse(char *s, char *stem)
{
if (streq(stem, "sec")) {
return((Dur)FloatParse(s, -1));
} else if (streq(stem, "min")) {
return((Dur)60.0*FloatParse(s, -1));
} else if (streq(stem, "hrs")) {
return((Dur)3600.0*FloatParse(s, -1));
} else if (streq(stem, "day")) {
return((Dur)24.0*3600.0*FloatParse(s, -1));
} else if (streq(stem, "yrs")) {
return((Dur)DAYSPERYEAR*24.0*3600.0*FloatParse(s, -1));
} else if (streq(stem, "hhmm")) {
return(IntParse(s, 2)*3600L + IntParse(s+2, 2)*60L);
} else if (streq(stem, "hhmmss")) {
return(IntParse(s, 2)*3600L + IntParse(s+2, 2)*60L +
(long)IntParse(s+4, 2));
} else if (streq(stem, "ns")) {
return((Dur)0); /* todo */
} else if (streq(stem, "ms")) {
return((Dur)0); /* todo */
} else {
Dbg(DBGGEN, DBGBAD, "DurParse: trouble parsing <%s> <%s>", s, stem);
return(DURERR);
}
}
Tod TodParse(char *s)
{
int len;
if (streq(s, "na")) return(TODNA);
len = strlen(s);
switch (len) {
case 2: return(atoi(s)*3600L);
case 4: return(IntParse(s, 2)*3600L + IntParse(s+2, 2)*60L);
case 6: return(IntParse(s, 2)*3600L + IntParse(s+2, 2)*60L +
(long)IntParse(s+4, 2));
default:
Dbg(DBGGEN, DBGBAD, "TodParse: trouble parsing <%s>", s);
return(TODERR);
}
}
Bool TsParse(Ts *ts, char *s)
{
int len;
ts->flag = 0;
ts->cx = ContextRoot;
ts->unixts = UNIXTSNA; /* REDUNDANT */
if (streq(s, "na") || streq(s, "?")) {
ts->unixts = UNIXTSNA; return(1);
} else if (streq(s, "Inf") || streq(s, "+Inf")) {
ts->unixts = UNIXTSPOSINF; return(1);
} else if (streq(s, "-Inf")) {
ts->unixts = UNIXTSNEGINF; return(1);
}
len = strlen(s);
switch (len) {
case 4: /* YYYY */
ts->unixts = YMDHMSToUnixTs(atoi(s), 1, 1, 0, 0, 0);
break;
case 6: /* YYYYMM */
ts->unixts = YMDHMSToUnixTs(IntParse(s, 4), IntParse(s+4, 2), 1, 0, 0, 0);
break;
case 8: /* YYYYMMDD */
ts->unixts = YMDHMSToUnixTs(IntParse(s, 4), IntParse(s+4, 2),
IntParse(s+6, 2), 0, 0, 0);
break;
case 15: /* YYYYMMDDTHHMMSS */
/* todo: time zones */
ts->unixts = YMDHMSToUnixTs(IntParse(s, 4), IntParse(s+4, 2),
IntParse(s+6, 2), IntParse(s+9, 2),
IntParse(s+11, 2), IntParse(s+13, 2));
break;
case 16: /* YYYYMMDDTHHMMSSZ */
/* todo: time zones */
ts->unixts = YMDHMSToUnixTs(IntParse(s, 4), IntParse(s+4, 2),
IntParse(s+6, 2), IntParse(s+9, 2),
IntParse(s+11, 2), IntParse(s+13, 2));
break;
case 20: /* YYYYMMDDTHHMMSS-HHMM */
/* todo: time zones */
ts->unixts = YMDHMSToUnixTs(IntParse(s, 4), IntParse(s+4, 2),
IntParse(s+6, 2), IntParse(s+9, 2),
IntParse(s+11, 2), IntParse(s+13, 2));
break;
default:
Dbg(DBGGEN, DBGBAD, "TsParse: trouble parsing <%s>", s);
ts->unixts = UNIXTSNA;
return(0);
}
return(1);
}
int SaveXrefs(FILE *fp, int index, int gap, int from, int to, int newStartxref,int boundary)
{
List writer = InitList();
LPSTR str = "";
int cursor,i,len;
Xref xref;
for (cursor = from; cursor < to; cursor++)
{
//xref之前
xref=(Xref)GetNodeData(xrefs,cursor);
while (index < xref->xrefOffset)
{
fputc(PDFInByte[index++],fp);
}
//xref地址部分
while (PDFInByte[index] != 13 && PDFInByte[index] != 10) fputc(PDFInByte[index++],fp);
while (PDFInByte[index] == 13 || PDFInByte[index] == 10) fputc(PDFInByte[index++],fp);
len=xref->sections->length;
Section sec;
for (i=0;i<len;i++)
{
int count,j;
sec=(Section)GetNodeData(xref->sections,i);
while (PDFInByte[index] != 13 && PDFInByte[index] != 10) fputc(PDFInByte[index++],fp);
while (PDFInByte[index] == 13 || PDFInByte[index] == 10) fputc(PDFInByte[index++],fp);
for (count = 0; count < sec->count; count++)
{
char Addr[11];
for (j = 0; j < 10; j++)
{
Addr[j] = PDFInByte[index + j];
}
Addr[j]=0;
int num=IntParse(Addr);
if (num > boundary)
{
num += gap;
for (int j = 0; j < 10; j++)
{
Addr[9 - j] = num % 10 + 48;
num /= 10;
}
}
fputs(Addr,fp);
index += 10;
for (j = 0; j < 10; j++)
{
fputc(PDFInByte[index++],fp);
}
}
}
//trailer的处理
while (PDFInByte[index] != 62 || PDFInByte[index + 1] != 62)
{
if (PDFInByte[index] == 80 && PDFInByte[index + 1] == 114 && PDFInByte[index + 2] == 101 && PDFInByte[index + 3] == 118)//Prev
{
fputs("Prev ",fp);
index +=(int)strlen("Prev ");
str=IntToString(xref->newPrev);
fputs(str,fp);
index += (int)strlen(str);
continue;
}
if (PDFInByte[index] == 88 && PDFInByte[index + 1] == 82 && PDFInByte[index + 2] == 101 && PDFInByte[index + 3] == 102)//XRef
{
writer=InitList();
int len = (int)strlen("XRefStm ")+(int)strlen(IntToString(xref->trailer->XRefStm));
index += len;
fputs("DelTag ",fp);
for (i = 0; i < len - 7; i++)fputc('0',fp);
continue;
}
fputc(PDFInByte[index++],fp);
}//结束时index指向>>的第一个>
//看是否有startxref
if (cursor == xrefs->length - 1)
{
int see = index + 2;
while (PDFInByte[see] == 13 || PDFInByte[see] == 10) see++; //结束时指向下一行第一个字符
//若有start
if (PDFInByte[see] == 115 && PDFInByte[see + 1] == 116 && PDFInByte[see + 2] == 97 && PDFInByte[see + 3] == 114 && PDFInByte[see + 4] == 116)
{
while (PDFInByte[see] < 48 || PDFInByte[see] > 57) see++;//see定位在第一个数字
while (index < see) fputc(PDFInByte[index++],fp);//结束时index指向第一个数字
str = IntToString(newStartxref);
fputs(str,fp);
index += (int)strlen(str);
if (cursor == xrefs->length- 1)
{
while (index < FileLength)
{
fputc(PDFInByte[index++],fp);
}
break;
}
}
}
}
return index;
}
//读取PDF的信息
void ReadPath(char *PDFPath)
{
FILE *fp;
if((fp=fopen(PDFPath,"rb"))==0)
{
printf("cannot open %s\n",PDFPath);
}
Title=L"";
Author=L"";
Subject=L"";
Keywords=L"";
CreationDate=L"";
ModDate=L"";
Creator=L"";
Producer=L"";
xrefs =InitList();
Tails=InitList();
Attrs=InitList();
FileLength=getFileSize(PDFPath);
printf("FileLength: %d\n",FileLength);
PDFInByte=(unsigned char*)malloc(sizeof(char)*FileLength);
fread(PDFInByte,1,FileLength,fp);
fclose(fp);
Version[0]=PDFInByte[5];
Version[1]= '.' ;
Version[2]=PDFInByte[7];
Version[3]=0;
printf("Version: %s\n",Version);
int index = FileLength;
index = findStringUp(PDFInByte,FileLength,index, "%%EOF") - 5;
while (PDFInByte[index] > 57 || PDFInByte[index] < 48) index--;
int i = 1,j;
StartXref = 0;
while (PDFInByte[index] <= 57 && PDFInByte[index] >= 48)
{
StartXref += (PDFInByte[index--] - 48) * i;
i *= 10;
}
printf("StartXref: %d\n",StartXref);
ReadXref(StartXref);
LPSTR* Addrs=(LPSTR*)malloc(sizeof(char*)*Size);
Xref xref;
Section section;
for(i=0;i<xrefs->length;i++)
{
xref=(Xref)GetNodeData(xrefs,i);
for(j=0;j< xref->sections->length;j++)
{
section=(Section)GetNodeData(xref->sections,j);
for (int cursor = 0; cursor < section->count; cursor++)
{
Addrs[section->startID+cursor]=(LPSTR)GetNodeData(section->Addrs,cursor); //用新的覆盖旧的
}
}
}
//读取Info
if (Info == 0) return;
index=InfoAt = IntParse(Substring(Addrs[Info],0, 10));
while (PDFInByte[index++] != 47) ; //找到第一个/,结束时index指向tag的第一个字符
List tag = InitList();
LPSTR tagString;
LPWSTR valueString;
valueString=(LPWSTR)malloc(sizeof(wchar_t));
while (PDFInByte[index] != 62)//>
{
//tag.Clear();
tag = InitList();
while (PDFInByte[index] != 32 && PDFInByte[index] != 40) //找到空格或者(，结束时index指向空格或(
{
AddNode(tag,(void*)PDFInByte[index]);
index++;
}
index = omitBlank(PDFInByte,index);
tagString =CharListGetString(tag);
//开始读取value
if (PDFInByte[index] != 40) //指向其他obj的情况
{
int tagID = getNumber(PDFInByte, &index);
index++;
while (PDFInByte[index] != 47 && PDFInByte[index] != 62) index++; //找下一个/或者结束符>，结束时index指向下一个/或结束符>
char *tagAddr = Substring(Addrs[tagID],0, 10);
int cursor = IntParse(tagAddr);
while (PDFInByte[cursor++] != 40) ; //找(，结束时指向属性的第一个字符
ReadValue(cursor,&valueString);
}
else //跟随的括号中有属性的情况
{
index = ReadValue(++index, &valueString) + 1; //index此时指向)后面一个字符
while (PDFInByte[index] != 47 && PDFInByte[index] != 62) index++; //找下一个/或者结束符>，结束时index指向下一个/或结束符>
}
index++;
//index最终定位在下一个tag的第一个字符或结束符>
int TagNum=-1,i;
for(i=0;i<8;i++)
{
if(strcmp(tagString,InfoTags[i])==0)
{
TagNum=i;
break;
}
}
switch (TagNum)
{
case 0: Title = valueString; break;//Title
case 1: Author = valueString; break;//Author
case 2: Subject = valueString; break;//Subject
case 3: Keywords = valueString; break;//Keywords
case 4: CreationDate = valueString; break;
case 5: ModDate =valueString; break;//ModDate
case 6: Creator = valueString; break;//Creator
case 7: Producer = valueString; break;//Producer
default:
Attr attr=(Attr)malloc(sizeof(_Attr));
attr->Name=tagString;
attr->Value=valueString;
AddNode(Attrs,attr); break;
}
}
while (PDFInByte[index] != 106) index++;//将index指向endobj的j
InfoEnd = index + 1;
//读取Tail
if (Tail == 0) return;
index=TailAt = IntParse(Substring(Addrs[Tail],0, 10));
while (PDFInByte[index++] != 47) ; //找到第一个/,结束时index指向tag的第一个字符
while (PDFInByte[index] != 62)//>
{
//tag.Clear();
tag=InitList();
while (PDFInByte[index] != 32 && PDFInByte[index] != 40) //找到空格或者(，结束时index指向空格或(
{
AddNode(tag,(void*)PDFInByte[index]);
index++;
}
index = omitBlank(PDFInByte,index);
tagString =CharListGetString(tag);
//开始读取value，默认Tail的属性只以空号的形式保存
index = ReadValue(++index, &valueString) + 1; //index此时指向)后面一个字符
while (PDFInByte[index] != 47 && PDFInByte[index] != 62) index++; //找下一个/或者结束符>，结束时index指向下一个/或结束符>
index++;
//index最终定位在下一个tag的第一个字符或结束符>
Attr tail=(Attr)malloc(sizeof(_Attr));
tail->Name=tagString;
tail->Value=valueString;
AddNode(Tails,tail);
}
while (PDFInByte[index] != 106) index++;//将index指向endobj的j
TailEnd = index + 1;
fclose(fp);//已经关闭过了？
}
