// called by only proc_hyperlink()
// get hyperlink title: <a href="/aaa/t.html"> good link </a>, "good link" is the title
// html: HTML source, title: save the result(==title)
// 타이틀 길이 제한
int htmlParse::get_hyperlink_title(char *html, char *title, int maxtitlelen)
{
char *ptr=html, *start=title;
int n=0;
#ifdef DEB
PRN("get_hyperlink_title(): html=%X, title=%X\n", html, title);
#endif
if (html==0|| title==0) return 0;
*title=0;
while (*ptr && *ptr!='>')
ptr++;
if (*ptr) ptr++;
ptr = StrUtil::skip_space(ptr);
while (*ptr && *ptr!='<' && n < maxtitlelen-2) {
*title++ = *ptr++;
n++;
}
#ifdef DEB
PRN("get_hyperlink_title(): n=%d, ptr=%X\n", n, ptr);
#endif
while (n>0 && ISSPACE(*(title-1)) ) {
title--;
n--;
}
#ifdef DEB
PRN("get_hyperlink_title(): n=%d, ptr=%X\n", n, ptr);
PRN("get_hyperlink_title(): html=%X, title=%X\n", html, title);
#endif
*title=0;
return n;
}
void TravExpr::print(TRVexpr *u)
{
PRN("URL EXPR: %s\n", u->org_url_expr);
//if ( u->ctg_name ) PRN("\tCategory= %s\n", u->ctg_name);
//if ( u->dtype_name ) PRN("\tDocType= %s\n", u->dtype_name);
PRN("\tCTG=%X, DocType=%X\n", u->ctg_id, u->doctype_id);
PRN("\tDepth= %d\n", u->depth);
if (u->host_span) PRN("\tAll host\n");
if (u->loop_type==INT_LOOP)
PRN("\t[%d..%d]\n", u->from, u->to);
else if (u->loop_type==CHAR_LOOP)
PRN("\t[%c..%c]\n", u->from, u->to);
if (u->user)
PRN("\tUser/Passwd: %s %s\n", u->user, u->passwd);
u->key_dirs.print("KEY: ");
u->inc_dirs.print("INC: ");
u->exc_dirs.print("EXC: ");
if (u->filename)
PRN("\tFile: %s\n", u->filename);
}
void htmlParse::print_maxlen_tag(char *textbuf)
{
char tmpch=0;
//tagStackType *maxtag = find_maxlen_tag();
tagStackType *maxtag = maxdelta_tag;
PRN("tagStack: stack_top = %d\n", stack_top);
if (maxtag)
PRN("tagStack: tag =%s, start=%d, leng=%d\n",
maxtag->tag->name, maxtag->start_pos, maxtag->len_txt);
tmpch = textbuf[maxtag->start_pos + maxtag->len_txt];
textbuf[maxtag->start_pos + maxtag->len_txt] = 0;
FileUtil::save_file_2_from_mem(textbuf + maxtag->start_pos, "D:/SVN/ad2005/trimmed.txt");
//PRN("%s\n", textbuf + maxtag->start_pos);
textbuf[maxtag->start_pos + maxtag->len_txt] = tmpch;
}
/* decode navigation data (Galileo E1B page) -----------------------------------
*
* args : uint8_t *buff1 I navigation data bits (even page)
* uint8_t *buff2 I navigation data bits (odd page)
* sdreph_t *eph I/O sdr ephemeris structure
* return : int word type
*-----------------------------------------------------------------------------*/
extern int decode_page_e1b(const uint8_t *buff1, const uint8_t *buff2,
sdrnav_t *nav)
{
int id;
/* buff is 240 bits (30 bytes) of composed two page parts */
/* see Galileo SISICD Table 39, pp. 34 */
uint8_t buff[30];
memcpy(buff,buff1,15); memcpy(&buff[15],buff2,15);
id=getbitu(buff,2,6); /* word type */
Ephemeris[nav->sat].PageN((id >= 7)? 999:id);
switch (id) {
case 0: decode_word0(buff,nav); break;
case 1: decode_word1(buff,nav); break;
case 2: decode_word2(buff,nav); break;
case 3: decode_word3(buff,nav); break;
case 4: decode_word4(buff,nav); break;
case 5: decode_word5(buff,nav); break;
case 6: decode_word6(buff,nav); break;
case 7: case 8: case 9: break; // almanac
case 10: decode_word10(buff,nav); break;
case 63: break; // dummy page (we've actually seen these)
default:
printf("%s UNKNOWN W%d\n", PRN(nav->sat), id);
break;
}
return id;
}
/**
HTML 내용을 text 형식으로 변환.
return : text 길이.
*/
int htmlParse::html2text(char *html_mem, int html_size, char *txt_mem, int max_txt_size)
{
stToken tok;
int tmp_txt_n=0, txt_n=0;
char * tmp_txt=0;
htmlTitle[0] = 0;
memset(&tok, 0, sizeof(tok));
tok.src_mem = html_mem;
tok.src_len = html_size;
txt_mem[0] = 0;
if ( (tmp_txt = new char [max_txt_size+1]) == NULL) {
PRN("memory fail\n");
return 0;
}
tmp_txt_n = parse(&tok, tmp_txt, max_txt_size);
#ifdef MAIN_TEXT_ONLY
print_maxlen_tag(tmp_txt);
#endif
#ifdef DEB
PRN("html --> tmp_txt: %d --> %d\n",
html_size, tmp_txt_n);
PRN("tmp_txt strlen=%d, max_txt_size=%d\n", strlen(tmp_txt), max_txt_size );
//printf("%s", tmp_txt);
#endif
txt_n = StrUtil::trim_text(txt_mem, tmp_txt, max_txt_size);
#ifdef DEB
PRN("trim text: %d --> %d\n",
tmp_txt_n, txt_n);
PRN("txt_mem strlen=%d\n", strlen(txt_mem) );
#endif
delete [] tmp_txt;
return txt_n;
}
int extractCron::loadTable_DB()
{
int tabsize = 0;
int n=0;
DbQuery query;
char *cronjob[6];
tabsize = dbStore::countRows("cronjob");
PRN("NUMBER of cronjob = %d\n", tabsize);
if (tabsize < 0) {
PRN("DB query fail\n");
}
tabsize += 2;
this->allocTable(tabsize);
if ( query.exec(dbStore::dbConn, "select * from cronjob") < 0) {
PRN("DB query error!\n");
return 0;
}
while (query.fetch()) {
int jobtype = 0;
jobtype = query.colIntVal(6);;
/*
@todo
jobtype == 1 : HARVEST
jobtype == 2 : EXTRACT
*/
if (jobtype!=2) continue;
for (int i=0; i<5; i++)
cronjob[i] = query.colStrVal(i+1);
//cronjob[5] = query.colStrVal(7);
cronjob[5] = query.colStrVal_trimmed(7);
this->insertCronJob( cronjob );
}
return n;
}
void ArchiveEditable::add(const _TCHAR *fn)
{
ListItr li = find(fn);
if(li == list_.end()) {
struct _stat st;
if( 0 == _tstat(fn, &st) ) {
if(st.st_mode & _S_IFDIR) { // ディレクトリなら
TDIR *dir; tdirent *ent;
dir = topendir(fn);
while(dir) {
ent = treaddir(dir);
if(!ent) break;
if(!_tcscmp(ent->d_name, _T(".")) && !_tcscmp(ent->d_name, _T(".."))) {
basic_string<_TCHAR> path(fn); (path += _T("/")) += ent->d_name;
add(path.c_str());
PRN(path.c_str());
}
}
tclosedir(dir);
}
else {
// 最後尾に追加
ArchiveEntry newe(fn);
newe.time = st.st_mtime;
newe.size = st.st_size;
if(list_.size() > 0) newe.pos = list_.back().pos+list_.back().size;
else newe.pos = 0;
list_.push_back(EntryAttr(newe, EntryAttr::NEW));
}
}
}
else {
struct _stat st;
if( 0 == _tstat(fn, &st) && (li->time != st.st_mtime || li->size != st.st_size)) {
li->attr = EntryAttr::UPDATE;
li->time = st.st_mtime; li->size = st.st_size;
}
else if(li->attr == EntryAttr::REMOVE) li->attr = EntryAttr::ALREADY;
}
}
int main(int argc, char *argv[]) {
//declare the time period
extern int monthDay[12];
extern int periodStartYear, periodStartMonth, periodStartDay, periodStartHour, periodStartMinute;
extern int periodEndYear, periodEndMonth, periodEndDay, periodEndHour, periodEndMinute;
FILE *fp, *tmpfp, *dest;
int intbuf,qtm;
qtm = 0;
//type of scheduler
char* scheduleType = (char*)malloc(10);
//current number of array slot
int eventSize = 80;
//declare the event array, can be expanded its size twice to contain all events
Event *event = (Event*)calloc(eventSize, sizeof(*event));
//printf("allocate calloc space + %d",sizeof(*event));
int i, j,k,l,m,havevalidday;
int pid = 0;
// initialize all the data in event array (may not needed)
for (i=0; i<eventSize; i++) {
strcpy(event[i].type, "");
strcpy(event[i].name, "");
event[i].startYear = 0;
event[i].startMonth = 0;
event[i].startDay = 0;
event[i].startHour = 0;
event[i].dueYear = 0;
event[i].dueMonth = 0;
event[i].dueDay = 0;
event[i].dueHour = 0;
event[i].duration = 0;
event[i].priority = 0;
event[i].handled = 0;
event[i].startTimeValue = 0;
event[i].dueTimeValue = 0;
}
printf(" ~~WELCOME TO ASKS~~\n\n");
//declare the buffer to contain the input string
char input[100];
printf("Please enter:\n");
fgets(input, 100, stdin);
//initialize the period of schedule
char* tmp;
char* tmpforrr;
tmp = strtok(input, " ");
tmp = strtok(NULL, "-");
periodStartYear = atoi(tmp);
tmp = strtok(NULL, "-");
periodStartMonth = atoi(tmp);
tmp = strtok(NULL, " ");
periodStartDay = atoi(tmp);
tmp = strtok(NULL, ":");
periodStartHour = atoi(tmp);
tmp = strtok(NULL, " ");
periodStartMinute = atoi(tmp);
tmp = strtok(NULL, "-");
periodEndYear = atoi(tmp);
tmp = strtok(NULL, "-");
periodEndMonth = atoi(tmp);
tmp = strtok(NULL, " ");
periodEndDay = atoi(tmp);
tmp = strtok(NULL, ":");
periodEndHour = atoi(tmp);
tmp = strtok(NULL, " ");
periodEndMinute = atoi(tmp);
// time table array
int duryear = periodEndYear-periodStartYear+1;
int timetable [duryear][12][31][24];
int max = (periodEndYear-periodStartYear)*1000000+periodEndMonth*10000+periodEndDay*100+periodEndHour;
int min = periodStartMonth*10000+periodStartDay*100+periodStartHour;
int dayAmount = 0;
// initialization
int tmptime;
for(i=0; i<duryear; i++) {
for(j=0; j<12; j++) {
for(k=0; k<31; k++) {
havevalidday = 0;
for(l=0; l<24; l++) {
tmptime = i*1000000+(j+1)*10000+(k+1)*100+l;
if(tmptime<min||tmptime>max)
timetable[i][j][k][l] = -2; //day do not need, disable
else
timetable[i][j][k][l] = -1; //valid day, give -1
if(k>monthDay[j]-1)
timetable[i][j][k][l] = -2; //day do not exist, disable
if(timetable[i][j][k][l] == -1)
havevalidday = 1;
}
if(havevalidday == 1)
dayAmount++;
}
}
}
int timetablebyday[dayAmount][24];
// for counting the number of used array slot to decide whether there is need to expand the array
int eventCounter = 0;
//initialize the formal input event and put them into correct Event data type.
while(1) {
printf("Please enter:\n");
char input[100];
fgets(input, 100, stdin);
//decide whether this is the end of input
if (input[0]=='a'&&input[1]=='d'&&input[2]=='d') {
//initialize the type of event
tmp = strtok(input, " ");
for (i=3; i<strlen(tmp)+1; i++) {
*(tmp+i-3) = *(tmp+i);
}
strcpy(event[eventCounter].type, tmp);
//printf("%s\n",event[eventCounter].type);
//initialize the name of event
tmp = strtok(NULL, " ");
strcpy(event[eventCounter].name, tmp);
//printf("%s\n",event[eventCounter].name);
if (strcmp(event[eventCounter].type, "Project") != 0) {
//initialize the year of start time of this event
tmp = strtok(NULL, "-");
event[eventCounter].startYear = atoi(tmp);
//printf("%d\n",event[eventCounter].startYear);
//initialize the month of start time of this event
tmp = strtok(NULL, "-");
event[eventCounter].startMonth = atoi(tmp);
//printf("%d\n",event[eventCounter].startMonth);
//initialize the day of start time of this event
tmp = strtok(NULL, " ");
event[eventCounter].startDay = atoi(tmp);
//printf("%d\n",event[eventCounter].startDay);
//initialize the hour of start time of this event
tmp = strtok(NULL, ":");
event[eventCounter].startHour = atoi(tmp);
//printf("%d\n",event[eventCounter].startHour);
tmp = strtok(NULL, " ");
}
//test whether the event is project or assignment
if (strcmp(event[eventCounter].type, "Project") == 0 || strcmp(event[eventCounter].type, "Assignment") == 0) {
//initialize the year of due time of this event
tmp = strtok(NULL, "-");
event[eventCounter].dueYear = atoi(tmp);
//initialize the month of due time of this event
tmp = strtok(NULL, "-");
event[eventCounter].dueMonth = atoi(tmp);
//initialize the day of due time of this event
tmp = strtok(NULL, " ");
event[eventCounter].dueDay = atoi(tmp);
//initialize the hour of due time of this event
tmp = strtok(NULL, ":");
event[eventCounter].dueHour = atoi(tmp);
tmp = strtok(NULL, " ");
int minute = atoi(tmp);
if (minute > 0) {
event[eventCounter].dueHour++;
}
}
//initialize the duration
tmp = strtok(NULL, " ");
event[eventCounter].duration = atoi(tmp);
//printf("%d\n",event[eventCounter].duration);
tmp = strtok(NULL, " ");
event[eventCounter].priority = atoi(tmp);
//printf("%d\n",event[eventCounter].priority);
//expand dynamic array to contain more event (may need more tests)
eventCounter++;
if (eventCounter == eventSize) {
Event *tmpEvent = (Event*)calloc(eventSize, 80);
memcpy(tmpEvent, event, eventSize*80);
eventSize *= 2;
event = (Event*)calloc(eventSize, 80);
memcpy(event, tmpEvent, eventSize*40);
free(tmpEvent);
}
} else if(strcmp(strtok(input, "-"), "runASKS ") == 0) {
tmp = strtok(NULL, "|");
strcpy(scheduleType, tmp);
// printf("---%s---\n",scheduleType);
tmp = strtok(NULL, " ");
tmp = strtok(NULL, " ");
tmp = strtok(NULL, " ");
tmp[strlen(tmp)-2] = '\0';
fp = fopen("report.dat", "wt");
tmpfp = fopen("analysis.dat","wt");
break;
}
//printf("%d\n", eventCounter); //testing code
}//end of input module
// put all class and sleep into the timetable
int theYear;
int theMonth;
int theDay;
int theHour;
int timetablevalue;
int periodStartTimeValue = periodStartMonth*10000+periodStartDay*100+periodStartHour;
int periodEndTimeValue = duryear*1000000+periodEndMonth*10000+periodEndDay*100+periodEndHour;
for(m=0; m<eventCounter; m++) {
theYear = event[m].startYear - periodStartYear;
theMonth = event[m].startMonth;
theDay = event[m].startDay;
theHour = event[m].startHour;
if((event[m].startYear+event[m].startMonth+event[m].startDay+event[m].startHour)!=0) {
// if start time has been input, calculate the starttimevalue
event[m].startTimeValue = theYear*1000000+theMonth*10000+theDay*100+theHour;
}
else { // if start time not input, set it to the start of the period and calculate
event[m].startYear = periodStartYear;
event[m].startMonth = periodStartMonth;
event[m].startDay = periodStartDay;
event[m].startHour = periodStartHour;
theYear = event[m].startYear - periodStartYear;
theMonth = event[m].startMonth;
theDay = event[m].startDay;
theHour = event[m].startHour;
event[m].startTimeValue = theYear*1000000+theMonth*10000+theDay*100+theHour;
}
if((event[m].dueYear+event[m].dueMonth+event[m].dueDay+event[m].dueHour)!=0)
// calculate duetimevalue
event[m].dueTimeValue = (event[m].dueYear-periodStartYear)*1000000+event[m].dueMonth*10000+event[m].dueDay*100+event[m].dueHour;
// printf("%d %d %d %d\n",event[m].startYear,event[m].startMonth,event[m].startDay,event[m].startHour);
// printf("%d %d\n ",event[m].startTimeValue,event[m].dueTimeValue);
if(strcmp(event[m].type,"Class")==0||strcmp(event[m].name,"Sleep")==0) { // for class and sleep
if(event[m].startTimeValue<periodStartTimeValue) {
//printf("%s %d %d\n",event[m].name,event[m].startTimeValue,periodStartTimeValue);
continue;
}
//printf("%s %s \n",event[m].type,event[m].name);
for(i=theYear; i<duryear; i++) { //year
for(j=0; j<12; j++) { //month
for(k=0; k<31; k++) { //day
for(l=0; l<24; l++) { //hour
timetablevalue = (i)*1000000+(j+1)*10000+(k+1)*100+l+1;//time in this round of loop
if(timetablevalue > event[m].startTimeValue &&timetable[i][j][k][l] == -1) {
event[m].handled++;
// 啊啊啊腰疼疼疼疼 //printf("handled\n");
timetable[i][j][k][l] = m;
if(event[m].handled == event[m].duration) {
// printf("%d %d %d %d %s\n",i+1,j+1,k+1,l+1,event[m].type);
break;
}
}//if
if(event[m].handled == event[m].duration) {
break;
}
}//l
if(event[m].handled == event[m].duration) {
break;
}
}//k
if(event[m].handled == event[m].duration) {
break;
}
}//j
if(event[m].handled == event[m].duration) {
break;
}
}//i
}
}
//printf("%s\n",scheduleType);
if(strcmp(scheduleType, "FCFS ") == 0) {
// printf("fcfs\n");
FCFS(event,eventCounter,duryear,timetable);
scheduleType = "First Come First Serve";
}
else if(strcmp(scheduleType, "SJF ") == 0) {
// printf("sjf\n");
SJF(event,eventCounter,duryear,timetable);
scheduleType = "Short Job First";
}
else if(strcmp(scheduleType, "SRT ") == 0) {
// printf("srt\n");
SRT(event,eventCounter,duryear,timetable);
scheduleType = "Shortest Remaining Time";
}
else if(strcmp(scheduleType, "PR p ") == 0) {
// printf("pr p\n");
PRP(event,eventCounter,duryear,timetable);
scheduleType = "priority with preemption";
}
else if(strcmp(scheduleType, "PR n ") == 0) {
// printf("pr n\n");
PRN(event,eventCounter,duryear,timetable);
scheduleType = "priority without preemption";
}
else {
tmpforrr = strtok(scheduleType, " ");
if(strcmp(tmpforrr,"RR ")) {
qtm = atoi(strtok(NULL," "));
RRX(event,eventCounter,duryear,timetable,qtm);
scheduleType = "Round Robin with quantum of ";
}
}
/*for(i=0;i<duryear;i++){
for(j=0;j<12;j++){
// printf("month %d\n",j+1);
for(k=0;k<31;k++){
// printf("day %d\n",k+1);
for(l=0;l<24;l++){
if(timetable[i][j][k][l] != -2)
printf("%d ",timetable[i][j][k][l]);
}
// printf("\n");
}
}
}
printf("\n-----\n");
*/
//prepare for multi-process
int fd[2];
if(pipe(fd)<0) {
printf("pipe creation error!\n");
exit(1);
}
pid = fork();
if(pid == 0) { //child for output...
//printf("outer child %d %d\n",getpid(),getppid());
close(fd[1]);
for(i=0; i<dayAmount; i++) {
for(j=0; j<24; j++) {
if(i==0&&j<periodStartHour)
timetablebyday[i][j] = -2;
else if(i==dayAmount-1&&j>periodEndHour)
timetablebyday[i][j] = -2;
else {
read(fd[0],&intbuf,sizeof(int));
timetablebyday[i][j] = intbuf;
//printf(" %d",timetablebyday[i][j]);
}
}
}
pid = fork();
if(pid == 0) { //child of child for output
//printf("inner child %d %d\n",getpid(),getppid());
fprintf(fp, "Alex Timetable\n");
fprintf(fp, "%d-%02d-%02d %02d:%02d to %d-%02d-%02d %02d:%02d\n", periodStartYear, periodStartMonth, periodStartDay,periodStartHour, periodStartMinute, periodEndYear, periodEndMonth, periodEndDay, periodEndHour, periodEndMinute);
fprintf(fp, "time\t");
for(i=0; i<duryear; i++) {
for(j=0; j<12; j++) {
for(k=0; k<31; k++) {
havevalidday = 0;
for(l=0; l<24; l++) {
if(timetable[i][j][k][l] != -2)
havevalidday = 1;
}
if(havevalidday == 1)
fprintf(fp,"%d-%02d-%02d\t",i+periodStartYear,j+1,k+1);
}//k
}//j
}//i
fprintf(fp,"\n");
for(i=0; i<24; i++) {
fprintf(fp,"%02d:00\t",i);
for(j=0; j<dayAmount; j++) {
if(timetablebyday[j][i] == -1)
fprintf(fp,"N/A\t\t");
else if(timetablebyday[j][i] > -1)
fprintf(fp,"%s \t",event[timetablebyday[j][i]].name);
else
fprintf(fp,"invalid time\t");
}
fprintf(fp,"\n");
}
fclose(fp);
exit(0);
}
else { //parent of child for analysis
//printf("child parent %d %d\n",getpid(),getppid());
fprintf(tmpfp,"\n\n**Summary of assignments allocation**\n\n");
fprintf(tmpfp,"Algorithms used: %s",scheduleType);
if(qtm != 0)
fprintf(tmpfp,"%d",qtm);
fprintf(tmpfp,"\n");
int cnt=0;
for(i=0; i<eventCounter; i++) {
if(event[i].duration == event[i].handled)
cnt++;
}
int turntime = 0;
int tmpturntime = 0;
int tmpwaittime = 0;
int waittime = 0;
fprintf(tmpfp,"There are %d assignments scheduled. Details are as follows\n",cnt);
fprintf(tmpfp,"ASSIGNMENT\tSTART\t\t\tEND\t\t\tTURNAROUND TIME\n");
fprintf(tmpfp,"===========================================================================\n");
//finished calculate turnaround time and wait time
for(m=0; m<eventCounter; m++) {
tmpturntime = 0;
tmpwaittime = 0;
if(event[m].duration != event[m].handled)// not finished
continue;//pass
event[m].handled = 0;
for(i=0; i<duryear; i++) { //year
for(j=0; j<12; j++) { //month
for(k=0; k<31; k++) { //day
for(l=0; l<24; l++) { //hour
if(timetable[i][j][k][l]==-2)
continue;
timetablevalue = (i)*1000000+(j+1)*10000+(k+1)*100+l;//time in this round of loop
if(event[m].startTimeValue<timetablevalue+1) {
tmpwaittime++;
tmpturntime++;
}
if(timetable[i][j][k][l] == m) {
if(event[m].handled==0) {
event[m].startYear = i+periodStartYear;
event[m].startMonth = j+1;
event[m].startDay = k+1;
event[m].startHour = l;
}
event[m].handled++;
tmpwaittime--;
if(event[m].handled == event[m].duration) {
fprintf(tmpfp,"%s\t%d-%02d-%02d %02d:00\t%d-%02d-%02d %02d:00\t%dHrs\t\n",event[m].name,event[m].startYear,event[m].startMonth,event[m].startDay,event[m].startHour,i+periodStartYear,j+1,k+1,l,tmpturntime);
turntime = turntime + tmpturntime;
waittime = waittime + tmpwaittime;
break;
}
}
}//l
if(event[m].handled == event[m].duration) {
break;
}
}//k
if(event[m].handled == event[m].duration) {
break;
}
}//j
if(event[m].handled == event[m].duration) {
break;
}
}//i
}
fprintf(tmpfp," -end-\n");
fprintf(tmpfp,"\n===========================================================================\n");
fprintf(tmpfp,"**PERFORMANCE\n\n");
fprintf(tmpfp,"AVERAGE TURNAROUND TIME FOR ASSIGNMENTS: %d HRS\n",turntime/cnt);
fprintf(tmpfp,"AVERAGE WAITINGTIME FOR ASSIGNMENTS: %d HRS\n",waittime/cnt);
fprintf(tmpfp,"\n**Outstanding/Rejected List**\n\n");
m=0;
for(i=0; i<eventCounter; i++)
if(event[i].duration!=event[i].handled) {
m++;
fprintf(tmpfp,"%s %s\n",event[i].type,event[i].name);
}
fprintf(tmpfp,"There are %d events rejected.\n\n\n",m);
fclose(tmpfp);
//remove("analysis.dat");
waitpid();
exit(0);
}
}
else { //parent
close(fd[0]);
//printf("parent %d %d\n",getpid(),getppid());
for(i=0; i<duryear; i++) {
for(j=0; j<12; j++) {
for(k=0; k<31; k++) {
for(l=0; l<24; l++) {
if(timetable[i][j][k][l] != -2) {
intbuf = timetable[i][j][k][l];
// printf(" %d \n ",intbuf);
write(fd[1],&intbuf,sizeof(int));
}
}
}
}
}
waitpid();
sleep(1);
fp = fopen("report.dat", "r");
tmpfp = fopen("analysis.dat","r");
dest = fopen(tmp,"a+");
char byte;
while (!feof(fp)) {
fread(&byte, sizeof(char), 1, fp);
fwrite(&byte, sizeof(char), 1, dest);
}
close(fp);
while (!feof(tmpfp)) {
fread(&byte, sizeof(char), 1, tmpfp);
fwrite(&byte, sizeof(char), 1, dest);
}
close(tmpfp);
close(dest);
remove("report.dat");
remove("analysis.dat");
exit(0);
}
}
/**
HTML 문서를 parsing 하면서 text로 변환.
*/
int htmlParse::parse(stToken *tok, char *text, int maxlen)
{
int tok_val=0, prev_tok_val=0, prev_ch=0;
int is_in_tag=0;
uint4 tmp=0;
int not_moved=0;
char *start=text;
stTagProc *curr_tag=0;
//stTagProc **prev_tag=0;
stEntityProc *ent_proc;
int tag_depth=0;
int is_in_BODY_tag=0;
int is_in_A_tag=0;
int is_in_PRE_tag=0; // 2003.11.8
int hlink_chars = 0;
//int newline_delayed=0; // Optional Pair && Newline tag가 시작한 경우 TRUE
//int dont_newline=0; // TRUE means "dont write newline"
int pair_mark=0;
int words_in_line=0;
int b_tag_first_in_line=0; // 라인 첫머리에 B tag가 나온 경우
int glossary_marked=0;
int is_glossary=0;
tok->offset = 0;
tok->line_num=1;
*text = 0;
maxlen -= 20;
while(tok->offset < tok->src_len && (int)((uint4)text-(uint4)start) < maxlen) {
tmp = tok->offset;
htmlLex::get_token_mem(tok);
#ifdef BUG
if (tok->tok_len>=20) {
printf("long tok: line=%d, tok_val=%d, CH=%c\n",
tok->line_num, tok->tok_val, tok->src_mem[tok->offset-1]);
fflush(stdout);
printf("tok=%s\n", tok->tok_str);
fflush(stdout);
}
#endif
#ifdef DEB
/*
if (tok->tok_val==0 || tok->line_num==1) {
printf("line=%d, tok_val=%d, CH=%c\n",
tok->line_num, tok->tok_val, tok->src_mem[tok->offset-1]);
printf("offset=%d, text-start=%d\n", tok->offset, text-start);
}
*/
#endif
if (text > start) {
prev_ch = *(text-1);
if (prev_ch=='\n') {
words_in_line = 0;
b_tag_first_in_line = 0;
glossary_marked = 0;
}
}
if (tok->tok_val == TOK_EOF) break;
if (tok->tok_val == URL_CMNT) {
//printf("%s", tok->tok_str);
continue;
}
if (tok->offset==tmp) {
#ifdef BUG
//printf("parse(): offset not moved: line=%d, CH=%d(%c)\n",
// tok->line_num, tok->src_mem[tok->offset-1], tok->src_mem[tok->offset-1]);
#endif
tok->offset++;
if (++not_moved >= 2) break;
else continue;
}
tok_val = tok->tok_val;
if (tok_val == BTAG_BEGIN) {
skip_to_two_token(tok, STRING, TAG_END);
if (tok->tok_val != STRING) { // must be a HTML tag
skip_to_token(tok, TAG_END);
continue;
}
//prn_tabs();
//fprintf(log,"%s-->:%d :%d\n", tok->tok_str, tok->line_num, tok->offset);
curr_tag = htmlTagEntity::tag_idx(tok->tok_str);
if (curr_tag == NULL) {
skip_to_token(tok, TAG_END);
continue;
}
tag_depth++;
#ifdef MAIN_TEXT_ONLY
if (curr_tag->pair != Not_PAIR && is_in_BODY_tag && curr_tag!=htmlTagEntity::A_tag)
push_tag(curr_tag, tag_depth, (int)((uint4)text-(uint4)start));
if (curr_tag==htmlTagEntity::BODY_tag) {
is_in_BODY_tag = 1;
}
#endif
if (curr_tag==htmlTagEntity::A_tag) {
is_in_A_tag = 1;
}
else if (curr_tag==htmlTagEntity::PRE_tag) {
is_in_PRE_tag = 1;
}
else if(curr_tag==htmlTagEntity::B_tag||curr_tag==htmlTagEntity::DT_tag) {
if (words_in_line==0)
b_tag_first_in_line = 1;
}
else if (curr_tag==htmlTagEntity::TITLE_tag) {
get_hyperlink_title(tok->src_mem + tok->offset , htmlTitle, HTMLTITLE_LEN);
#ifdef DEB
PRN("tok->scr_mem=%X, offset=%d, title=%X\n",
tok->src_mem, tok->offset, htmlTitle);
PRN("title: %s (%d)\n", htmlTitle, strlen(htmlTitle) );
#endif
is_glossary = is_glossay_mode(htmlTitle);
}
else if (curr_tag==htmlTagEntity::BR_tag) {
*text++ = '\n';
}
if (curr_tag->proc == 0) { // 0
if (curr_tag->pair != Not_PAIR)
skip_ignore_part(tok, curr_tag);
else
skip_to_token(tok, TAG_END);
if (curr_tag->newline) {
if (prev_ch != '\n') *text++ = '\n';
}
continue;
}
if (curr_tag==htmlTagEntity::TITLE_tag) {
if (prev_ch != '\n') *text++ = '\n';
*text++ = '[';
*text++ = '[';
}
else if (curr_tag->proc == 1) {
}
else if (!pair_mark && curr_tag->proc == 2) {
pair_mark = 2;
*text++ = '<';
}
else if (!pair_mark && curr_tag->proc == 3) {
pair_mark = 3;
*text++ = '\'';
}
else if (!pair_mark && curr_tag->proc == 4) {
pair_mark = 4;
*text++ = '[';
}
else if (curr_tag->proc == 5) {
if (prev_ch != '\n') *text++ = '\n';
*text++ = '*';
*text++ = ' ';
words_in_line = 0;
}
else if (curr_tag->proc == 6) {
if (prev_ch != '\n') *text++ = '\n';
*text++ = '\n';
}
skip_to_token(tok, TAG_END);
if (!pair_mark && prev_ch != ' ') *text++ = ' ';
//if (!ISSPACE(*text)) *text++ = ' ';
}// BTAG_BEGIN '<'
else if (tok_val == ETAG_BEGIN) {
skip_to_two_token(tok, STRING, TAG_END);
if (tok->tok_val != STRING) { // must be a HTML tag
skip_to_token(tok, TAG_END);
continue;
}
curr_tag = htmlTagEntity::tag_idx(tok->tok_str);
if (curr_tag == NULL) {
skip_to_token(tok, TAG_END);
continue;
}
#ifdef MAIN_TEXT_ONLY
tag_depth--;
if (curr_tag==htmlTagEntity::BODY_tag) {
is_in_BODY_tag = 0;
}
if (curr_tag->pair != Not_PAIR && is_in_BODY_tag && curr_tag!=htmlTagEntity::A_tag)
end_tag(curr_tag, (int)((uint4)text-(uint4)start));
#endif
if (curr_tag==htmlTagEntity::A_tag) {
is_in_A_tag = 0;
#ifdef MARK_A_LINK
*text++ = '}';
#endif
}
else if (curr_tag==htmlTagEntity::PRE_tag) {
is_in_PRE_tag = 0;
}
else if(curr_tag==htmlTagEntity::DT_tag ||
(curr_tag==htmlTagEntity::B_tag && (b_tag_first_in_line && is_glossary)))
{
if (glossary_marked==0) {
*text++ = ' ';
*text++ = ':';
//*text++ = ' ';
b_tag_first_in_line = 0;
glossary_marked = 1;
}
}
else if (curr_tag==htmlTagEntity::TITLE_tag) {
*text++ = ']';
*text++ = ']';
*text++ = '\n';
}
else if (curr_tag->proc == 1) {
}
else if (pair_mark==2 && curr_tag->proc == 2) {
pair_mark = 0;
*text++ = '>';
}
else if (pair_mark==3 && curr_tag->proc == 3) {
pair_mark = 0;
*text++ = '\'';
}
else if (pair_mark==4 && curr_tag->proc == 4) {
pair_mark = 0;
*text++ = ']';
}
else if (curr_tag->proc == 5) {
}
else if (curr_tag->proc == 6) {
if (prev_ch != '\n') *text++ = '\n';
*text++ = '\n';
}
if (curr_tag->newline) {
*text++ = '\n';
}
skip_to_token(tok, TAG_END);
}// ETAG_BEGIN '</'
else if (tok_val == STAG_END) {
skip_to_token(tok, TAG_END);
is_in_tag = 0;
tag_depth--;
}
else if (tok_val == TAG_EXC) {
tmp = skip_to_token(tok, TAG_END);
//fprintf(log,"<- > skipped %d\n", tmp);
}
else if (tok_val == CMNT_BEGIN) {
//fprintf(log,"Cmnt -->: %d :%d\n", tok->line_num, tok->offset);
htmlLex::skip_to_cmnt_end(tok);
//fprintf(log,"Cmnt <--: %d :%d\n", tok->line_num, tok->offset);
}
else if (tok_val == ENTITY_STR) {
if (prev_ch !=' ') *text++ = ' '; //2002.12.2
ent_proc = htmlTagEntity::entity_idx(tok->tok_str);
if (ent_proc && ent_proc->conv[0]) {
#ifdef BUG
//prn_ent_proc(ent_proc);
#endif
strcpy(text, ent_proc->conv);
text += strlen(ent_proc->conv);
}
else {
//fprintf(log,"ignored entity = %s\n", tok->tok_str);
}
}
else if (tok_val == ENTITY_NUM) {
if (prev_ch !=' ') *text++ = ' '; //2002.12.2
ent_proc = htmlTagEntity::entity_id_idx(tok->tok_realval);
if (ent_proc && ent_proc->conv[0]) {
#ifdef BUG
//prn_ent_proc(ent_proc);
#endif
strcpy(text, ent_proc->conv);
text += strlen(ent_proc->conv);
}
else {
if (tok->tok_realval < 0x80) {
// 2005.7.19
*text = (char)tok->tok_realval ;
text++;
}
}
}
else {
char *t = text;
if (is_in_PRE_tag || tok->tok_len>1 || tok_val == STRING || tok_val == NUMBER) {
//if (prev_tok_val == TAG_END && *text != ' ') *text++ = ' ';
// 2003. 3.20
//if (prev_tok_val == TAG_END && isalnum(prev_ch) )
// *text++ = ' ';
words_in_line++;
strcpy(text, tok->tok_str);
text += tok->tok_len;
}
// 2002.10.16 한글은 space로 전환
else if (tok->tok_val == FR_STR) {
if (do_prn_hangul) {
if (prev_ch != ' ') *text++ = ' ';
strcpy(text, tok->tok_str);
text += tok->tok_len;
words_in_line++;
}
else {
if (prev_ch != ' ') *text++ = ' ';
}
}
else if (tok->tok_len==1) {
if (curr_tag==htmlTagEntity::PRE_tag) {
#ifdef ODD_CHAR
if ( (tok_val & 0xF0) != 0x90)
*text++ = tok_val;
else if (tok_val==0x92) {
if (prev_tok_val != 0x92)
*text++ = '\'';
}
#else
*text++ = tok_val;
#endif
}
//else if (tok_val == '\n' || tok_val == '\r') {
else if (ISSPACE(tok_val)) {
if (prev_ch != ' ') *text++ = ' ';
}
else {
if (prev_ch==':' && tok_val==':') { }
else *text++ = tok_val;
}
}
if (is_in_A_tag)
hlink_chars += (int)(text - t);
}
prev_tok_val = tok->tok_val;
}// while(1)
*text = 0;
if ((int)(text-start) >= maxlen-5) {
PRN("parse(): too far !! maxlen=%d, %d\n", maxlen, text-start);
}
if ((int)(text-start) < maxlen-5)
memset(text, 0, 4);
#ifdef DEB
PRN("text=%X, start=%X, text=%d, start=%d\n",
text, start, (int)text % 10000, (int)start % 10000);
PRN("start[0]=%d text[0]=%d\n", start[0], text[0]);
#endif
return ((int)text-(int)start);
}
int main ()
{
PRN(Hello from main());
return 0 ;
}
static void meta(void)
{
register int i;
register char *s, *r;
register char *prn_bufp;
memcpy(set[0], "abcdefghijklmno", 15);
prn_bufp = prn_buf;
level = 1;
stack[0] = 1;
iter:
if (level == 0) {
PRN_LASTFLUSH();
return;
}
s = set[level-1];
if (level == n) {
PRN('w'); PRN('r'); PRN('i'); PRN('t'); PRN('e');
PRN('l'); PRN('n'); PRN('(');
PRN(s[0]);
for (i = 1; i < n; i++) {
PRN(',');
PRN(s[i]);
}
PRN(')'); PRN('\n');
level--;
goto iter;
}
r = set[level];
if ((k = stack[level - 1]) < 0) {
level--;
goto iter;
}
if (k > 0) {
if (k != level) {
/* printf("else "); */
PRN('e'); PRN('l'); PRN('s'); PRN('e'); PRN(' ');
}
/* printf("if %c < %c then\n", s[k-1], s[level]); */
PRN('i'); PRN('f'); PRN(' '); PRN(s[k-1]); PRN(' ');
PRN('<'); PRN(' '); PRN(s[level]); PRN(' ');
PRN('t'); PRN('h'); PRN('e'); PRN('n');
PRN('\n');
} else {
/* printf("else\n"); */
PRN('e'); PRN('l'); PRN('s'); PRN('e'); PRN('\n');
PRN_CHKFLUSH();
}
r[k] = s[level];
for (i = 0; i < k; i++)
*r++ = *s++;
for (r++; i < level; i++)
*r++ = *s++;
for (i++,s++; i < n; i++)
*r++ = *s++;
stack[level - 1]--;
stack[level] = level + 1;
level++;
goto iter;
}
/**
테스트 결과.
root가 다음과 같은 경우: "http://www.cnn.com/ASIA -D 1 -I /WORLD,/2005/SPORT"
ASIA를 화일로 인식함.
*/
int TravExpr::match(URLParse &linkurl)
{
int i, n;
int match=LINK_MATCH_UNDEF;
char *ptr;
URLst *hlink = &(linkurl.url);
URLst *rootURL = &(rootURLparse.url);
// host check
if (expr.host_span==0 && hlink->host &&
(strncmp(hlink->host, rootURL->host, rootURL->host_len)!=0))
{
if (debug) PRN("match(): false by host\n");
return LINK_MATCH_FALSE;
}
// exclude dir
for(i=0; i<(const int)expr.exc_dirs.num(); i++) {
if (isSubdir (hlink->path, expr.exc_dirs.str(i)) )
{
match = LINK_MATCH_FALSE;
if (debug) PRN("match(): false by exclude dir\n");
return LINK_MATCH_FALSE;
}
}
//화일 이름에 wild match $* $? :: keyword dir, inc dir에 대해 AND 관게 (relation)
if (expr.filename) {
if (StrCmp::URLwildcmp(expr.filename, hlink->file)==0)
match = LINK_MATCH_TRUE;
else {
match = LINK_MATCH_FALSE;
if (debug) PRN("match(): false by filename\n");
return LINK_MATCH_FALSE;
}
}
// keyword 를 포함하는 dir
for(i=0; i<(const int)expr.key_dirs.num(); i++) {
n = strlen(expr.key_dirs.str(i));
ptr = strstr(expr.key_dirs.str(i), hlink->path);
// dir 이름중에 keyword 가 나타나면..
if (ptr!=0 && !isalpha(ptr[n]) &&
(ptr==hlink->path || !isalpha( *(ptr-1) ) ) ) {
match = LINK_MATCH_TRUE; break;
}
}
if (match!=LINK_MATCH_TRUE)
{
//---- URL expr의 ROOT URL의 같은 dir이하만 포함 --> 변경
n = rootURL->dir_len;
if (n<=1 && rootURL->file_len!=0) {
// http://news.nationalgeographic.com/iraq.html 같은 경우
// '/' 이하가 다 포함되는 문제 발생.
// 그래서 DIR이 '/'인 경우는 URL type이 file 이 아닌 경우만 true 가능
match=LINK_MATCH_FALSE;
}
else if (n<=1 || linkurl.isSubDir(rootURLparse) )
{
match=LINK_MATCH_TRUE;
return LINK_MATCH_TRUE;
}
else match=LINK_MATCH_FALSE;
if (debug) PRN("match(): root path match = %d\n", match);
}
// include dir
if (match != LINK_MATCH_TRUE) {
// keyword dir에서 TRUE이면 여기 검사 필요없이 TRUE
// root URL과 link.path가 다른 경우 INC_DIR 검사
for(i=0; i<(const int)expr.inc_dirs.num(); i++) {
if (isSubdir (hlink->path, expr.inc_dirs.str(i)) )
{
match = LINK_MATCH_TRUE; break;
}
}
// include dir에 해당하는 dir만 (Only) (2002.9.1)
if (expr.inc_dirs.num() && match != LINK_MATCH_TRUE)
match = LINK_MATCH_FALSE;
}
return match;
}
