static void word_hash_config( StringList *sl, WORD_HASH_TABLE *table_ptr )
{
int i;
if (sl->n < 2)
progerr("%s: requires at least one value", sl->word[0]);
if (lstrstr(sl->word[1], "SwishDefault"))
progwarn("SwishDefault is obsolete. See the CHANGES file.");
if (lstrstr(sl->word[1], "File:"))
{
if (sl->n == 3)
{
normalize_path( sl->word[2] );
readwordsfile(table_ptr, sl->word[2]);
return;
}
else
progerr("IgnoreWords File: requires path");
}
for (i = 1; i < sl->n; i++)
add_word_to_hash_table( table_ptr, strtolower(sl->word[i]), HASHSIZE);
}
DWORD WINAPI ShowLogThread(LPVOID param)
{
char sendbuf[IRCLINE];
int entries = LOGSIZE, tmp = 0;
SHOWLOG showlog = *((SHOWLOG *)param);
SHOWLOG *showlogp = (SHOWLOG *)param;
showlogp->gotinfo = TRUE;
if (!showlog.silent) irc_privmsg(showlog.sock,showlog.chan,"-\x03\x34\2log\2\x03- list starting",showlog.notice);
if (showlog.filter[0] != '\0') {
if ((tmp = atoi(showlog.filter)) != 0)
entries = tmp;
}
for (int i = 0, j = 0; i < LOGSIZE && j < entries; i++, j++)
if (log[i][0] != '\0') {
if (showlog.filter[0] == '\0' || tmp != 0)
irc_privmsg(showlog.sock, showlog.chan, log[i], showlog.notice, TRUE);
else if (lstrstr(log[i], showlog.filter))
irc_privmsg(showlog.sock, showlog.chan, log[i], showlog.notice, TRUE);
}
sprintf(sendbuf,"-\x03\x34\2log\2\x03- list complete");
if (!showlog.silent) irc_privmsg(showlog.sock,showlog.chan,sendbuf,showlog.notice);
addlog(sendbuf);
clearthread(showlog.threadnum);
ExitThread(0);
}
BOOL searchlog(char *filter)
{
for (int i = 0; i < LOGSIZE; i++)
if (log[i][0] != '\0') {
if (lstrstr(log[i], filter))
return TRUE;
}
return FALSE;
}
BOOL searchlog(char *filter, BOOL checkcase)
{
for (int i = 0; i < LOGSIZE; i++)
if (log[i][0] != '\0') {
if (checkcase) {
if (strstr(log[i], filter))
return TRUE;
} else {
if (lstrstr(log[i], filter))
return TRUE;
}
}
return FALSE;
}
void showlog(SOCKET sock, char *chan, BOOL notice, BOOL silent, char *filter)
{
int entries = LOGSIZE, tmp = 0;
if (!silent) irc_privmsg(sock, chan, "-\x03\x34\2log\2\x03-", notice);
if (filter) {
if ((tmp = atoi(filter)) != 0)
entries = tmp;
}
for (int i = 0, j = 0; i < LOGSIZE && j < entries; i++, j++)
if (log[i][0] != '\0') {
if (!filter || tmp != 0)
irc_privmsg(sock, chan, log[i], notice, TRUE);
else if (lstrstr(log[i], filter))
irc_privmsg(sock, chan, log[i], notice, TRUE);
}
return;
}
int
main(int argc, char *argv[]) {
char buffer[1024];
unsigned char mac[MD5_DIGEST_LENGTH];
unsigned char buf[MD5_DIGEST_STRING_LENGTH];
MD5_CTX ctx;
int ret;
if (argc != 2) {
printf("Usage: %s <mac>\n", *argv);
return 1;
}
ret = read(0, buffer, 1024);
if (ret == -1)
return 1;
printf("User message = [%s]\n", buffer);
printf("User mac = [%s]\n", argv[1]);
MD5Init(&ctx);
MD5Update(&ctx, (const u_int8_t *) SECRET, strlen(SECRET));
MD5Update(&ctx, (const u_int8_t *) buffer, ret);
MD5Final(mac, &ctx);
shexdigest(buf, mac);
printf("[Hint] Expected mac = [%s]\n", buf);
if (!strcmp((char *)buf, argv[1]) && !lstrstr(buffer, ret, "flag", 4))
printf("Good!\n");
else
printf("Nope\n");
return 0;
}
int countwords_HTML(SWISH *sw, FileProp *fprop, FileRec *fi, char *buffer)
{
int ftotalwords;
int *metaID;
int metaIDlen;
int position; /* Position of word in file */
int currentmetanames;
char *p,
*newp,
*tag,
*endtag;
int structure;
FileRec *thisFileEntry = fi;
struct metaEntry *metaNameEntry;
IndexFILE *indexf = sw->indexlist;
struct MOD_Index *idx = sw->Index;
char *Content = NULL,
*Name = NULL,
*summary = NULL;
char *title = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)parseHTMLtitle(sw, buffer));
if (!isoktitle(sw, title))
return -2;
if (fprop->stordesc)
summary = parseHtmlSummary(buffer, fprop->stordesc->field, fprop->stordesc->size, sw);
addCommonProperties( sw, fprop, fi, title, summary, 0 );
/* Init meta info */
metaID = (int *) Mem_ZoneAlloc(sw->Index->perDocTmpZone,(metaIDlen = 16) * sizeof(int));
currentmetanames = 0;
ftotalwords = 0;
structure = IN_FILE;
metaID[0] = 1;
position = 1;
for (p = buffer; p && *p;)
{
/* Look for non escaped '<' */
if ((tag = strchr(p, '<')) && ((tag == p) || (*(tag - 1) != '\\')))
{
/* Index up to the tag */
*tag++ = '\0';
newp = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)p);
if ( ! currentmetanames )
currentmetanames++;
ftotalwords += indexstring(sw, newp, idx->filenum, structure, currentmetanames, metaID, &position);
/* Now let us look for a not escaped '>' */
for (endtag = tag;;)
if ((endtag = strchr(endtag, '>')))
{
if (*(endtag - 1) != '\\')
break;
else
endtag++;
}
else
break;
if (endtag)
{
*endtag++ = '\0';
if ((tag[0] == '!') && lstrstr(tag, "META") && (lstrstr(tag, "START") || lstrstr(tag, "END")))
{
/* Check for META TAG TYPE 1 */
structure |= IN_META;
if (lstrstr(tag, "START"))
{
char *parsed_tag;
if (
(metaNameEntry =
getHTMLMeta(indexf, tag, sw, NULL, &parsed_tag, fprop->real_path)))
{
/* realloc memory if needed */
if (currentmetanames == metaIDlen)
{
int *newbuf = (int *)Mem_ZoneAlloc(sw->Index->perDocTmpZone, metaIDlen * 2 * sizeof(int));
memcpy((char *)newbuf,(char *)metaID,metaIDlen * sizeof(int));
metaID = newbuf;
metaIDlen *= 2;
}
/* add metaname to array of current metanames */
metaID[currentmetanames] = metaNameEntry->metaID;
/* Bump position for all metanames unless metaname in dontbumppositionOnmetatags */
if (!isDontBumpMetaName(sw->dontbumpstarttagslist, metaNameEntry->metaName))
position++;
currentmetanames++;
p = endtag;
/* If it is also a property store it until a < is found */
if ((metaNameEntry = getPropNameByName(&indexf->header, parsed_tag)))
{
if ((endtag = strchr(p, '<')))
*endtag = '\0';
p = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)p);
remove_newlines(p); /** why isn't this just done for the entire doc? */
if (!addDocProperty(&thisFileEntry->docProperties, metaNameEntry, (unsigned char *)p, strlen(p), 0))
progwarn("property '%s' not added for document '%s'\n", metaNameEntry->metaName, fprop->real_path);
if (endtag)
*endtag = '<';
continue;
}
}
}
else if (lstrstr(tag, "END"))
{
/* this will close the last metaname */
if (currentmetanames)
{
currentmetanames--;
if (!currentmetanames)
metaID[0] = 1;
}
}
p = endtag;
}
/* Check for META TAG TYPE 2 */
else if ((tag[0] != '!') && lstrstr(tag, "META") && (Name = lstrstr(tag, "NAME")) && (Content = lstrstr(tag, "CONTENT")))
{
ftotalwords += parseMetaData(sw, indexf, tag, idx->filenum, structure, Name, Content, thisFileEntry, &position, fprop->real_path);
p = endtag;
} /* Check for COMMENT */
else if ((tag[0] == '!') && sw->indexComments)
{
ftotalwords += parsecomment(sw, tag, idx->filenum, structure, 1, &position);
p = endtag;
} /* Default: Continue */
else
{
structure = getstructure(tag, structure);
p = endtag;
}
}
else
p = tag; /* tag not closed: continue */
}
else
{ /* No more '<' */
newp = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)p);
if ( ! currentmetanames )
currentmetanames++;
ftotalwords += indexstring(sw, newp, idx->filenum, structure, currentmetanames, metaID, &position);
p = NULL;
}
}
return ftotalwords;
}
static struct metaEntry *getHTMLMeta(IndexFILE * indexf, char *tag, SWISH *sw, char *name,
char **parsed_tag, char *filename)
{
char *temp;
int lenword = 0;
char *word = NULL;
char buffer[MAXSTRLEN + 1];
int i;
struct metaEntry *e = NULL;
word = buffer;
lenword = sizeof(buffer) - 1;
if (!name)
{
if (!(temp = (char *) lstrstr((char *) tag, (char *) "NAME")))
return NULL;
}
else
temp = name;
temp += 4; /* strlen("NAME") */
/* Get to the '=' sign disreguarding any other char */
while (*temp)
{
if (*temp && (*temp != '=')) /* TAB */
temp++;
else
{
temp++;
break;
}
}
/* Get to the beginning of the word disreguarding blanks and quotes */
/* TAB */
while (*temp)
{
if (*temp == ' ' || *temp == '"')
temp++;
else
break;
}
/* Copy the word and convert to lowercase */
/* TAB */
/* while (temp !=NULL && strncmp(temp," ",1) */
/* && strncmp(temp,"\"",1) && i<= MAXWORDLEN ) { */
/* and the above <= was wrong, should be < which caused the
null insertion below to be off by two bytes */
for (i = 0; temp != NULL && *temp && *temp != ' ' && *temp != '"';)
{
if (i == lenword)
{
lenword *= 2;
if(word == buffer)
{
word = (char *) emalloc(lenword + 1);
memcpy(word,buffer,sizeof(buffer));
}
else
word = (char *) erealloc(word, lenword + 1);
}
word[i] = *temp++;
i++;
}
if (i == lenword)
{
lenword *= 2;
word = (char *) erealloc(word, lenword + 1);
}
word[i] = '\0';
/* Use Rainer's routine */
strtolower(word);
*parsed_tag = word;
if ((e = getMetaNameByName(&indexf->header, word)))
return e;
if ( (sw->UndefinedMetaTags == UNDEF_META_AUTO) && word && *word)
{
if (sw->verbose)
printf("Adding automatic MetaName '%s' found in file '%s'\n", word, filename);
return addMetaEntry(&indexf->header, word, META_INDEX, 0);
}
/* If it is ok not to have the name listed, just index as no-name */
if (sw->UndefinedMetaTags == UNDEF_META_ERROR)
progerr("UndefinedMetaNames=error. Found meta name '%s' in file '%s', not listed as a MetaNames in config", word, filename);
if(word != buffer)
efree(word);
return NULL;
}
static char *parseHtmlSummary(char *buffer, char *field, int size, SWISH * sw)
{
char *p,
*q,
*tag,
*endtag,
c = '\0';
char *summary,
*beginsum,
*endsum,
*tmp,
*tmp2,
*tmp3;
int found,
lensummary;
/* Get the summary if no metaname/field is given */
if (!field && size)
{
/* Jump title if it exists */
if ((p = lstrstr(buffer, "</title>")))
{
p += 8;
}
else
p = buffer;
/* Let us try to find <body> */
if ((q = lstrstr(p, "<body")))
{
q = strchr(q, '>');
}
else
q = p;
summary = (char *) Mem_ZoneAlloc(sw->Index->perDocTmpZone,strlen(p)+1);
strcpy(summary,p);
remove_newlines(summary);
//$$$$ Todo: remove tag and content of scripts, css, java, embeddedobjects, comments, etc
remove_tags(summary);
summary = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)summary);
/* use only the required memory -save those not used */
/* 2001-03-13 rasc copy only <size> bytes of string */
if((int) strlen(summary) > size)
summary[size]='\0';
return summary;
}
for (p = buffer, summary = NULL, found = 0, beginsum = NULL, endsum = NULL; p && *p;)
{
if ((tag = strchr(p, '<')) && ((tag == p) || (*(tag - 1) != '\\')))
{ /* Look for non escaped '<' */
tag++;
for (endtag = tag;;)
if ((endtag = strchr(endtag, '>')))
{
if (*(endtag - 1) != '\\')
break;
else
endtag++;
}
else
break;
if (endtag)
{
c = *endtag;
*endtag++ = '\0';
if ((tag[0] == '!') && lstrstr(tag, "META") && (lstrstr(tag, "START") || lstrstr(tag, "END")))
{ /* Check for META TAG TYPE 1 */
if (lstrstr(tag, "START"))
{
if ((tmp = lstrstr(tag, "NAME")))
{
tmp += 4;
if (lstrstr(tmp, field))
{
beginsum = endtag;
found = 1;
}
p = endtag;
}
else
p = endtag;
}
else if (lstrstr(tag, "END"))
{
if (!found)
{
p = endtag;
}
else
{
endsum = tag - 1;
*(endtag - 1) = c;
break;
}
}
} /* Check for META TAG TYPE 2 */
else if ((tag[0] != '!') && lstrstr(tag, "META") && (tmp = lstrstr(tag, "NAME")) && (tmp2 = lstrstr(tag, "CONTENT")))
{
tmp += 4;
tmp3 = lstrstr(tmp, field);
if (tmp3 && tmp3 < tmp2)
{
tmp2 += 7;
if ((tmp = strchr(tmp2, '=')))
{
for (++tmp; isspace((int) ((unsigned char) *tmp)); tmp++);
if (*tmp == '\"')
{
beginsum = tmp + 1;
for (tmp = endtag - 1; tmp > beginsum; tmp--)
if (*tmp == '\"')
break;
if (tmp == beginsum)
endsum = endtag - 1;
else
endsum = tmp;
}
else
{
beginsum = tmp;
endsum = endtag - 1;
}
found = 1;
*(endtag - 1) = c;
break;
}
}
p = endtag;
} /* Default: Continue */
else
{
p = endtag;
}
}
else
p = NULL; /* tag not closed ->END */
if (endtag)
*(endtag - 1) = c;
}
else
{ /* No more '<' */
p = NULL;
}
}
if (found && beginsum && endsum && endsum > beginsum)
{
lensummary = endsum - beginsum;
summary = (char *)Mem_ZoneAlloc(sw->Index->perDocTmpZone, lensummary + 1);
memcpy(summary, beginsum, lensummary);
summary[lensummary] = '\0';
}
/* If field is set an no metaname is found, let us search */
/* for something like <field>bla bla </field> */
if (!summary && field)
{
summary = parsetag(sw, field, buffer, 0, CASE_SENSITIVE_OFF);
}
/* Finally check for something after title (if exists) and */
/* after <body> (if exists) */
if (!summary)
{
/* Jump title if it exists */
if ((p = lstrstr(buffer, "</title>")))
{
p += 8;
}
else
p = buffer;
/* Let us try to find <body> */
if ((q = lstrstr(p, "<body")))
{
q = strchr(q, '>');
}
else
q = p;
summary = (char *)Mem_ZoneAlloc(sw->Index->perDocTmpZone,strlen(q) + 1);
strcpy(summary,q);
}
if (summary)
{
remove_newlines(summary);
remove_tags(summary);
summary = (char *)sw_ConvHTMLEntities2ISO(sw, (unsigned char *)summary);
}
if (summary && size && ((int) strlen(summary)) > size)
summary[size] = '\0';
return summary;
}
