2011-11-14 15:57:52 +01:00
|
|
|
/*
|
2011-11-16 14:47:47 +01:00
|
|
|
* parser_utils.c
|
2011-11-14 15:57:52 +01:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <ctype.h>
|
|
|
|
#include <sys/socket.h>
|
|
|
|
#include <sys/types.h>
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#include <netinet/in.h>
|
|
|
|
#include <arpa/inet.h>
|
|
|
|
#include <fcntl.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
#include <time.h>
|
|
|
|
#include <piler.h>
|
|
|
|
#include "trans.h"
|
|
|
|
#include "html.h"
|
|
|
|
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
void init_state(struct _state *state){
|
2011-11-14 15:57:52 +01:00
|
|
|
int i;
|
|
|
|
|
|
|
|
state->message_state = MSG_UNDEF;
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
state->line_num = 0;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
state->is_header = 1;
|
2011-11-22 12:31:54 +01:00
|
|
|
state->is_1st_header = 1;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
state->textplain = 1; /* by default we are a text/plain message */
|
2011-11-14 15:57:52 +01:00
|
|
|
state->texthtml = 0;
|
|
|
|
state->message_rfc822 = 0;
|
|
|
|
|
|
|
|
state->base64 = 0;
|
|
|
|
state->utf8 = 0;
|
|
|
|
|
|
|
|
state->qp = 0;
|
|
|
|
|
|
|
|
state->htmltag = 0;
|
|
|
|
state->style = 0;
|
|
|
|
|
|
|
|
state->skip_html = 0;
|
|
|
|
|
|
|
|
state->content_type_is_set = 0;
|
|
|
|
|
|
|
|
memset(state->message_id, 0, SMALLBUFSIZE);
|
2011-11-16 14:47:47 +01:00
|
|
|
memset(state->miscbuf, 0, MAX_TOKEN_LEN);
|
2011-12-07 15:24:52 +01:00
|
|
|
memset(state->qpbuf, 0, MAX_TOKEN_LEN);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-22 12:31:54 +01:00
|
|
|
memset(state->filename, 0, TINYBUFSIZE);
|
|
|
|
memset(state->type, 0, TINYBUFSIZE);
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
state->has_to_dump = 0;
|
|
|
|
state->fd = -1;
|
2011-11-19 21:25:44 +01:00
|
|
|
state->mfd = -1;
|
2011-11-14 15:57:52 +01:00
|
|
|
state->realbinary = 0;
|
2011-11-16 14:47:47 +01:00
|
|
|
state->octetstream = 0;
|
2011-11-22 12:31:54 +01:00
|
|
|
state->pushed_pointer = 0;
|
2011-11-19 21:25:44 +01:00
|
|
|
state->saved_size = 0;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
state->boundaries = NULL;
|
2011-11-28 14:21:14 +01:00
|
|
|
state->rcpt = NULL;
|
2012-01-14 09:53:26 +01:00
|
|
|
state->rcpt_domain = NULL;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
state->n_attachments = 0;
|
|
|
|
|
|
|
|
for(i=0; i<MAX_ATTACHMENTS; i++){
|
|
|
|
state->attachments[i].size = 0;
|
2011-11-16 14:47:47 +01:00
|
|
|
memset(state->attachments[i].type, 0, TINYBUFSIZE);
|
|
|
|
memset(state->attachments[i].filename, 0, TINYBUFSIZE);
|
2011-11-19 21:25:44 +01:00
|
|
|
memset(state->attachments[i].internalname, 0, TINYBUFSIZE);
|
|
|
|
memset(state->attachments[i].digest, 0, 2*DIGEST_LENGTH+1);
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
memset(state->b_from, 0, SMALLBUFSIZE);
|
2012-01-14 09:53:26 +01:00
|
|
|
memset(state->b_from_domain, 0, SMALLBUFSIZE);
|
2012-01-29 09:02:21 +01:00
|
|
|
memset(state->b_to, 0, MAXBUFSIZE);
|
2012-01-14 09:53:26 +01:00
|
|
|
memset(state->b_to_domain, 0, SMALLBUFSIZE);
|
2011-11-14 15:57:52 +01:00
|
|
|
memset(state->b_subject, 0, MAXBUFSIZE);
|
|
|
|
memset(state->b_body, 0, BIGBUFSIZE);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2012-01-07 00:00:36 +01:00
|
|
|
unsigned long parse_date_header(char *datestr){
|
|
|
|
int n=0;
|
|
|
|
char *p, *q, *r, s[SMALLBUFSIZE];
|
2011-11-16 14:47:47 +01:00
|
|
|
unsigned long ts=0;
|
|
|
|
struct tm tm;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2012-01-07 00:00:36 +01:00
|
|
|
datestr += 5;
|
|
|
|
p = datestr;
|
2011-11-16 14:47:47 +01:00
|
|
|
|
|
|
|
|
2012-01-07 00:00:36 +01:00
|
|
|
for(; *datestr; datestr++){
|
|
|
|
if(isspace(*datestr)) *datestr = ' ';
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
if(*p == ' '){ p++; }
|
|
|
|
|
|
|
|
do {
|
|
|
|
p = split_str(p, " ", s, sizeof(s)-1);
|
|
|
|
if(strlen(s) > 0){
|
|
|
|
n++;
|
|
|
|
|
|
|
|
q = strchr(s, ','); if(q) *q='\0';
|
|
|
|
|
|
|
|
if(strlen(s) <= 2){ tm.tm_mday = atoi(s); continue; }
|
|
|
|
|
|
|
|
if(strlen(s) == 4){ tm.tm_year = atoi(s) - 1900; continue; }
|
|
|
|
|
|
|
|
if(strlen(s) == 3){
|
|
|
|
if(strcmp(s, "Mon") == 0) tm.tm_wday = 1;
|
|
|
|
else if(strcmp(s, "Tue") == 0) tm.tm_wday = 2;
|
|
|
|
else if(strcmp(s, "Wed") == 0) tm.tm_wday = 3;
|
|
|
|
else if(strcmp(s, "Thu") == 0) tm.tm_wday = 4;
|
|
|
|
else if(strcmp(s, "Fri") == 0) tm.tm_wday = 5;
|
|
|
|
else if(strcmp(s, "Sat") == 0) tm.tm_wday = 6;
|
|
|
|
else if(strcmp(s, "Sun") == 0) tm.tm_wday = 0;
|
|
|
|
|
|
|
|
|
|
|
|
if(strcmp(s, "Jan") == 0) tm.tm_mon = 0;
|
|
|
|
else if(strcmp(s, "Feb") == 0) tm.tm_mon = 1;
|
|
|
|
else if(strcmp(s, "Mar") == 0) tm.tm_mon = 2;
|
|
|
|
else if(strcmp(s, "Apr") == 0) tm.tm_mon = 3;
|
|
|
|
else if(strcmp(s, "May") == 0) tm.tm_mon = 4;
|
|
|
|
else if(strcmp(s, "Jun") == 0) tm.tm_mon = 5;
|
|
|
|
else if(strcmp(s, "Jul") == 0) tm.tm_mon = 6;
|
|
|
|
else if(strcmp(s, "Aug") == 0) tm.tm_mon = 7;
|
|
|
|
else if(strcmp(s, "Sep") == 0) tm.tm_mon = 8;
|
|
|
|
else if(strcmp(s, "Oct") == 0) tm.tm_mon = 9;
|
|
|
|
else if(strcmp(s, "Nov") == 0) tm.tm_mon = 10;
|
|
|
|
else if(strcmp(s, "Dec") == 0) tm.tm_mon = 11;
|
|
|
|
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(strlen(s) == 8){
|
|
|
|
r = &s[0];
|
|
|
|
|
|
|
|
q = strchr(r, ':'); if(!q) break;
|
|
|
|
*q = '\0'; tm.tm_hour = atoi(r); r = q+1;
|
|
|
|
|
|
|
|
q = strchr(r, ':'); if(!q) break;
|
|
|
|
*q = '\0'; tm.tm_min = atoi(r); r = q+1;
|
|
|
|
|
|
|
|
tm.tm_sec = atoi(r);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
} while(p);
|
|
|
|
|
|
|
|
tm.tm_isdst = -1;
|
2011-11-16 14:47:47 +01:00
|
|
|
ts = mktime(&tm);
|
|
|
|
|
|
|
|
return ts;
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
int isHexNumber(char *p){
|
|
|
|
for(; *p; p++){
|
|
|
|
if(!(*p == '-' || (*p >= 0x30 && *p <= 0x39) || (*p >= 0x41 && *p <= 0x46) || (*p >= 0x61 && *p <= 0x66)) )
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
int extract_boundary(char *p, struct _state *state){
|
|
|
|
char *q;
|
|
|
|
|
|
|
|
p += strlen("boundary");
|
|
|
|
|
|
|
|
q = strchr(p, '"');
|
|
|
|
if(q) *q = ' ';
|
|
|
|
|
|
|
|
/*
|
|
|
|
* I've seen an idiot spammer using the following boundary definition in the header:
|
|
|
|
*
|
|
|
|
* Content-Type: multipart/alternative;
|
|
|
|
* boundary=3D"b1_52b92b01a943615aff28b7f4d2f2d69d"
|
|
|
|
*/
|
|
|
|
|
|
|
|
if(strncmp(p, "=3D", 3) == 0){
|
|
|
|
*(p+3) = '=';
|
|
|
|
p += 3;
|
|
|
|
}
|
|
|
|
|
|
|
|
p = strchr(p, '=');
|
|
|
|
if(p){
|
|
|
|
p++;
|
|
|
|
for(; *p; p++){
|
|
|
|
if(isspace(*p) == 0)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
q = strrchr(p, '"');
|
|
|
|
if(q) *q = '\0';
|
|
|
|
|
|
|
|
q = strrchr(p, '\r');
|
|
|
|
if(q) *q = '\0';
|
|
|
|
|
|
|
|
q = strrchr(p, '\n');
|
|
|
|
if(q) *q = '\0';
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
append_list(&(state->boundaries), p);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void fixupEncodedHeaderLine(char *buf){
|
2011-11-28 14:21:14 +01:00
|
|
|
char *sb, *sq, *p, *q, *r, *s, v[SMALLBUFSIZE], puf[MAXBUFSIZE];
|
2011-11-14 15:57:52 +01:00
|
|
|
char *start, *end;
|
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
memset(puf, 0, sizeof(puf));
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
q = buf;
|
|
|
|
|
|
|
|
do {
|
2011-11-28 14:21:14 +01:00
|
|
|
q = split_str(q, " ", v, sizeof(v)-1);
|
|
|
|
|
|
|
|
p = v;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
do {
|
|
|
|
start = strstr(p, "=?");
|
|
|
|
if(start){
|
2011-11-28 14:21:14 +01:00
|
|
|
*start = '\0';
|
|
|
|
if(strlen(p) > 0){
|
|
|
|
strncat(puf, p, sizeof(puf)-1);
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
start++;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
s = NULL;
|
|
|
|
sb = strcasestr(start, "?B?"); if(sb) s = sb;
|
|
|
|
sq = strcasestr(start, "?Q?"); if(sq) s = sq;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
if(s){
|
|
|
|
end = strstr(s+3, "?=");
|
|
|
|
if(end){
|
|
|
|
*end = '\0';
|
2011-12-05 17:18:03 +01:00
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
if(sb){ decodeBase64(s+3); }
|
|
|
|
if(sq){ decodeQP(s+3); r = s + 3; for(; *r; r++){ if(*r == '_') *r = ' '; } }
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-12-07 15:24:52 +01:00
|
|
|
/* encode everything if it's not utf-8 encoded */
|
|
|
|
if(strncasecmp(start+1, "utf-8", 5)) utf8_encode((unsigned char*)s+3);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
strncat(puf, s+3, sizeof(puf)-1);
|
|
|
|
|
|
|
|
p = end + 2;
|
|
|
|
}
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
else {
|
2011-11-28 14:21:14 +01:00
|
|
|
strncat(puf, start, sizeof(puf)-1);
|
|
|
|
|
|
|
|
break;
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
}
|
2011-11-28 14:21:14 +01:00
|
|
|
else {
|
|
|
|
strncat(puf, p, sizeof(puf)-1);
|
|
|
|
break;
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
} while(p);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
if(q) strncat(puf, " ", sizeof(puf)-1);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
} while(q);
|
|
|
|
|
|
|
|
snprintf(buf, MAXBUFSIZE-1, "%s", puf);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-12-07 15:24:52 +01:00
|
|
|
void fixupSoftBreakInQuotedPritableLine(char *buf, struct _state *state){
|
|
|
|
int i=0;
|
|
|
|
char *p, puf[MAXBUFSIZE];
|
|
|
|
|
|
|
|
if(strlen(state->qpbuf) > 0){
|
|
|
|
memset(puf, 0, MAXBUFSIZE);
|
|
|
|
strncpy(puf, state->qpbuf, MAXBUFSIZE-1);
|
|
|
|
strncat(puf, buf, MAXBUFSIZE-1);
|
|
|
|
|
|
|
|
memset(buf, 0, MAXBUFSIZE);
|
|
|
|
memcpy(buf, puf, MAXBUFSIZE);
|
|
|
|
|
|
|
|
memset(state->qpbuf, 0, MAX_TOKEN_LEN);
|
|
|
|
}
|
|
|
|
|
|
|
|
if(buf[strlen(buf)-1] == '='){
|
|
|
|
buf[strlen(buf)-1] = '\0';
|
|
|
|
i = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(i == 1){
|
|
|
|
p = strrchr(buf, ' ');
|
|
|
|
if(p){
|
|
|
|
memset(state->qpbuf, 0, MAX_TOKEN_LEN);
|
|
|
|
if(strlen(p) < MAX_TOKEN_LEN-1){
|
|
|
|
//snprintf(state->qpbuf, MAX_TOKEN_LEN-1, "%s", p);
|
|
|
|
memcpy(&(state->qpbuf[0]), p, MAX_TOKEN_LEN-1);
|
|
|
|
|
|
|
|
*p = '\0';
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
void fixupBase64EncodedLine(char *buf, struct _state *state){
|
|
|
|
char *p, puf[MAXBUFSIZE];
|
|
|
|
|
|
|
|
if(strlen(state->miscbuf) > 0){
|
2011-11-16 14:47:47 +01:00
|
|
|
memset(puf, 0, sizeof(puf));
|
|
|
|
strncpy(puf, state->miscbuf, sizeof(puf)-1);
|
|
|
|
strncat(puf, buf, sizeof(puf)-1);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
memset(buf, 0, MAXBUFSIZE);
|
|
|
|
memcpy(buf, puf, MAXBUFSIZE);
|
|
|
|
|
|
|
|
memset(state->miscbuf, 0, MAX_TOKEN_LEN);
|
|
|
|
}
|
|
|
|
|
|
|
|
if(buf[strlen(buf)-1] != '\n'){
|
|
|
|
p = strrchr(buf, ' ');
|
|
|
|
if(p){
|
2011-11-16 14:47:47 +01:00
|
|
|
//strncpy(state->miscbuf, p+1, MAX_TOKEN_LEN-1);
|
|
|
|
memcpy(&(state->miscbuf[0]), p+1, MAX_TOKEN_LEN-1);
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
*p = '\0';
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void markHTML(char *buf, struct _state *state){
|
|
|
|
char *s, puf[MAXBUFSIZE], html[SMALLBUFSIZE];
|
|
|
|
int k=0, j=0, pos=0;
|
|
|
|
|
|
|
|
memset(puf, 0, MAXBUFSIZE);
|
|
|
|
memset(html, 0, SMALLBUFSIZE);
|
|
|
|
|
|
|
|
s = buf;
|
|
|
|
|
|
|
|
for(; *s; s++){
|
|
|
|
if(*s == '<'){
|
|
|
|
state->htmltag = 1;
|
|
|
|
puf[k] = ' ';
|
|
|
|
k++;
|
|
|
|
|
|
|
|
memset(html, 0, SMALLBUFSIZE); j=0;
|
|
|
|
|
|
|
|
pos = 0;
|
|
|
|
|
|
|
|
//printf("start html:%c\n", *s);
|
|
|
|
}
|
|
|
|
|
|
|
|
if(state->htmltag == 1){
|
|
|
|
|
|
|
|
if(j == 0 && *s == '!'){
|
|
|
|
state->skip_html = 1;
|
|
|
|
//printf("skiphtml=1\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
if(state->skip_html == 0){
|
|
|
|
if(*s != '>' && *s != '<' && *s != '"'){
|
|
|
|
//printf("j=%d/%c", j, *s);
|
|
|
|
html[j] = tolower(*s);
|
|
|
|
if(j < SMALLBUFSIZE-10) j++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(isspace(*s)){
|
|
|
|
if(j > 0){
|
|
|
|
k += appendHTMLTag(puf, html, pos, state);
|
|
|
|
memset(html, 0, SMALLBUFSIZE); j=0;
|
|
|
|
}
|
|
|
|
pos++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
if(state->style == 0){
|
|
|
|
puf[k] = *s;
|
|
|
|
k++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if(*s == '>'){
|
|
|
|
state->htmltag = 0;
|
|
|
|
state->skip_html = 0;
|
|
|
|
|
|
|
|
//printf("skiphtml=0\n");
|
|
|
|
//printf("end html:%c\n", *s);
|
|
|
|
|
|
|
|
//strncat(html, " ", SMALLBUFSIZE-1);
|
|
|
|
|
|
|
|
if(j > 0){
|
|
|
|
strncat(html, " ", SMALLBUFSIZE-1);
|
|
|
|
k += appendHTMLTag(puf, html, pos, state);
|
|
|
|
memset(html, 0, SMALLBUFSIZE); j=0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
//printf("append last in line:*%s*, html=+%s+, j=%d\n", puf, html, j);
|
|
|
|
if(j > 0){ k += appendHTMLTag(puf, html, pos, state); }
|
|
|
|
|
|
|
|
strcpy(buf, puf);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
int appendHTMLTag(char *buf, char *htmlbuf, int pos, struct _state *state){
|
|
|
|
char *p, html[SMALLBUFSIZE];
|
|
|
|
int len;
|
|
|
|
|
|
|
|
if(pos == 0 && strncmp(htmlbuf, "style ", 6) == 0) state->style = 1;
|
|
|
|
if(pos == 0 && strncmp(htmlbuf, "/style ", 7) == 0) state->style = 0;
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
return 0;
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
//printf("appendHTML: pos:%d, +%s+\n", pos, htmlbuf);
|
|
|
|
|
|
|
|
if(state->style == 1) return 0;
|
|
|
|
|
|
|
|
if(strlen(htmlbuf) == 0) return 0;
|
|
|
|
|
|
|
|
snprintf(html, SMALLBUFSIZE-1, "HTML*%s", htmlbuf);
|
|
|
|
len = strlen(html);
|
|
|
|
|
|
|
|
if(len > 8 && strchr(html, '=')){
|
|
|
|
p = strstr(html, "cid:");
|
|
|
|
if(p){
|
|
|
|
*(p+3) = '\0';
|
|
|
|
strncat(html, " ", SMALLBUFSIZE-1);
|
|
|
|
}
|
|
|
|
|
|
|
|
strncat(buf, html, MAXBUFSIZE-1);
|
|
|
|
return len;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(strstr(html, "http") ){
|
|
|
|
strncat(buf, html+5, MAXBUFSIZE-1);
|
|
|
|
return len-5;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void translateLine(unsigned char *p, struct _state *state){
|
|
|
|
int url=0;
|
|
|
|
|
|
|
|
for(; *p; p++){
|
|
|
|
|
|
|
|
if( (state->message_state == MSG_RECEIVED || state->message_state == MSG_FROM || state->message_state == MSG_TO || state->message_state == MSG_CC) && *p == '@'){ continue; }
|
|
|
|
|
2011-11-28 14:21:14 +01:00
|
|
|
if(state->message_state == MSG_SUBJECT && (*p == '%' || *p == '_' || *p == '&') ){ continue; }
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
if(state->message_state == MSG_CONTENT_TYPE && *p == '_' ){ continue; }
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
if(*p == '.' || *p == '-'){ continue; }
|
2011-11-14 15:57:52 +01:00
|
|
|
|
|
|
|
if(strncasecmp((char *)p, "http://", 7) == 0){ p += 7; url = 1; continue; }
|
|
|
|
if(strncasecmp((char *)p, "https://", 8) == 0){ p += 8; url = 1; continue; }
|
|
|
|
|
|
|
|
if(url == 1 && (*p == '.' || *p == '-' || *p == '_' || *p == '/' || isalnum(*p)) ) continue;
|
|
|
|
if(url == 1) url = 0;
|
|
|
|
|
|
|
|
if(state->texthtml == 1 && state->message_state == MSG_BODY && strncmp((char *)p, "HTML*", 5) == 0){
|
|
|
|
p += 5;
|
|
|
|
while(isspace(*p) == 0){
|
|
|
|
p++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-12-07 15:24:52 +01:00
|
|
|
if(delimiter_characters[(unsigned int)*p] != ' ')
|
2011-11-14 15:57:52 +01:00
|
|
|
*p = ' ';
|
|
|
|
else {
|
2011-12-09 15:24:15 +01:00
|
|
|
// commented out because it breaks utf-8 encoding, 2011.12.07.
|
|
|
|
//*p = tolower(*p);
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-12-05 17:18:03 +01:00
|
|
|
void fix_email_address_for_sphinx(char *s){
|
|
|
|
for(; *s; s++){
|
2012-01-28 20:52:13 +01:00
|
|
|
if(*s == '@' || *s == '.' || *s == '+' || *s == '-' || *s == '_') *s = 'X';
|
2011-12-05 17:18:03 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2012-01-09 23:15:39 +01:00
|
|
|
int does_it_seem_like_an_email_address(char *email){
|
|
|
|
char *p;
|
|
|
|
|
|
|
|
if(email == NULL) return 0;
|
|
|
|
|
|
|
|
if(strlen(email) < 5) return 0;
|
|
|
|
|
|
|
|
p = strchr(email, '@');
|
|
|
|
if(!p) return 0;
|
|
|
|
|
|
|
|
if(strlen(p+1) < 3) return 0;
|
|
|
|
|
|
|
|
if(!strchr(p+1, '.')) return 0;
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-11-14 15:57:52 +01:00
|
|
|
/*
|
|
|
|
* reassemble 'V i a g r a' to 'Viagra'
|
|
|
|
*/
|
|
|
|
|
|
|
|
void reassembleToken(char *p){
|
|
|
|
int i, k=0;
|
|
|
|
|
|
|
|
for(i=0; i<strlen(p); i++){
|
|
|
|
|
|
|
|
if(isspace(*(p+i-1)) && !isspace(*(p+i)) && isspace(*(p+i+1)) && !isspace(*(p+i+2)) && isspace(*(p+i+3)) && !isspace(*(p+i+4)) && isspace(*(p+i+5)) ){
|
|
|
|
p[k] = *(p+i); k++;
|
|
|
|
p[k] = *(p+i+2); k++;
|
|
|
|
p[k] = *(p+i+4); k++;
|
|
|
|
|
|
|
|
i += 5;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
p[k] = *(p+i);
|
|
|
|
k++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
p[k] = '\0';
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void degenerateToken(unsigned char *p){
|
|
|
|
int i=1, d=0, dp=0;
|
|
|
|
unsigned char *s;
|
|
|
|
|
|
|
|
/* quit if this the string does not end with a punctuation character */
|
|
|
|
|
|
|
|
if(!ispunct(*(p+strlen((char *)p)-1)))
|
|
|
|
return;
|
|
|
|
|
|
|
|
s = p;
|
|
|
|
|
|
|
|
for(; *p; p++){
|
|
|
|
if(ispunct(*p)){
|
|
|
|
d = i;
|
|
|
|
|
|
|
|
if(!ispunct(*(p-1)))
|
|
|
|
dp = d;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
d = dp = i;
|
|
|
|
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
|
|
|
|
*(s+dp) = '\0';
|
|
|
|
|
|
|
|
if(*(s+dp-1) == '.' || *(s+dp-1) == '!' || *(s+dp-1) == '?') *(s+dp-1) = '\0';
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void fixURL(char *url){
|
2011-11-16 14:47:47 +01:00
|
|
|
char *p, *q, fixed_url[SMALLBUFSIZE];
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
memset(fixed_url, 0, sizeof(fixed_url));
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
p = url;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
if(strncasecmp(url, "http://", 7) == 0) p += 7;
|
|
|
|
if(strncasecmp(url, "https://", 8) == 0) p += 8;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
q = strchr(p, '/');
|
|
|
|
if(q) *q = '\0';
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-12-07 15:24:52 +01:00
|
|
|
snprintf(fixed_url, sizeof(fixed_url)-1, "__URL__%s ", p);
|
2012-01-28 20:52:13 +01:00
|
|
|
fix_email_address_for_sphinx(fixed_url+7);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
strcpy(url, fixed_url);
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
int extractNameFromHeaderLine(char *s, char *name, char *resultbuf){
|
|
|
|
int rc=0;
|
|
|
|
char buf[TINYBUFSIZE], *p, *q;
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
snprintf(buf, sizeof(buf)-1, "%s", s);
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
p = strstr(buf, name);
|
2011-11-14 15:57:52 +01:00
|
|
|
if(p){
|
2011-11-16 14:47:47 +01:00
|
|
|
p += strlen(name);
|
|
|
|
p = strchr(p, '=');
|
|
|
|
if(p){
|
|
|
|
p++;
|
|
|
|
q = strrchr(p, ';');
|
|
|
|
if(q) *q = '\0';
|
|
|
|
q = strrchr(p, '"');
|
|
|
|
if(q){
|
|
|
|
*q = '\0';
|
|
|
|
p = strchr(p, '"');
|
|
|
|
if(p){
|
|
|
|
p++;
|
|
|
|
}
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
2011-11-16 14:47:47 +01:00
|
|
|
snprintf(resultbuf, TINYBUFSIZE-1, "%s", p);
|
|
|
|
rc = 1;
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
2011-11-16 14:47:47 +01:00
|
|
|
}
|
2011-11-14 15:57:52 +01:00
|
|
|
|
2011-11-16 14:47:47 +01:00
|
|
|
return rc;
|
2011-11-14 15:57:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-12-30 15:52:59 +01:00
|
|
|
char *determine_attachment_type(char *filename, char *type){
|
|
|
|
char *p;
|
|
|
|
|
|
|
|
if(strncasecmp(type, "text/", strlen("text/")) == 0) return "text,";
|
|
|
|
if(strncasecmp(type, "image/", strlen("image/")) == 0) return "image,";
|
|
|
|
if(strncasecmp(type, "audio/", strlen("audio/")) == 0) return "audio,";
|
|
|
|
if(strncasecmp(type, "video/", strlen("video/")) == 0) return "video,";
|
|
|
|
if(strncasecmp(type, "text/x-card", strlen("text/x-card")) == 0) return "vcard,";
|
|
|
|
|
|
|
|
if(strncasecmp(type, "application/pdf", strlen("application/pdf")) == 0) return "pdf,";
|
|
|
|
|
|
|
|
if(strncasecmp(type, "application/msword", strlen("application/msword")) == 0) return "word,";
|
|
|
|
if(strncasecmp(type, "application/vnd.ms-excel", strlen("application/vnd.ms-excel")) == 0) return "excel,";
|
|
|
|
if(strncasecmp(type, "application/vnd.ms-powerpoint", strlen("application/vnd.ms-powerpoint")) == 0) return "powerpoint,";
|
|
|
|
|
|
|
|
if(strncasecmp(type, "application/x-shockwave-flash", strlen("application/x-shockwave-flash")) == 0) return "flash,";
|
|
|
|
|
|
|
|
if(strcasestr(type, "opendocument")) return "odf,";
|
|
|
|
|
|
|
|
if(strcasecmp(type, "application/octet-stream") == 0){
|
|
|
|
|
|
|
|
p = strrchr(type, '.');
|
|
|
|
if(p){
|
|
|
|
p++;
|
|
|
|
|
|
|
|
if(strncasecmp(p, "zip", 3) == 0) return "compressed,";
|
|
|
|
if(strncasecmp(p, "rar", 3) == 0) return "compressed,";
|
|
|
|
|
|
|
|
if(strncasecmp(p, "doc", 3) == 0) return "word,";
|
|
|
|
if(strncasecmp(p, "docx", 4) == 0) return "word,";
|
|
|
|
if(strncasecmp(p, "xls", 3) == 0) return "excel,";
|
|
|
|
if(strncasecmp(p, "xlsx", 4) == 0) return "excel,";
|
|
|
|
if(strncasecmp(p, "ppt", 3) == 0) return "powerpoint,";
|
|
|
|
if(strncasecmp(p, "pptx", 4) == 0) return "powerpoint,";
|
|
|
|
|
|
|
|
if(strncasecmp(p, "png", 3) == 0) return "image,";
|
|
|
|
if(strncasecmp(p, "gif", 3) == 0) return "image,";
|
|
|
|
if(strncasecmp(p, "jpg", 3) == 0) return "image,";
|
|
|
|
if(strncasecmp(p, "jpeg", 4) == 0) return "image,";
|
|
|
|
if(strncasecmp(p, "tiff", 4) == 0) return "image,";
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return "other,";
|
|
|
|
}
|
2011-11-14 15:57:52 +01:00
|
|
|
|