2 * SARG Squid Analysis Report Generator http://sarg.sourceforge.net
6 * please look at http://sarg.sourceforge.net/donations.php
8 * http://sourceforge.net/projects/sarg/forums/forum/363374
9 * ---------------------------------------------------------------------
11 * This program is free software; you can redistribute it and/or modify
12 * it under the terms of the GNU General Public License as published by
13 * the Free Software Foundation; either version 2 of the License, or
14 * (at your option) any later version.
16 * This program is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 * GNU General Public License for more details.
21 * You should have received a copy of the GNU General Public License
22 * along with this program; if not, write to the Free Software
23 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.
27 #include "include/conf.h"
28 #include "include/defs.h"
30 static char **files_done
= NULL
;
31 static int nfiles_done
= 0;
33 //! The number of invalid lines found in the redirector report.
34 static int RedirectorErrors
=0;
35 //! The file containing the sorted entries.
36 static char redirector_sorted
[MAXLEN
]="";
38 static void parse_log(FILE *fp_ou
,char *buf
)
40 char leks
[5], sep
[2], res
[MAXLEN
];
42 char source
[128], list
[128];
43 char full_url
[MAX_URL_LEN
];
45 char user
[MAX_USER_LEN
];
47 char userlabel
[MAX_USER_LEN
];
48 long long int lmon
, lday
, lyear
;
52 struct getwordstruct gwarea
;
53 struct getwordstruct gwarea1
;
54 struct userinfostruct
*uinfo
;
56 getword_start(&gwarea
,buf
);
57 if(RedirectorLogFormat
[0] != '\0') {
58 getword_start(&gwarea1
,RedirectorLogFormat
);
60 if (getword(leks
,sizeof(leks
),&gwarea1
,'#')<0) {
61 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (too many characters before first tag)\n"));
73 while(strcmp(leks
,"end") != 0) {
74 if (getword(leks
,sizeof(leks
),&gwarea1
,'#')<0) {
75 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (missing # at end of tag)\n"));
78 if (getword(sep
,sizeof(sep
),&gwarea1
,'#')<0) {
79 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (too many characters in column separator)\n"));
82 if(strcmp(leks
,"end") != 0) {
83 if (getword_limit(res
,sizeof(res
),&gwarea
,sep
[0])<0) {
84 debuga(_("Parsing of tag \"%s\" in redirector log %s returned no result\n"),leks
,wentp
);
88 if(strcmp(leks
,"year") == 0) {
90 } else if(strcmp(leks
,"mon") == 0) {
92 } else if(strcmp(leks
,"day") == 0) {
94 } else if(strcmp(leks
,"hour") == 0) {
95 if (strlen(res
)>=sizeof(hour
)) {
96 debuga(_("Hour string too long in redirector log file %s\n"),wentp
);
101 } else if(strcmp(leks
,"source") == 0) {
102 if (strlen(res
)>=sizeof(source
)) {
103 debuga(_("Banning source name too long in redirector log file %s\n"),wentp
);
108 } else if(strcmp(leks
,"list") == 0) {
109 if (strlen(res
)>=sizeof(list
)) {
110 debuga(_("Banning list name too long in redirector log file %s\n"),wentp
);
115 } else if(strcmp(leks
,"ip") == 0) {
116 if (strlen(res
)>=sizeof(ip
)) {
117 debuga(_("IP address too long in redirector log file %s\n"),wentp
);
122 } else if(strcmp(leks
,"user") == 0) {
123 if (strlen(res
)>=sizeof(user
)) {
124 debuga(_("User ID too long in redirector log file %s\n"),wentp
);
129 } else if(strcmp(leks
,"url") == 0) {
131 * Don't worry about the url being truncated as we only keep the host name
134 safe_strcpy(full_url
,res
,sizeof(full_url
));
139 if (getword_atoll(&lyear
,&gwarea
,'-')<0 || getword_atoll(&lmon
,&gwarea
,'-')<0 ||
140 getword_atoll(&lday
,&gwarea
,' ')<0) {
141 debuga(_("Invalid date found in file %s\n"),wentp
);
148 if (getword(hour
,sizeof(hour
),&gwarea
,' ')<0) {
149 debuga(_("Invalid time found in file %s\n"),wentp
);
153 if (getword_skip(MAXLEN
,&gwarea
,'(')<0 || getword(source
,sizeof(source
),&gwarea
,'/')<0) {
154 debuga(_("Invalid redirected source in file %s\n"),wentp
);
158 if (getword(list
,sizeof(list
),&gwarea
,'/')<0) {
159 debuga(_("Invalid redirected list in file %s\n"),wentp
);
163 if (getword_skip(MAXLEN
,&gwarea
,' ')<0 || getword_limit(full_url
,sizeof(full_url
),&gwarea
,' ')<0) {
164 debuga(_("Invalid URL in file %s\n"),wentp
);
168 if (getword(ip
,sizeof(ip
),&gwarea
,'/')<0) {
169 debuga(_("Invalid source IP in file %s\n"),wentp
);
173 if (getword_skip(MAXLEN
,&gwarea
,' ')<0 || getword(user
,sizeof(user
),&gwarea
,' ')<0) {
174 debuga(_("Invalid user in file %s\n"),wentp
);
179 url
=process_url(full_url
,false);
181 //sprintf(warea,"%04d%02d%02d",year,mon,day);
183 if(RedirectorFilterOutDate
) {
184 idata
= year
*10000+mon
*100+day
;
185 if(idata
< dfrom
|| idata
> duntil
)
194 if (user
[0]=='\0' || (user
[1]=='\0' && (user
[0]=='-' || user
[0]==' '))) {
195 if(RecordsWithoutUser
== RECORDWITHOUTUSER_IP
) {
199 if(RecordsWithoutUser
== RECORDWITHOUTUSER_IGNORE
)
201 if(RecordsWithoutUser
== RECORDWITHOUTUSER_EVERYBODY
)
202 strcpy(user
,"everybody");
205 uinfo
=userinfo_find_from_id(user
);
207 uinfo
=userinfo_create(user
);
208 uinfo
->id_is_ip
=id_is_ip
;
209 uinfo
->no_report
=true;
210 if(Ip2Name
&& id_is_ip
) ip2name(user
,sizeof(user
));
211 user_find(userlabel
,MAX_USER_LEN
, user
);
212 userinfo_label(uinfo
,userlabel
);
214 fprintf(fp_ou
,"%s\t%04d%02d%02d\t%s\t%s\t%s\t",uinfo
->id
,year
,mon
,day
,hour
,ip
,url
);
215 if (source
[0] && list
[0])
216 fprintf(fp_ou
,"%s/%s\n",source
,list
);
218 fprintf(fp_ou
,"%s\n",source
);
220 fprintf(fp_ou
,"%s\n",list
);
224 static void read_log(const char *wentp
, FILE *fp_ou
,int dfrom
,int duntil
)
232 debuga(_("Reading redirector log file %s\n"),wentp
);
235 /* With squidGuard, you can log groups in only one log file.
236 We must parse each log files only one time. Example :
238 domainlist porn/domains
243 domainlist aggressive/domains
244 urllist aggressive/urls
248 domainlist audio-video/domains
249 urllist audio-video/urls
253 for (i
=0; i
<nfiles_done
; i
++)
254 if (!strcmp(wentp
, files_done
[i
])) return;
257 files_done
= realloc(files_done
, nfiles_done
*sizeof(char *));
259 debuga(_("Not enough memory to store the name of the new redirector log to be read - %s\n"),strerror(errno
));
262 files_done
[nfiles_done
-1] = strdup(wentp
);
263 if (!files_done
[nfiles_done
-1]) {
264 debuga(_("Not enough memory to store the name of the new redirector log to be read - %s\n"),strerror(errno
));
268 if ((fp_in
=fopen(wentp
,"r"))==NULL
) {
269 debuga(_("(squidguard) Cannot open log file %s: %s\n"),wentp
,strerror(errno
));
273 if ((line
=longline_create())==NULL
) {
274 debuga(_("Not enough memory to read the redirector log\n"));
278 while ((buf
=longline_read(fp_in
,line
)) != NULL
) {
279 parse_log(fp_ou
,buf
);
282 longline_destroy(&line
);
287 void redirector_log(void)
289 FILE *fp_ou
= NULL
, *fp_guard
= NULL
;
291 char guard_in
[MAXLEN
];
304 if(SquidGuardConf
[0] == '\0' && NRedirectorLogs
== 0) {
305 if (debugz
) debugaz(_("No redirector logs provided to produce that kind of report\n"));
309 snprintf(guard_in
,sizeof(guard_in
),"%s/redirector.int_unsort",tmp
);
310 if((fp_ou
=fopen(guard_in
,"w"))==NULL
) {
311 debuga(_("(squidguard) Cannot open log file %s: %s\n"),guard_in
,strerror(errno
));
315 dfrom
=(period
.start
.tm_year
+1900)*10000+(period
.start
.tm_mon
+1)*100+period
.start
.tm_mday
;
316 duntil
=(period
.end
.tm_year
+1900)*10000+(period
.end
.tm_mon
+1)*100+period
.end
.tm_mday
;
318 if (NRedirectorLogs
>0) {
319 for (i
=0 ; i
<NRedirectorLogs
; i
++)
320 read_log(RedirectorLogs
[i
],fp_ou
,dfrom
,duntil
);
322 if(access(SquidGuardConf
, R_OK
) != 0) {
323 debuga(_("Cannot open squidGuard config file: %s\n"),SquidGuardConf
);
327 if((fp_guard
=fopen(SquidGuardConf
,"r"))==NULL
) {
328 debuga(_("(squidguard) Cannot open log file %s: %s\n"),SquidGuardConf
,strerror(errno
));
333 while(fgets(buf
,sizeof(buf
),fp_guard
)!=NULL
) {
335 if((str
=get_param_value("logdir",buf
))!=NULL
) {
337 We want to tolerate spaces inside the directory name but we must also
338 remove the trailing spaces left by the editor after the directory name.
339 This should not be a problem as nobody use a file name with trailing spaces.
341 for (y
=strlen(str
)-1 ; y
>=0 && (unsigned char)str
[y
]<=' ' ; y
--);
342 if (y
>=sizeof(logdir
)-1) y
=sizeof(logdir
)-2;
348 } else if((str
=get_param_value("log",buf
))!=NULL
) {
349 if((str2
=get_param_value("anonymous",str
))!=NULL
)
353 If logdir is defined, we prepend it to the log file name, otherwise, we assume
354 the log directive provides an absolute file name to the log file. Therefore,
355 we don't need to add an additionnal / at the beginning of the log file name.
357 y
=(logdir
[0]) ? sprintf(wentp
,"%s/",logdir
) : 0;
359 Spaces are allowed in the name of the log file. The file name ends at the first #
360 because it is assumed it is an end of line comment. Any space before the # is then
361 removed. Any control character (i.e. a character with a code lower than 32) ends
362 the file name. That includes the terminating zero.
364 while((unsigned char)*str
>=' ' && *str
!='#' && y
<sizeof(wentp
)-1)
368 while(*str
==' ' && y
>0) {
374 read_log(wentp
,fp_ou
,dfrom
,duntil
);
379 if (fp_guard
) fclose(fp_guard
);
380 if (fp_ou
&& fclose(fp_ou
)==EOF
) {
381 debuga(_("Write error in %s: %s\n"),guard_in
,strerror(errno
));
386 for (y
=0; y
<nfiles_done
; y
++)
387 if (files_done
[y
]) free(files_done
[y
]);
391 if (redirector_count
) {
392 snprintf(redirector_sorted
,sizeof(redirector_sorted
),"%s/redirector.int_log",tmp
);
394 debuga(_("Sorting file: %s\n"),redirector_sorted
);
397 if (snprintf(tmp6
,sizeof(tmp6
),"sort -t \"\t\" -k 1,1 -k 2,2 -k 4,4 \"%s\" -o \"%s\"",guard_in
, redirector_sorted
)>=sizeof(tmp6
)) {
398 debuga(_("Sort command too long when sorting file \"%s\" to \"%s\"\n"),guard_in
,redirector_sorted
);
401 cstatus
=system(tmp6
);
402 if (!WIFEXITED(cstatus
) || WEXITSTATUS(cstatus
)) {
403 debuga(_("sort command return status %d\n"),WEXITSTATUS(cstatus
));
404 debuga(_("sort command: %s\n"),tmp6
);
409 if (!KeepTempLog
&& unlink(guard_in
)) {
410 debuga(_("Cannot delete \"%s\": %s\n"),guard_in
,strerror(errno
));
416 static void show_ignored_redirector(FILE *fp_ou
,int count
)
420 snprintf(ignored
,sizeof(ignored
),ngettext("%d more redirector entry not shown here…","%d more redirector entries not shown here…",count
),count
);
421 fprintf(fp_ou
,"<tr><td class=\"data\"></td><td class=\"data\"></td><td class=\"data\"></td><td class=\"data2 more\">%s</td><td class=\"data\"></td></tr>\n",ignored
);
424 void redirector_report(void)
426 FILE *fp_in
= NULL
, *fp_ou
= NULL
;
444 struct getwordstruct gwarea
;
445 const struct userinfostruct
*uinfo
;
452 if(!redirector_count
) {
454 if (redirector_sorted
[0])
455 debugaz(_("Redirector report not generated because it is empty\n"));
460 snprintf(report
,sizeof(report
),"%s/redirector.html",outdirname
);
462 if((fp_in
=fopen(redirector_sorted
,"r"))==NULL
) {
463 debuga(_("(squidguard) Cannot open log file %s: %s\n"),redirector_sorted
,strerror(errno
));
467 if((fp_ou
=fopen(report
,"w"))==NULL
) {
468 debuga(_("(squidguard) Cannot open log file %s: %s\n"),report
,strerror(errno
));
472 if ((line
=longline_create())==NULL
) {
473 debuga(_("Not enough memory to read the processed redirector log\n"));
477 write_html_header(fp_ou
,(IndexTree
== INDEX_TREE_DATE
) ? 3 : 1,_("Redirector report"),HTML_JS_NONE
);
478 fputs("<tr><td class=\"header_c\">",fp_ou
);
479 fprintf(fp_ou
,_("Period: %s"),period
.html
);
480 fputs("</td></tr>\n",fp_ou
);
481 fprintf(fp_ou
,"<tr><th class=\"header_c\">%s</th></tr>\n",_("Redirector report"));
482 close_html_header(fp_ou
);
484 fputs("<div class=\"report\"><table cellpadding=1 cellspacing=2>\n",fp_ou
);
485 fprintf(fp_ou
,"<tr><th class=\"header_l\">%s</th><th class=\"header_l\">%s</th><th class=\"header_l\">%s</th><th class=\"header_l\">%s</th><th class=\"header_l\">%s</th></tr>\n",_("USERID"),_("IP/NAME"),_("DATE/TIME"),_("ACCESSED SITE"),_("RULE"));
487 while((buf
=longline_read(fp_in
,line
))!=NULL
) {
488 getword_start(&gwarea
,buf
);
489 if (getword(user
,sizeof(user
),&gwarea
,'\t')<0) {
490 debuga(_("Invalid user in file %s\n"),redirector_sorted
);
493 if (getword_atoll(&data2
,&gwarea
,'\t')<0) {
494 debuga(_("Invalid date in file %s\n"),redirector_sorted
);
497 if (getword(hora
,sizeof(hora
),&gwarea
,'\t')<0) {
498 debuga(_("Invalid time in file %s\n"),redirector_sorted
);
501 if (getword(ip
,sizeof(ip
),&gwarea
,'\t')<0) {
502 debuga(_("Invalid IP address in file %s\n"),redirector_sorted
);
505 if (getword_ptr(buf
,&url
,&gwarea
,'\t')<0) {
506 debuga(_("Invalid URL in file %s\n"),redirector_sorted
);
509 if (getword(rule
,sizeof(rule
),&gwarea
,'\n')<0) {
510 debuga(_("Invalid rule in file %s\n"),redirector_sorted
);
514 uinfo
=userinfo_find_from_id(user
);
516 debuga(_("Unknown user ID %s in file %s\n"),user
,redirector_sorted
);
520 computedate(data2
/10000,(data2
/100)%10,data2
%100,&t
);
521 strftime(data
,sizeof(data
),"%x",&t
);
528 if (Ip2Name
&& !uinfo
->id_is_ip
) ip2name(oname
,sizeof(oname
));
532 if(strcmp(ouser
,user
) != 0) {
536 if(strcmp(oip
,ip
) != 0) {
539 if (Ip2Name
&& !uinfo
->id_is_ip
) ip2name(oname
,sizeof(oname
));
544 if(SquidGuardReportLimit
) {
545 if(strcmp(ouser2
,uinfo
->label
) == 0) {
548 if(count
>SquidGuardReportLimit
&& SquidGuardReportLimit
>0)
549 show_ignored_redirector(fp_ou
,count
-SquidGuardReportLimit
);
551 strcpy(ouser2
,uinfo
->label
);
553 if(count
> SquidGuardReportLimit
)
558 fprintf(fp_ou
,"<tr><td class=\"data2\">%s</td><td class=\"data2\">%s</td>",uinfo
->label
,ip
);
560 fputs("<tr><td class=\"data2\"></td><td class=\"data2\"></td>",fp_ou
);
561 fprintf(fp_ou
,"<td class=\"data2\">%s-%s</td><td class=\"data2\">",data
,hora
);
562 output_html_link(fp_ou
,url
,100);
563 fprintf(fp_ou
,"</td><td class=\"data2\">%s</td></tr>\n",rule
);
566 longline_destroy(&line
);
568 if(count
>SquidGuardReportLimit
&& SquidGuardReportLimit
>0)
569 show_ignored_redirector(fp_ou
,count
-SquidGuardReportLimit
);
571 fputs("</table>\n",fp_ou
);
573 if (RedirectorErrors
>0)
575 fputs("<div class=\"warn\"><span>",fp_ou
);
576 fprintf(fp_ou
,ngettext("%d error found in the log file. Some entries may be missing.","%d errors found in the log file. Some entries may be missing.",RedirectorErrors
),RedirectorErrors
);
577 fputs("</span></div>\n",fp_ou
);
580 fputs("</div>\n",fp_ou
);
581 if (write_html_trailer(fp_ou
)<0)
582 debuga(_("Write error in file %s\n"),report
);
583 if (fclose(fp_ou
)==EOF
) {
584 debuga(_("Write error in %s: %s\n"),report
,strerror(errno
));
588 if (!KeepTempLog
&& unlink(redirector_sorted
)) {
589 debuga(_("Cannot delete \"%s\": %s\n"),redirector_sorted
,strerror(errno
));