]> git.ipfire.org Git - thirdparty/sarg.git/blob - redirector.c
Merge commit 'debeace19822478022d24437a'
[thirdparty/sarg.git] / redirector.c
1 /*
2 * SARG Squid Analysis Report Generator http://sarg.sourceforge.net
3 * 1998, 2012
4 *
5 * SARG donations:
6 * please look at http://sarg.sourceforge.net/donations.php
7 * Support:
8 * http://sourceforge.net/projects/sarg/forums/forum/363374
9 * ---------------------------------------------------------------------
10 *
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.
15 *
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.
20 *
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.
24 *
25 */
26
27 #include "include/conf.h"
28 #include "include/defs.h"
29
30 static char **files_done = NULL;
31 static int nfiles_done = 0;
32
33 //! The number of invalid lines found in the redirector report.
34 static int RedirectorErrors=0;
35
36 static void parse_log(FILE *fp_ou,char *buf)
37 {
38 char leks[5], sep[2], res[MAXLEN];
39 char hour[15];
40 char source[128], list[128];
41 char full_url[MAX_URL_LEN];
42 const char *url;
43 char user[MAX_USER_LEN];
44 char ip[45];
45 long long int lmon, lday, lyear;
46 int mon, day, year;
47 int idata=0;
48 bool id_is_ip;
49 struct getwordstruct gwarea;
50 struct getwordstruct gwarea1;
51 struct userinfostruct *uinfo;
52
53 getword_start(&gwarea,buf);
54 if(RedirectorLogFormat[0] != '\0') {
55 getword_start(&gwarea1,RedirectorLogFormat);
56 leks[0]='\0';
57 if (getword(leks,sizeof(leks),&gwarea1,'#')<0) {
58 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (too many characters before first tag)\n"));
59 exit(EXIT_FAILURE);
60 }
61 year=0;
62 mon=0;
63 day=0;
64 hour[0]='\0';
65 source[0]='\0';
66 list[0]='\0';
67 ip[0]='\0';
68 user[0]='\0';
69 full_url[0]='\0';
70 while(strcmp(leks,"end") != 0) {
71 if (getword(leks,sizeof(leks),&gwarea1,'#')<0) {
72 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (missing # at end of tag)\n"));
73 exit(EXIT_FAILURE);
74 }
75 if (getword(sep,sizeof(sep),&gwarea1,'#')<0) {
76 debuga(_("Invalid \"redirector_log_format\" option in your sarg.conf (too many characters in column separator)\n"));
77 exit(EXIT_FAILURE);
78 }
79 if(strcmp(leks,"end") != 0) {
80 if (getword_limit(res,sizeof(res),&gwarea,sep[0])<0) {
81 debuga(_("Parsing of tag \"%s\" in redirector log %s returned no result\n"),leks,wentp);
82 RedirectorErrors++;
83 return;
84 }
85 if(strcmp(leks,"year") == 0) {
86 year=atoi(res);
87 } else if(strcmp(leks,"mon") == 0) {
88 mon=atoi(res);
89 } else if(strcmp(leks,"day") == 0) {
90 day=atoi(res);
91 } else if(strcmp(leks,"hour") == 0) {
92 if (strlen(res)>=sizeof(hour)) {
93 debuga(_("Hour string too long in redirector log file %s\n"),wentp);
94 RedirectorErrors++;
95 return;
96 }
97 strcpy(hour,res);
98 } else if(strcmp(leks,"source") == 0) {
99 if (strlen(res)>=sizeof(source)) {
100 debuga(_("Banning source name too long in redirector log file %s\n"),wentp);
101 RedirectorErrors++;
102 return;
103 }
104 strcpy(source,res);
105 } else if(strcmp(leks,"list") == 0) {
106 if (strlen(res)>=sizeof(list)) {
107 debuga(_("Banning list name too long in redirector log file %s\n"),wentp);
108 RedirectorErrors++;
109 return;
110 }
111 strcpy(list,res);
112 } else if(strcmp(leks,"ip") == 0) {
113 if (strlen(res)>=sizeof(ip)) {
114 debuga(_("IP address too long in redirector log file %s\n"),wentp);
115 RedirectorErrors++;
116 return;
117 }
118 strcpy(ip,res);
119 } else if(strcmp(leks,"user") == 0) {
120 if (strlen(res)>=sizeof(user)) {
121 debuga(_("User ID too long in redirector log file %s\n"),wentp);
122 RedirectorErrors++;
123 return;
124 }
125 strcpy(user,res);
126 } else if(strcmp(leks,"url") == 0) {
127 /*
128 * Don't worry about the url being truncated as we only keep the host name
129 * any way...
130 */
131 safe_strcpy(full_url,res,sizeof(full_url));
132 }
133 }
134 }
135 } else {
136 if (getword_atoll(&lyear,&gwarea,'-')<0 || getword_atoll(&lmon,&gwarea,'-')<0 ||
137 getword_atoll(&lday,&gwarea,' ')<0) {
138 debuga(_("Invalid date found in file %s\n"),wentp);
139 RedirectorErrors++;
140 return;
141 }
142 year=(int)lyear;
143 mon=(int)lmon;
144 day=(int)lday;
145 if (getword(hour,sizeof(hour),&gwarea,' ')<0) {
146 debuga(_("Invalid time found in file %s\n"),wentp);
147 RedirectorErrors++;
148 return;
149 }
150 if (getword_skip(MAXLEN,&gwarea,'(')<0 || getword(source,sizeof(source),&gwarea,'/')<0) {
151 debuga(_("Invalid redirected source in file %s\n"),wentp);
152 RedirectorErrors++;
153 return;
154 }
155 if (getword(list,sizeof(list),&gwarea,'/')<0) {
156 debuga(_("Invalid redirected list in file %s\n"),wentp);
157 RedirectorErrors++;
158 return;
159 }
160 if (getword_skip(MAXLEN,&gwarea,' ')<0 || getword_limit(full_url,sizeof(full_url),&gwarea,' ')<0) {
161 debuga(_("Invalid URL in file %s\n"),wentp);
162 RedirectorErrors++;
163 return;
164 }
165 if (getword(ip,sizeof(ip),&gwarea,'/')<0) {
166 debuga(_("Invalid source IP in file %s\n"),wentp);
167 RedirectorErrors++;
168 return;
169 }
170 if (getword_skip(MAXLEN,&gwarea,' ')<0 || getword(user,sizeof(user),&gwarea,' ')<0) {
171 debuga(_("Invalid user in file %s\n"),wentp);
172 RedirectorErrors++;
173 return;
174 }
175 }
176 url=process_url(full_url,false);
177
178 //sprintf(warea,"%04d%02d%02d",year,mon,day);
179
180 if(RedirectorFilterOutDate) {
181 idata = year*10000+mon*100+day;
182 if(idata < dfrom || idata > duntil)
183 return;
184 }
185
186 if(UserIp) {
187 strcpy(user,ip);
188 id_is_ip=true;
189 } else {
190 id_is_ip=false;
191 if(strcmp(user,"-") == 0 || strcmp(user," ") == 0 || strcmp(user,"") == 0) {
192 if(RecordsWithoutUser == RECORDWITHOUTUSER_IP) {
193 strcpy(user,ip);
194 id_is_ip=true;
195 }
196 if(RecordsWithoutUser == RECORDWITHOUTUSER_IGNORE)
197 return;
198 if(RecordsWithoutUser == RECORDWITHOUTUSER_EVERYBODY)
199 strcpy(user,"everybody");
200 }
201 }
202 uinfo=userinfo_find_from_id(user);
203 if (!uinfo) {
204 uinfo=userinfo_create(user);
205 uinfo->id_is_ip=id_is_ip;
206 uinfo->no_report=true;
207 if(Ip2Name && id_is_ip) ip2name(user,sizeof(user));
208 user_find(uinfo->label,MAX_USER_LEN, user);
209 }
210 fprintf(fp_ou,"%s\t%04d%02d%02d\t%s\t%s\t%s\t",uinfo->id,year,mon,day,hour,ip,url);
211 if (source[0] && list[0])
212 fprintf(fp_ou,"%s/%s\n",source,list);
213 else if (source[0])
214 fprintf(fp_ou,"%s\n",source);
215 else
216 fprintf(fp_ou,"%s\n",list);
217 redirector_count++;
218 }
219
220 static void read_log(const char *wentp, FILE *fp_ou,int dfrom,int duntil)
221 {
222 FILE *fp_in = NULL;
223 char *buf;
224 int i;
225 longline line;
226
227 if(debug) {
228 debuga(_("Reading redirector log file %s\n"),wentp);
229 }
230
231 /* With squidGuard, you can log groups in only one log file.
232 We must parse each log files only one time. Example :
233 dest porn {
234 domainlist porn/domains
235 urllist porn/urls
236 log file1.log
237 }
238 dest aggressive {
239 domainlist aggressive/domains
240 urllist aggressive/urls
241 log file2.log
242 }
243 dest audio-video {
244 domainlist audio-video/domains
245 urllist audio-video/urls
246 log file1.log
247 }
248 */
249 for (i=0; i<nfiles_done; i++)
250 if (!strcmp(wentp, files_done[i])) return;
251
252 nfiles_done++;
253 files_done = realloc(files_done, nfiles_done*sizeof(char *));
254 if (!files_done) {
255 debuga(_("Not enough memory to store the name of the new redirector log to be read - %s\n"),strerror(errno));
256 exit(EXIT_FAILURE);
257 }
258 files_done[nfiles_done-1] = strdup(wentp);
259 if (!files_done[nfiles_done-1]) {
260 debuga(_("Not enough memory to store the name of the new redirector log to be read - %s\n"),strerror(errno));
261 exit(EXIT_FAILURE);
262 }
263
264 if ((fp_in=fopen(wentp,"r"))==NULL) {
265 debuga(_("(squidguard) Cannot open log file %s\n"),wentp);
266 exit(EXIT_FAILURE);
267 }
268
269 if ((line=longline_create())==NULL) {
270 debuga(_("Not enough memory to read the redirector log\n"));
271 exit(EXIT_FAILURE);
272 }
273
274 while ((buf=longline_read(fp_in,line)) != NULL) {
275 parse_log(fp_ou,buf);
276 }
277 fclose(fp_in);
278 longline_destroy(&line);
279 return;
280 }
281
282
283 void redirector_log(void)
284 {
285 FILE *fp_ou = NULL, *fp_guard = NULL;
286 char buf[MAXLEN];
287 char guard_in[MAXLEN];
288 char guard_ou[MAXLEN];
289 char logdir[MAXLEN];
290 char user[MAXLEN];
291 char tmp6[MAXLEN];
292 int i;
293 int y;
294 int cstatus;
295 int dfrom, duntil;
296 char *str;
297 char *str2;
298
299 str2 = user;
300
301 if(SquidGuardConf[0] == '\0' && NRedirectorLogs == 0) {
302 if (debugz) debugaz(_("No redirector logs provided to produce that kind of report\n"));
303 return;
304 }
305
306 snprintf(guard_in,sizeof(guard_in),"%s/redirector.int_unsort",tmp);
307 snprintf(guard_ou,sizeof(guard_ou),"%s/redirector.int_log",tmp);
308 if((fp_ou=fopen(guard_in,"a"))==NULL) {
309 debuga(_("(squidguard) Cannot open log file %s\n"),guard_in);
310 exit(EXIT_FAILURE);
311 }
312
313 dfrom=(period.start.tm_year+1900)*10000+(period.start.tm_mon+1)*100+period.start.tm_mday;
314 duntil=(period.end.tm_year+1900)*10000+(period.end.tm_mon+1)*100+period.end.tm_mday;
315
316 if (NRedirectorLogs>0) {
317 for (i=0 ; i<NRedirectorLogs ; i++)
318 read_log(RedirectorLogs[i],fp_ou,dfrom,duntil);
319 } else {
320 if(access(SquidGuardConf, R_OK) != 0) {
321 debuga(_("Cannot open squidGuard config file: %s\n"),SquidGuardConf);
322 exit(EXIT_FAILURE);
323 }
324
325 if((fp_guard=fopen(SquidGuardConf,"r"))==NULL) {
326 debuga(_("(squidguard) Cannot open log file %s\n"),SquidGuardConf);
327 exit(EXIT_FAILURE);
328 }
329
330 logdir[0]=0;
331 while(fgets(buf,sizeof(buf),fp_guard)!=NULL) {
332 fixendofline(buf);
333 if((str=get_param_value("logdir",buf))!=NULL) {
334 /*
335 We want to tolerate spaces inside the directory name but we must also
336 remove the trailing spaces left by the editor after the directory name.
337 This should not be a problem as nobody use a file name with trailing spaces.
338 */
339 for (y=strlen(str)-1 ; y>=0 && (unsigned char)str[y]<=' ' ; y--);
340 if (y>=sizeof(logdir)-1) y=sizeof(logdir)-2;
341 logdir[y+1] = '\0';
342 while (y>=0) {
343 logdir[y] = str[y];
344 y--;
345 }
346 } else if((str=get_param_value("log",buf))!=NULL) {
347 if((str2=get_param_value("anonymous",str))!=NULL)
348 str=str2;
349
350 /*
351 If logdir is defined, we prepend it to the log file name, otherwise, we assume
352 the log directive provides an absolute file name to the log file. Therefore,
353 we don't need to add an additionnal / at the beginning of the log file name.
354 */
355 y=(logdir[0]) ? sprintf(wentp,"%s/",logdir) : 0;
356 /*
357 Spaces are allowed in the name of the log file. The file name ends at the first #
358 because it is assumed it is an end of line comment. Any space before the # is then
359 removed. Any control character (i.e. a character with a code lower than 32) ends
360 the file name. That includes the terminating zero.
361 */
362 while((unsigned char)*str>=' ' && *str!='#' && y<sizeof(wentp)-1)
363 wentp[y++]=*str++;
364 if(*str=='#') {
365 str--;
366 while(*str==' ' && y>0) {
367 str--;
368 y--;
369 }
370 }
371 wentp[y]=0;
372 read_log(wentp,fp_ou,dfrom,duntil);
373 }
374 }
375 }
376
377 if (fp_guard) fclose(fp_guard);
378 if (fp_ou) fclose(fp_ou);
379
380 if (files_done) {
381 for (y=0; y<nfiles_done; y++)
382 if (files_done[y]) free(files_done[y]);
383 free(files_done);
384 }
385
386 if(debug) {
387 debuga(_("Sorting file: %s\n"),guard_ou);
388 }
389
390 if (snprintf(tmp6,sizeof(tmp6),"sort -t \"\t\" -k 1,1 -k 2,2 -k 4,4 \"%s\" -o \"%s\"",guard_in, guard_ou)>=sizeof(tmp6)) {
391 debuga(_("Sort command too long when sorting file \"%s\" to \"%s\"\n"),guard_in,guard_ou);
392 exit(EXIT_FAILURE);
393 }
394 cstatus=system(tmp6);
395 if (!WIFEXITED(cstatus) || WEXITSTATUS(cstatus)) {
396 debuga(_("sort command return status %d\n"),WEXITSTATUS(cstatus));
397 debuga(_("sort command: %s\n"),tmp6);
398 exit(EXIT_FAILURE);
399 }
400
401 if (unlink(guard_in)) {
402 debuga(_("Cannot delete %s - %s\n"),guard_in,strerror(errno));
403 exit(EXIT_FAILURE);
404 }
405 return;
406 }
407
408 static void show_ignored_redirector(FILE *fp_ou,int count)
409 {
410 char ignored[80];
411
412 snprintf(ignored,sizeof(ignored),ngettext("%d more redirector entry not shown here&hellip;","%d more redirector entries not shown here&hellip;",count),count);
413 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);
414 }
415
416 void redirector_report(void)
417 {
418 FILE *fp_in = NULL, *fp_ou = NULL;
419
420 char *buf;
421 char *url;
422 char squidguard_in[MAXLEN];
423 char report[MAXLEN];
424 char ip[45];
425 char rule[255];
426 char oip[45];
427 char user[MAXLEN];
428 char ouser[MAXLEN];
429 char data[15];
430 char hora[15];
431 char ouser2[255];
432 char oname[MAXLEN];
433 bool z=false;
434 int count=0;
435 long long int data2;
436 bool new_user;
437 struct getwordstruct gwarea;
438 const struct userinfostruct *uinfo;
439 struct tm t;
440 longline line;
441
442 ouser[0]='\0';
443 ouser2[0]='\0';
444
445 snprintf(squidguard_in,sizeof(squidguard_in),"%s/redirector.int_log",tmp);
446 if(!redirector_count) {
447 unlink(squidguard_in);
448 if (debugz) debugaz(_("Redirector report not generated because it is empty\n"));
449 return;
450 }
451
452 snprintf(report,sizeof(report),"%s/redirector.html",outdirname);
453
454 if((fp_in=fopen(squidguard_in,"r"))==NULL) {
455 debuga(_("(squidguard) Cannot open log file %s\n"),squidguard_in);
456 exit(EXIT_FAILURE);
457 }
458
459 if((fp_ou=fopen(report,"w"))==NULL) {
460 debuga(_("(squidguard) Cannot open log file %s\n"),report);
461 exit(EXIT_FAILURE);
462 }
463
464 if ((line=longline_create())==NULL) {
465 debuga(_("Not enough memory to read the processed redirector log\n"));
466 exit(EXIT_FAILURE);
467 }
468
469 write_html_header(fp_ou,(IndexTree == INDEX_TREE_DATE) ? 3 : 1,_("Redirector report"),HTML_JS_NONE);
470 fputs("<tr><td class=\"header_c\">",fp_ou);
471 fprintf(fp_ou,_("Period: %s"),period.html);
472 fputs("</td></tr>\n",fp_ou);
473 fprintf(fp_ou,"<tr><th class=\"header_c\">%s</th></tr>\n",_("Redirector report"));
474 close_html_header(fp_ou);
475
476 fputs("<div class=\"report\"><table cellpadding=1 cellspacing=2>\n",fp_ou);
477 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"));
478
479 while((buf=longline_read(fp_in,line))!=NULL) {
480 getword_start(&gwarea,buf);
481 if (getword(user,sizeof(user),&gwarea,'\t')<0) {
482 debuga(_("Invalid user in file %s\n"),squidguard_in);
483 exit(EXIT_FAILURE);
484 }
485 if (getword_atoll(&data2,&gwarea,'\t')<0) {
486 debuga(_("Invalid date in file %s\n"),squidguard_in);
487 exit(EXIT_FAILURE);
488 }
489 if (getword(hora,sizeof(hora),&gwarea,'\t')<0) {
490 debuga(_("Invalid time in file %s\n"),squidguard_in);
491 exit(EXIT_FAILURE);
492 }
493 if (getword(ip,sizeof(ip),&gwarea,'\t')<0) {
494 debuga(_("Invalid IP address in file %s\n"),squidguard_in);
495 exit(EXIT_FAILURE);
496 }
497 if (getword_ptr(buf,&url,&gwarea,'\t')<0) {
498 debuga(_("Invalid URL in file %s\n"),squidguard_in);
499 exit(EXIT_FAILURE);
500 }
501 if (getword(rule,sizeof(rule),&gwarea,'\n')<0) {
502 debuga(_("Invalid rule in file %s\n"),squidguard_in);
503 exit(EXIT_FAILURE);
504 }
505
506 uinfo=userinfo_find_from_id(user);
507 if (!uinfo) {
508 debuga(_("Unknown user ID %s in file %s\n"),user,squidguard_in);
509 exit(EXIT_FAILURE);
510 }
511
512 computedate(data2/10000,(data2/100)%10,data2%100,&t);
513 strftime(data,sizeof(data),"%x",&t);
514
515 new_user=false;
516 if(!z) {
517 strcpy(ouser,user);
518 strcpy(oip,ip);
519 strcpy(oname,ip);
520 if (Ip2Name && !uinfo->id_is_ip) ip2name(oname,sizeof(oname));
521 z=true;
522 new_user=true;
523 } else {
524 if(strcmp(ouser,user) != 0) {
525 strcpy(ouser,user);
526 new_user=true;
527 }
528 if(strcmp(oip,ip) != 0) {
529 strcpy(oip,ip);
530 strcpy(oname,ip);
531 if (Ip2Name && !uinfo->id_is_ip) ip2name(oname,sizeof(oname));
532 new_user=true;
533 }
534 }
535
536 if(SquidGuardReportLimit) {
537 if(strcmp(ouser2,uinfo->label) == 0) {
538 count++;
539 } else {
540 if(count>SquidGuardReportLimit && SquidGuardReportLimit>0)
541 show_ignored_redirector(fp_ou,count-SquidGuardReportLimit);
542 count=1;
543 strcpy(ouser2,uinfo->label);
544 }
545 if(count > SquidGuardReportLimit)
546 continue;
547 }
548
549 if (new_user)
550 fprintf(fp_ou,"<tr><td class=\"data2\">%s</td><td class=\"data2\">%s</td>",uinfo->label,ip);
551 else
552 fputs("<tr><td class=\"data2\"></td><td class=\"data2\"></td>",fp_ou);
553 fprintf(fp_ou,"<td class=\"data2\">%s-%s</td><td class=\"data2\">",data,hora);
554 output_html_link(fp_ou,url,100);
555 fprintf(fp_ou,"</td><td class=\"data2\">%s</td></tr>\n",rule);
556 }
557 fclose(fp_in);
558 longline_destroy(&line);
559
560 if(count>SquidGuardReportLimit && SquidGuardReportLimit>0)
561 show_ignored_redirector(fp_ou,count-SquidGuardReportLimit);
562
563 fputs("</table>\n",fp_ou);
564
565 if (RedirectorErrors>0)
566 {
567 fputs("<div class=\"warn\"><span>",fp_ou);
568 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);
569 fputs("</span></div>\n",fp_ou);
570 }
571
572 fputs("</div>\n",fp_ou);
573 if (write_html_trailer(fp_ou)<0)
574 debuga(_("Write error in file %s\n"),report);
575 if (fclose(fp_ou)==EOF)
576 debuga(_("Failed to close file %s - %s\n"),report,strerror(errno));
577
578 if (unlink(squidguard_in)) {
579 debuga(_("Cannot delete %s - %s\n"),squidguard_in,strerror(errno));
580 exit(EXIT_FAILURE);
581 }
582
583 return;
584 }