splitlog.c

   1 /*
   2  * SARG Squid Analysis Report Generator      http://sarg.sourceforge.net
   3  *                                                            1998, 2015
   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 /*
  31 Extract a date range from a squid log file and write it into a separate file.
  32
  33 It can optionally convert the date in human readable format.
  34
  35 The output can be split by day into separate files.
  36
  37 \param arq The squid log file to split.
  38 \param df The date format if the date is to be converted in human readable form. Only the first
  39 character is taken into account. It can be 'e' for European date format or anything else for
  40 US date format.
  41 \param dfrom The first date to output in the form (Year*10000+Month*100+Day).
  42 \param duntil The last date to output in the form (Year*10000+Month*100+Day).
  43 \param convert \c True if the date must be converted into human readable form.
  44 \param splitprefix If not empty, the output file is written in separate files (one for each day) and
  45 the files are named after the day they contain prefixed with the string contained in this variable.
  46 */
  47 void splitlog(const char *arq, char df, int dfrom, int duntil, int convert, const char *splitprefix)
  48 {
  49         FILE *fp_in;
  50         FILE *fp_ou=NULL;
  51         char *buf;
  52         char data[30];
  53         char dia[11];
  54         char output_file[MAXLEN];
  55         time_t tt;
  56         time_t min_tt;
  57         time_t max_tt=0;
  58         int idata=0;
  59         int autosplit=0;
  60         int output_prefix_len=0;
  61         int prev_year=0, prev_month=0, prev_day=0;
  62         struct tm *t;
  63         struct getwordstruct gwarea;
  64         longline line;
  65
  66         if (splitprefix[0]!='\0') {
  67                 // '/' + '-YYYY-mm-dd' + '\0' == 13
  68                 output_prefix_len=snprintf(output_file,sizeof(output_file)-12,"%s%s",outdir,splitprefix);
  69                 if (output_prefix_len>=sizeof(output_file)-12) {
  70                         debuga(__FILE__,__LINE__,_("Path too long: "));
  71                         debuga_more("%s%s-YYYY-mm-dd\n",outdir,splitprefix);
  72                         exit(EXIT_FAILURE);
  73                 }
  74                 autosplit=1;
  75         } else {
  76                 fp_ou=stdout;
  77         }
  78
  79         if(arq[0] == '\0')
  80                 arq="/var/log/squid/access.log";
  81
  82         if((fp_in=MY_FOPEN(arq,"r"))==NULL) {
  83                 debuga(__FILE__,__LINE__,_("Cannot open file \"%s\": %s\n"),arq,strerror(errno));
  84                 exit(EXIT_FAILURE);
  85         }
  86
  87         if ((line=longline_create())==NULL) {
  88                 debuga(__FILE__,__LINE__,_("Not enough memory to read file \"%s\"\n"),arq);
  89                 exit(EXIT_FAILURE);
  90         }
  91         time(&min_tt);
  92
  93         while((buf=longline_read(fp_in,line))!=NULL) {
  94                 getword_start(&gwarea,buf);
  95                 if (getword(data,sizeof(data),&gwarea,' ')<0) {
  96                         debuga(__FILE__,__LINE__,_("Invalid date in file \"%s\"\n"),arq);
  97                         exit(EXIT_FAILURE);
  98                 }
  99                 tt=atoi(data);
 100                 t=localtime(&tt);
 101
 102                 if(dfrom) {
 103                         idata=(t->tm_year+1900)*10000+(t->tm_mon+1)*100+t->tm_mday;
 104                         if(idata < dfrom || idata > duntil)
 105                                 continue;
 106                 }
 107
 108                 if (autosplit && (prev_year!=t->tm_year || prev_month!=t->tm_mon || prev_day!=t->tm_mday)) {
 109                         prev_year=t->tm_year;
 110                         prev_month=t->tm_mon;
 111                         prev_day=t->tm_mday;
 112                         if (fp_ou && fclose(fp_ou)==EOF) {
 113                                 debuga(__FILE__,__LINE__,_("Write error in \"%s\": %s\n"),output_file,strerror(errno));
 114                                 exit(EXIT_FAILURE);
 115                         }
 116                         strftime(output_file+output_prefix_len, sizeof(output_file)-output_prefix_len, "-%Y-%m-%d", t);
 117                         /*
 118                         The line must be added to a file we have already created. The file must be created if the date
 119                         is seen for the first time. The idea is to create the files from scratch if the split is started
 120                         a second time.
 121                         */
 122                         if ((fp_ou=MY_FOPEN(output_file,(tt>=min_tt && tt<=max_tt) ? "a" : "w"))==NULL) {
 123                                 debuga(__FILE__,__LINE__,_("Cannot open file \"%s\": %s\n"),output_file,strerror(errno));
 124                                 exit(EXIT_FAILURE);
 125                         }
 126                         if (tt<min_tt) min_tt=tt;
 127                         if (tt>max_tt) max_tt=tt;
 128                 }
 129
 130                 if(!convert) {
 131                         fprintf(fp_ou,"%s %s\n",data,gwarea.current);
 132                 } else {
 133                         if (df=='e')
 134                                 strftime(dia, sizeof(dia), "%d/%m/%Y", t);
 135                         else
 136                                 strftime(dia, sizeof(dia), "%m/%d/%Y", t);
 137
 138                         fprintf(fp_ou,"%s %02d:%02d:%02d %s\n",dia,t->tm_hour,t->tm_min,t->tm_sec,gwarea.current);
 139                 }
 140         }
 141
 142         longline_destroy(&line);
 143         if (fclose(fp_in)==EOF) {
 144                 debuga(__FILE__,__LINE__,_("Read error in \"%s\": %s\n"),arq,strerror(errno));
 145                 exit(EXIT_FAILURE);
 146         }
 147         if (autosplit && fp_ou) {
 148                 if (fclose(fp_ou)==EOF) {
 149                         debuga(__FILE__,__LINE__,_("Write error in \"%s\": %s\n"),output_file,strerror(errno));
 150                         exit(EXIT_FAILURE);
 151                 }
 152         }
 153 }