readlog_sarg.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 #include "include/readlog.h"
  30
  31 //! \c True if the current log is known to be a sarg parsed log.
  32 static bool InSargLog=false;
  33 //! \c True if the file name is invalid.
  34 static bool InvalidFileName=true;
  35 //! The last period extracted from the log file name.
  36 static struct periodstruct SargPeriod;
  37
  38 /*!
  39 A new file is being read. The name of the file is \a FileName.
  40 */
  41 static void Sarg_NewFile(const char *FileName)
  42 {
  43         InSargLog=false;
  44         InvalidFileName=(getperiod_fromsarglog(FileName,&SargPeriod)<0);
  45 }
  46
  47 /*!
  48 Read one entry from a sarg generated log.
  49
  50 \param Line One line from the input log file.
  51 \param Entry Where to store the information parsed from the line.
  52
  53 \retval RLRC_NoError One valid entry is parsed.
  54 \retval RLRC_Unknown The line is invalid.
  55 \retval RLRC_InternalError An internal error was encountered.
  56 */
  57 static enum ReadLogReturnCodeEnum Sarg_ReadEntry(char *Line,struct ReadLogStruct *Entry)
  58 {
  59         const char *Begin;
  60         int IpLen;
  61         int HttpCodeLen;
  62         int UrlLen;
  63         int UserLen;
  64         int Day;
  65         int Month;
  66         int Year;
  67         int Hour;
  68         int Minute;
  69         int Second;
  70         char *Ip;
  71         char *User;
  72
  73         if (strncmp(Line,"*** SARG Log ***",16)==0) {
  74                 if (InvalidFileName) {
  75                         debuga(__FILE__,__LINE__,_("The name of the file is invalid for a sarg log\n"));
  76                         exit(EXIT_FAILURE);
  77                 }
  78                 getperiod_merge(&period,&SargPeriod);
  79                 InSargLog=true;
  80                 return(RLRC_Ignore);
  81         }
  82         if (!InSargLog) return(RLRC_Unknown);
  83
  84         // get the date
  85         Day=0;
  86         while (isdigit(*Line)) Day=Day*10+(*Line++-'0');
  87         if (*Line!='/' || Day<1 || Day>31) return(RLRC_Unknown);
  88
  89         ++Line;
  90         Month=0;
  91         while (isdigit(*Line)) Month=Month*10+(*Line++-'0');
  92         if (*Line!='/') return(RLRC_Unknown);
  93         if (Month<=0 || Month>12) return(RLRC_Unknown);
  94
  95         ++Line;
  96         Year=0;
  97         while (isdigit(*Line)) Year=Year*10+(*Line++-'0');
  98         if (*Line!='\t' || Year<1900 || Year>2200) return(RLRC_Unknown);
  99
 100         // get the time
 101         ++Line;
 102         Hour=0;
 103         while (isdigit(*Line)) Hour=Hour*10+(*Line++-'0');
 104         if (*Line!=':' || Hour>=24) return(RLRC_Unknown);
 105         ++Line;
 106         Minute=0;
 107         while (isdigit(*Line)) Minute=Minute*10+(*Line++-'0');
 108         if (*Line!=':' || Minute>=60) return(RLRC_Unknown);
 109         ++Line;
 110         Second=0;
 111         while (isdigit(*Line)) Second=Second*10+(*Line++-'0');
 112         if (*Line!='\t' || Second>60) return(RLRC_Unknown); //second can be 60 due to a leap second
 113
 114         Entry->EntryTime.tm_year=Year-1900;
 115         Entry->EntryTime.tm_mon=Month-1;
 116         Entry->EntryTime.tm_mday=Day;
 117         Entry->EntryTime.tm_hour=Hour;
 118         Entry->EntryTime.tm_min=Minute;
 119         Entry->EntryTime.tm_sec=Second;
 120         Entry->EntryTime.tm_isdst=-1;
 121
 122         // the ID of the user
 123         Entry->User=User=++Line;
 124         for (UserLen=0 ; *Line && *Line!='\t' ; UserLen++) Line++;
 125         if (*Line!='\t' || UserLen==0) return(RLRC_Unknown);
 126
 127         // get IP address
 128         Entry->Ip=Ip=++Line;
 129         for (IpLen=0 ; *Line && *Line!='\t' ; IpLen++) Line++;
 130         if (*Line!='\t' || IpLen==0) return(RLRC_Unknown);
 131
 132         // get the URL
 133         Entry->Url=++Line;
 134         for (UrlLen=0 ; *Line && *Line!='\t' ; UrlLen++) Line++;
 135         if (*Line!='\t' || UrlLen==0) return(RLRC_Unknown);
 136
 137         // get the number of transfered bytes.
 138         Begin=++Line;
 139         Entry->DataSize=0LL;
 140         while (isdigit(*Line)) Entry->DataSize=Entry->DataSize*10+(*Line++-'0');
 141         if (*Line!='\t' || Begin==Line) return(RLRC_Unknown);
 142
 143         // get the HTTP code.
 144         Entry->HttpCode=++Line;
 145         for (HttpCodeLen=0 ; *Line && *Line!='\t' ; HttpCodeLen++) Line++;
 146         if (*Line!='\t' || HttpCodeLen==0) return(RLRC_Unknown);
 147
 148         // get the elapsed time.
 149         Begin=++Line;
 150         Entry->ElapsedTime=0L;
 151         while (isdigit(*Line)) Entry->ElapsedTime=Entry->ElapsedTime*10+(*Line++-'0');
 152         if (*Line!='\t' || Line==Begin) return(RLRC_Unknown);
 153
 154         // get the smart filter
 155         //! \bug Smart filter ignored from sarg log format.
 156
 157         // check the entry time
 158         if (mktime(&Entry->EntryTime)==-1) {
 159                 debuga(__FILE__,__LINE__,_("Invalid date or time found in the common log file\n"));
 160                 return(RLRC_InternalError);
 161         }
 162
 163         // it is safe to alter the line buffer now that we are returning a valid entry
 164         Ip[IpLen]='\0';
 165         Entry->HttpCode[HttpCodeLen]='\0';
 166         Entry->Url[UrlLen]='\0';
 167         User[UserLen]='\0';
 168
 169         return(RLRC_NoError);
 170 }
 171
 172 //! \brief Object to read a standard squid log format.
 173 const struct ReadLogProcessStruct ReadSargLog=
 174 {
 175         /* TRANSLATORS: This is the name of the log format displayed when this format is detected in an input log file. */
 176         N_("sarg log format"),
 177         Sarg_NewFile,
 178         Sarg_ReadEntry
 179 };