readlog_common.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 /*!
  32 A new file is being read. The name of the file is \a FileName.
  33 */
  34 static void Common_NewFile(const char *FileName)
  35 {
  36 }
  37
  38 /*!
  39 Extract a column containing a long long int from \a Line.
  40
  41 The extracted value is stored in \a Value.
  42
  43 The pointer to the next byte just after the number is returned
  44 by the function.
  45 */
  46 static char *Common_GetLongLongInt(char *Line,long long int *Value)
  47 {
  48         *Value=0LL;
  49         if (*Line=='-') {
  50                 ++Line;
  51         } else {
  52                 while (isdigit(*Line)) *Value=*Value*10+(*Line++-'0');
  53         }
  54         return(Line);
  55 }
  56
  57 /*!
  58 Read one entry from a standard squid log format.
  59
  60 \param Line One line from the input log file.
  61 \param Entry Where to store the information parsed from the line.
  62
  63 \retval RLRC_NoError One valid entry is parsed.
  64 \retval RLRC_Unknown The line is invalid.
  65 \retval RLRC_InternalError An internal error was encountered.
  66 */
  67 static enum ReadLogReturnCodeEnum Common_ReadEntry(char *Line,struct ReadLogStruct *Entry)
  68 {
  69         const char *Begin;
  70         int IpLen;
  71         int HttpCodeLen;
  72         int UrlLen;
  73         int UserLen;
  74         int Day;
  75         char MonthName[4];
  76         int MonthNameLen;
  77         int Month;
  78         int Year;
  79         int Hour;
  80         int Minute;
  81         int Second;
  82         char *Ip;
  83         char *User;
  84
  85         // get IP address
  86         Entry->Ip=Ip=Line;
  87         for (IpLen=0 ; *Line && *Line!=' ' ; IpLen++) Line++;
  88         if (*Line!=' ' || IpLen==0) return(RLRC_Unknown);
  89
  90         if (!squid24) {
  91                 // squid version <= 2.4 store the user ID in the second column: skip the first column here
  92                 Begin=++Line;
  93                 while (*Line && *Line!=' ') Line++;
  94                 if (*Line!=' '|| Line==Begin) return(RLRC_Unknown);
  95         }
  96
  97         // the ID of the user or - if the user is unidentified
  98         Entry->User=User=++Line;
  99         for (UserLen=0 ; *Line && *Line!=' ' ; UserLen++) Line++;
 100         if (*Line!=' ' || UserLen==0) return(RLRC_Unknown);
 101
 102         if (squid24) {
 103                 // squid version > 2.4 store the user ID in the first column: skip the second column here
 104                 Begin=++Line;
 105                 while (*Line && *Line!=' ') Line++;
 106                 if (*Line!=' '|| Line==Begin) return(RLRC_Unknown);
 107         }
 108
 109         // get the date enclosed within square brackets
 110         ++Line;
 111         if (*Line!='[') return(RLRC_Unknown);
 112         ++Line;
 113         Day=0;
 114         while (isdigit(*Line)) Day=Day*10+(*Line++-'0');
 115         if (*Line!='/' || Day<1 || Day>31) return(RLRC_Unknown);
 116
 117         ++Line;
 118         for (MonthNameLen=0 ; MonthNameLen<sizeof(MonthName)-1 && isalpha(*Line) ; MonthNameLen++) MonthName[MonthNameLen]=*Line++;
 119         if (*Line!='/') return(RLRC_Unknown);
 120         MonthName[MonthNameLen]='\0';
 121         Month=month2num(MonthName);
 122         if (Month>=12) return(RLRC_Unknown);
 123
 124         ++Line;
 125         Year=0;
 126         while (isdigit(*Line)) Year=Year*10+(*Line++-'0');
 127         if (*Line!=':' || Year<1900 || Year>2200) return(RLRC_Unknown);
 128
 129         // get the time
 130         ++Line;
 131         Hour=0;
 132         while (isdigit(*Line)) Hour=Hour*10+(*Line++-'0');
 133         if (*Line!=':' || Hour>=24) return(RLRC_Unknown);
 134         ++Line;
 135         Minute=0;
 136         while (isdigit(*Line)) Minute=Minute*10+(*Line++-'0');
 137         if (*Line!=':' || Minute>=60) return(RLRC_Unknown);
 138         ++Line;
 139         Second=0;
 140         while (isdigit(*Line)) Second=Second*10+(*Line++-'0');
 141         if (*Line!=' ' || Second>60) return(RLRC_Unknown); //second can be 60 due to a leap second
 142
 143         // skip the timezone up to the closing ]
 144         while (*Line && *Line!=']') Line++;
 145         if (*Line!=']') return(RLRC_Unknown);
 146
 147         Entry->EntryTime.tm_year=Year-1900;
 148         Entry->EntryTime.tm_mon=Month;
 149         Entry->EntryTime.tm_mday=Day;
 150         Entry->EntryTime.tm_hour=Hour;
 151         Entry->EntryTime.tm_min=Minute;
 152         Entry->EntryTime.tm_sec=Second;
 153         Entry->EntryTime.tm_isdst=-1;
 154
 155         // the URL is enclosed between double qhotes
 156         ++Line;
 157         if (*Line!=' ') return(RLRC_Unknown);
 158         ++Line;
 159         if (*Line!='\"') return(RLRC_Unknown);
 160
 161         // skip the HTTP function
 162         Begin=++Line;
 163         while (isalpha(*Line)) Line++;
 164         if (*Line!=' ' || Line==Begin) return(RLRC_Unknown);
 165
 166         // get the URL
 167         Entry->Url=++Line;
 168         for (UrlLen=0 ; *Line && *Line!=' ' ; UrlLen++) Line++;
 169         if (*Line!=' ' || UrlLen==0) return(RLRC_Unknown);
 170
 171         // skip the HTTP/...
 172         ++Line;
 173         while (*Line && *Line!='\"') Line++;
 174         if (*Line!='\"') return(RLRC_Unknown);
 175         ++Line;
 176         if (*Line!=' ') return(RLRC_Unknown);
 177
 178         // get the HTTP code.
 179         Entry->HttpCode=++Line;
 180         for (HttpCodeLen=0 ; *Line && *Line!=' ' ; HttpCodeLen++) Line++;
 181         if (*Line!=' ' || HttpCodeLen==0) return(RLRC_Unknown);
 182
 183         // get the number of transfered bytes.
 184         Begin=++Line;
 185         Line=Common_GetLongLongInt(Line,&Entry->DataSize);
 186         // some log contains more columns
 187         if ((*Line && *Line!=' ') || Begin==Line) return(RLRC_Unknown);
 188
 189         // check the entry time
 190         if (mktime(&Entry->EntryTime)==-1) {
 191                 debuga(__FILE__,__LINE__,_("Invalid date or time found in the common log file\n"));
 192                 return(RLRC_InternalError);
 193         }
 194
 195         // it is safe to alter the line buffer now that we are returning a valid entry
 196         Ip[IpLen]='\0';
 197         Entry->HttpCode[HttpCodeLen]='\0';
 198         Entry->Url[UrlLen]='\0';
 199         User[UserLen]='\0';
 200
 201         return(RLRC_NoError);
 202 }
 203
 204 //! \brief Object to read a standard common log format.
 205 const struct ReadLogProcessStruct ReadCommonLog=
 206 {
 207         /* TRANSLATORS: This is the name of the log format displayed when this format is detected in an input log file. */
 208         N_("common log format"),
 209         Common_NewFile,
 210         Common_ReadEntry
 211 };