/*
    The WebDruid - a web server log analysis program

    Copyright (C) 2003-2004  Fabien Chevalier (fabien@juliana-multimedia.com)

    Original webalizer copyright:
    Copyright (C) 1997-2001  Bradford L. Barrett (brad@mrunix.net)

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version, and provided that the above
    copyright and permission notice is included with all distributed
    copies of this or derived software.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA

    This software uses the gd graphics library, which is copyright by
    Quest Protein Database Center, Cold Spring Harbor Labs.  Please
    see the documentation supplied with the library for additional
    information and license terms, or visit www.boutell.com/gd/ for the
    most recent version of the library and supporting documentation.

    W3C log format suport generously donated by Klaus Reimer <k@ailis.de>

*/

/*********************************************/
/* STANDARD INCLUDES                         */
/*********************************************/

#include "config.h"

#include <time.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>                           /* normal stuff             */
#include <ctype.h>
#include <sys/utsname.h>
#include <sys/times.h>

/* ensure getopt */
#ifdef HAVE_GETOPT_H
#include <getopt.h>
#endif

/* ensure sys/types */
#ifndef _SYS_TYPES_H
#include <sys/types.h>
#endif

/* some systems need this */
#ifdef HAVE_MATH_H
#include <math.h>
#endif

/* SunOS 4.x Fix */
#ifndef CLK_TCK
#define CLK_TCK _SC_CLK_TCK
#endif

#include "webdruid.h"                        /* main header              */
#include "lang.h"
#include "parser.h"

/* internal function prototypes */

static void fmt_logrec(char *);
static int  parse_record_clf(char *, struct log_struct *);
static int  parse_record_w3c(char *buffer, struct log_struct * log_record);

/* month names used for parsing logfile (shouldn't be lang specific) */

static char *log_month[12]={ "jan", "feb", "mar",
                             "apr", "may", "jun",
                             "jul", "aug", "sep",
                             "oct", "nov", "dec"};

static char tmp_buf[BUFSIZE];

/*********************************************/
/* FMT_LOGREC - terminate log fields w/zeros */
/*********************************************/

static void fmt_logrec(char *buffer)
{
   char *cp=buffer;
   int  q=0,b=0,p=0;

   while (*cp != '\0')
   {
      /* break record up, terminate fields with '\0' */
      switch (*cp)
      {
       case ' ': if (b || q || p) break; *cp='\0'; break;
       case '"': q^=1;  break;
       case '[': if (q) break; b++; break;
       case ']': if (q) break; if (b>0) b--; break;
       case '(': if (q) break; p++; break;
       case ')': if (q) break; if (p>0) p--; break;
       case '\t': if (b || q || p) break; *cp='\0'; break;
      }
      cp++;
   }
}

/*********************************************/
/* PARSE_RECORD - uhhh, you know...          */
/*********************************************/

int parse_record(char *buffer, struct log_struct * log_record)
{
   int ret;

   /* clear out structure */
   memset(log_record, 0, sizeof(struct log_struct));

   /* dates initialisation */

   log_record->year  = 0;
   log_record->month = 0;
   log_record->day   = 0;
   log_record->hour  = 0;
   log_record->min   = 0;
   log_record->sec   = 0;

#ifdef USE_DNS
   #ifdef USE_IPV6
   memset(&log_record->addr,0,sizeof(struct sockaddr_storage));
   #else
   memset(&log_rec.addr,0,sizeof(struct in_addr));
   #endif
#endif

   /* call appropriate handler */
   switch (log_type)
   {
      default:
      case LOG_CLF:   ret = parse_record_clf(buffer, log_record);   break; /* clf   */
      case LOG_W3C:   ret = parse_record_w3c(buffer, log_record);   break; /* w3c   */
   }

   /*
   printf("%d %d %d %d %d %d\n", log_record->year, log_record->month, log_record->day,
            log_record->hour, log_record->min, log_record->sec);
   */

   if(ret == 1)
      return 0;
   else
      return 1;
}

/*********************************************/
/* PARSE_RECORD_CLF - web log handler        */
/*********************************************/

static int parse_record_clf(char *buffer, struct log_struct * log_record)
{
   int size, i;
   char *cp1, *cp2, *cpx, *eob, *eos;
   char datetime[29];  /* raw timestamp        */

   size = strlen(buffer);                 /* get length of buffer        */
   eob = buffer+size;                     /* calculate end of buffer     */
   fmt_logrec(buffer);                    /* seperate fields with \0's   */

   /* HOSTNAME */
   cp1 = cpx = buffer; cp2=log_record->hostname;
   eos = (cp1+MAXHOST)-1;
   if (eos >= eob) eos=eob-1;

   while ( (*cp1 != '\0') && (cp1 != eos) ) *cp2++ = *cp1++;
   *cp2 = '\0';
   if (*cp1 != '\0')
   {
      if (verbose)
      {
         fprintf(stderr,"%s",_("Warning: Truncating oversized hostname"));
         if (debug_mode) fprintf(stderr,": %s\n",cpx);
         else fprintf(stderr,"\n");
      }
      while (*cp1 != '\0') cp1++;
   }
   if (cp1 < eob) cp1++;

   /* skip next field (ident) */
   while ( (*cp1 != '\0') && (cp1 < eob) ) cp1++;
   if (cp1 < eob) cp1++;

   /* IDENT (authuser) field */
   cpx = cp1;
   cp2 = log_record->ident;
   eos = (cp1+MAXIDENT-1);
   if (eos >= eob) eos=eob-1;

   while ( (*cp1 != '[') && (cp1 < eos) ) /* remove embeded spaces */
   {
      if (*cp1=='\0') *cp1=' ';
      *cp2++=*cp1++;
   }
   *cp2--='\0';

   if (cp1 >= eob) return 0;

   /* check if oversized username */
   if (*cp1 != '[')
   {
      if (verbose)
      {
         fprintf(stderr,"%s",_("Warning: Truncating oversized username"));
         if (debug_mode) fprintf(stderr,": %s\n",cpx);
         else fprintf(stderr,"\n");
      }
      while ( (*cp1 != '[') && (cp1 < eob) ) cp1++;
   }

   /* strip trailing space(s) */
   while (*cp2==' ') *cp2--='\0';

   /* date/time string */
   cpx = cp1;
   cp2 = datetime;
   eos = (cp1+28);
   if (eos >= eob) eos=eob-1;

   while ( (*cp1 != '\0') && (cp1 != eos) ) *cp2++ = *cp1++;
   *cp2 = '\0';
   if (*cp1 != '\0')
   {
      if (verbose)
      {
         fprintf(stderr,"%s",_("Warning: Truncating oversized date field"));
         if (debug_mode) fprintf(stderr,": %s\n",cpx);
         else fprintf(stderr,"\n");
      }
      while (*cp1 != '\0') cp1++;
   }
   if (cp1 < eob) cp1++;

   /* minimal sanity check on timestamp */
   if ( (datetime[0] != '[') ||
        (datetime[3] != '/') ||
        (cp1 >= eob))  return 0;

   /* HTTP request */
   cpx = cp1;
   cp2 = log_record->url;
   eos = (cp1+MAXURL-1);
   if (eos >= eob) eos = eob-1;

   while ( (*cp1 != '\0') && (cp1 != eos) ) *cp2++ = *cp1++;
   *cp2 = '\0';
   if (*cp1 != '\0')
   {
      if (verbose)
      {
         fprintf(stderr,"%s",_("Warning: Truncating oversized request field"));
         if (debug_mode) fprintf(stderr,": %s\n",cpx);
         else fprintf(stderr,"\n");
      }
      while (*cp1 != '\0') cp1++;
   }
   if (cp1 < eob) cp1++;

   if ( (log_record->url[0] != '"') ||
        (cp1 >= eob) ) return 0;

   /* response code */
   log_record->resp_code = atoi(cp1);

   /* xfer size */
   while ( (*cp1 != '\0') && (cp1 < eob) ) cp1++;
   if (cp1 < eob) cp1++;
   if (*cp1<'0'||*cp1>'9') log_record->xfer_size=0;
   else log_record->xfer_size = strtoul(cp1,NULL,10);

   /* done with CLF record  -- check if we have more stuff to read*/
   if (cp1 < eob)
   {
      while ( (*cp1 != '\0') && (*cp1 != '\n') && (cp1 < eob) ) cp1++;
      if (cp1 < eob) cp1++;
      /* get referrer if present */
      cpx = cp1;
      cp2 = log_record->refer;
      eos = (cp1+MAXREF-1);
      if (eos >= eob) eos = eob-1;

      while ( (*cp1 != '\0') && (*cp1 != '\n') && (cp1 != eos) ) *cp2++ = *cp1++;
      *cp2 = '\0';
      if (*cp1 != '\0')
      {
         if (verbose)
         {
            fprintf(stderr,"%s",_("Warning: Truncating oversized referrer field"));
            if (debug_mode) fprintf(stderr,": %s\n",cpx);
            else fprintf(stderr,"\n");
         }
         while (*cp1 != '\0') cp1++;
      }
      if (cp1 < eob) cp1++;

      cpx = cp1;
      cp2 = log_record->agent;
      eos = cp1+(MAXAGENT-1);
      if (eos >= eob) eos = eob-1;

      while ( (*cp1 != '\0') && (cp1 != eos) ) *cp2++ = *cp1++;
      *cp2 = '\0';
   }

   /* do backup copy the buffer, in case it gets damaged by our tries to analyse it */
   strncpy(tmp_buf, buffer, BUFSIZE);

   /* convert month name to lowercase */
   for (i=4; i<7; i++)
         datetime[i]=tolower(datetime[i]);

   /* get year/month/day/hour/min/sec values    */
   log_record->month = 1;
   for (i=0; i<12; i++)
   {
         if (strncmp(log_month[i],&datetime[4],3)==0)
         { log_record->month = i+1; break; }
   }

   log_record->year = atoi(&datetime[8]);    /* get year number (int)   */
   log_record->day  = atoi(&datetime[1]);    /* get day number          */
   log_record->hour = atoi(&datetime[13]);   /* get hour number         */
   log_record->min  = atoi(&datetime[16]);   /* get minute number       */
   log_record->sec  = atoi(&datetime[19]);   /* get second number       */

   /* minimal sanity check on date */
   if ((i>=12)||(log_record->min>59)||(log_record->sec>59)||(log_record->year<1990))
   {
         total_bad++;                /* if a bad date, bump counter      */
         if (verbose)
         {
            fprintf(stderr,"%s: %s [%lu]",
               _("Error: bad record (bad date)"),datetime,total_rec);
            if (debug_mode)
               fprintf(stderr,":\n%s\n",tmp_buf);
            else
               fprintf(stderr,"\n");
         }
         return 0;                   /* and ignore this record           */
   }

   return 1;     /* maybe a valid record, return with TRUE */
}

/*********************************************/
/* PARSE_RECORD_W3C - w3c log handler        */
/*********************************************/

/*
   Patch: webalizer-w3c
   For: Webalizer 2.01.10
   From: Klaus Reimer <k@ailis.de>
   Desc: Implement W3C extended log file format support.
   Version: 3

   This patch implements W3C extended log file format support into webalizer.
   Thig log file format is used by Microsoft's Webserver IIS but other programs
   may also use this format which was invented by W3C. With this patch
   webalizer can read the W3C headers and can parse the data lines according to
   the specified Fields-header. Format-switching within a single logfile is
   supported. IIS is using this format-switching to write process accounting
   informations to the same logfile. These accounting informations are
   producing "bad record" warnings because these lines doesn't hold any
   informations usable for webalizer. You can just ignore them or disable
   process accounting logs in the IIS configuration if you don't need them

   IIS seems to be very buggy. I found three bugs in the current IIS version
   which affects logfiles:
   1. If output buffering is enabled in ASP scripts the sc-bytes field is
      always 0. This prevents accurate measurement of download traffic.
   2. Sometimes IIS writes nonsense into the log so the number of entries
      are not conform to the number of fields specified in the logfile header.
      This patch does its best to filter out these lines.
   3. IIS writes a lot of Null-characters to the end of each logfile. These
      lines are ignored by this patch.

   Known Bugs in this patch: The user-agent field is not completely
   "urldecoded". Only the '+' signs are replaced by space characters. Other
   url-encoded characters are not touched.
*/

/* field index structure */
struct  field_index_struct
{
   int date;     /* Date field index */
   int time;     /* Time field index */
   int ip;       /* IP field index */
   int username; /* Username field index */
   int method;   /* Method field index */
   int url;      /* URL field index    */
   int query;    /* Querystring field index */
   int status;   /* Status code field index */
   int size;     /* Size field index */
   int referer;  /* Referrer field index */
   int agent;    /* User agent field index */
   int fields;   /* Number of fields in this format */
};

/* field structure */
struct  fields_struct
{
   char *date;     /* Date field */
   char *time;     /* Time field */
   char *ip;       /* IP field */
   char *username; /* Username field */
   char *url;      /* URL field */
   char *query;    /* Querystring */
   char *status;   /* Status code */
   char *size;     /* Size field */
   char *referer;  /* Referrer field */
   char *agent;    /* User agent field */
};

static int parse_record_w3c(char *buffer, struct log_struct * log_record)
{
   int size;
   char *eob;
   char *cp;
   int index;
   static struct field_index_struct field_index;
   struct fields_struct fields;
   struct tm time;

   memset(&time, 0, sizeof(struct tm));
   size = strlen(buffer);                 /* get length of buffer        */
   eob = buffer + size;                   /* calculate end of buffer     */
   fmt_logrec(buffer);                    /* seperate fields with \0's   */

   cp = buffer;
   
   /* Check if the line is empty or a line suffers from the IIS
      Null-Character bug and abort parsing if found. */
   if (*cp == '\0') return 0;

   /* If it's a header line ignore it or parse the Fields header if found */
   if (*cp == '#')
   {
      cp++;
      if (!strcmp(cp, "Fields:"))
      {
         /* Reset the field indices */
         memset(&field_index, 0, sizeof(struct field_index_struct));
         while (*cp) cp++;
         cp++;
         index = 1;
         while ((cp < eob) && (*cp != '\r') && (*cp != '\n'))
         {
            /* Set the field index */
            if (!strcmp(cp, "date")) field_index.date = index;
            if (!strcmp(cp, "time")) field_index.time = index;
            if (!strcmp(cp, "c-ip")) field_index.ip = index;
            if (!strcmp(cp, "cs-uri-stem")) field_index.url = index;
            if (!strcmp(cp, "cs-uri-query")) field_index.query = index;
            if (!strcmp(cp, "sc-status")) field_index.status = index;
            if (!strcmp(cp, "cs(Referer)")) field_index.referer = index;
            if (!strcmp(cp, "sc-bytes")) field_index.size = index;
            if (!strcmp(cp, "cs(User-Agent)")) field_index.agent = index;
            if (!strcmp(cp, "cs-username")) field_index.username = index;

            /* Continue with the next field */
            while (*cp) cp++;
            cp++;
            index++;
         }
         field_index.fields = index -1;
      }
      

      /* Return because this header line is completely parsed */
      return 0;
   }
   
   /* A data line has been found */
   
   /* Check if the number of entries in this line are conform to the
      format specified in the header */
   index = 1;
   while ((cp < eob) && (*cp != '\r') && (*cp != '\n')) {
      while (*cp) cp++;
      cp++;
      index++;
   }
   if (index-1 != field_index.fields) return 0;
   
   /* Reset pointer */
   cp = buffer;

   /* Reset the field pointers and begin parsing the data line */
   memset(&fields, 0, sizeof(struct fields_struct));
   index = 1;
   while ((cp < eob) && (*cp != '\r') && (*cp != '\n'))
   {
      /* Set the field pointers */
      if (index == field_index.date) fields.date = cp;
      if (index == field_index.time) fields.time = cp;
      if (index == field_index.ip) fields.ip = cp;
      if (index == field_index.url) fields.url = cp;
      if (index == field_index.query) fields.query = cp;
      if (index == field_index.status) fields.status = cp;
      if (index == field_index.referer) fields.referer = cp;
      if (index == field_index.size) fields.size = cp;
      if (index == field_index.agent) fields.agent = cp;
      if (index == field_index.username) fields.username = cp;

      /* Continue with the next data field */
      while (*cp) cp++;
      cp++;
      index++;
   }

   /* Save URL */
   if (fields.url)
   {
      if (fields.query && (fields.query[0]!='-'))
         snprintf(log_record->url, MAXURL, "%s?%s", fields.url, fields.query);
      else
         strncpy(log_record->url, fields.url, MAXURL - 1);
   } else return 0;

   /* Save hostname */
   if (fields.ip) strncpy(log_record->hostname, fields.ip, MAXHOST - 1);

   /* Save response code */
   if (fields.status) log_record->resp_code = atoi(fields.status);

   /* Save referer */
   if (fields.referer) strncpy(log_record->refer, fields.referer, MAXREF - 1);

   /* Save transfer size */
   if (fields.size) log_record->xfer_size = strtoul(fields.size, NULL, 10);

   /* Save user agent */
   if (fields.agent)
   {
      cp = fields.agent;
      while (*cp)
      {
        if (*cp == '+') *cp = ' ';
        cp++;
      }
      strncpy(log_record->agent, fields.agent, MAXAGENT - 1);
   }

   /* Save auth username */
   if (fields.username) strncpy(log_record->ident, fields.username, MAXIDENT - 1);

   /* Parse date and time and save it */
   if (fields.date)
   {
      log_record->year = atoi(fields.date);
      while ((fields.date[0] != '\0') && (fields.date[0] != '-')) fields.date++;
      if (fields.date[0] == '\0') return 0;
      fields.date++;
      log_record->month = atoi(fields.date);
      while ((fields.date[0] != '\0') && (fields.date[0] != '-')) fields.date++;
      if (fields.date[0] == '\0') return 0;
      fields.date++;
      log_record->day = atoi(fields.date);
   }
   if (fields.time)
   {
      log_record->hour = atoi(fields.time);
      while ((fields.time[0] != '\0') && (fields.time[0] != ':')) fields.time++;
      if (fields.time[0] == '\0') return 0;
      fields.time++;
      log_record->min = atoi(fields.time);
      while ((fields.time[0] != '\0') && (fields.time[0] != ':')) fields.time++;
      if (fields.time[0] == '\0') return 0;
      fields.time++;
      log_record->sec = atoi(fields.time);
   }

   return 1;
}