Read longer lines from access.log and abort if line is too long

diff --git a/CMakeLists.txt b/CMakeLists.txt
index a802e0bff8fd193190b60e7fc90431067399e84f..e005ab5d0d0683a40c182000aabd40d62e66565e 100755 (executable)
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -3,7 +3,7 @@ PROJECT(sarg C)
 SET(sarg_VERSION 2)
 SET(sarg_REVISION 2)
 SET(sarg_BUILD "7")
-SET(sarg_BUILDDATE "Feb-02-2010")
+SET(sarg_BUILDDATE "Feb-04-2010")
 
 INCLUDE(AddFileDependencies)
 INCLUDE(CheckIncludeFile)

diff --git a/ChangeLog b/ChangeLog
index 08453da1911e5c04c8c87db1fcd1ac09da34807e..2301f50592dda0d3423e6e893fa5dc59654afcba 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,6 +1,6 @@
 SARG ChangeLog
 
-Jan-29-2010 Version 2.2.7
+Feb-04-2010 Version 2.2.7
                - Extra compile and run time protection (FORTIFY_SOURCE) fixed in configure.
                - Use tabulations as columns separator in intermediary files to avoid problems when a field of the log contains a space.
                - Input log file type detection partly rewritten to clearly distinguish which type is processed where.
@@ -25,6 +25,7 @@ Jan-29-2010 Version 2.2.7
                - Does not report URLs consisting of only a host name ending with .com as a downloaded file.
                - Mangle the $ and @ signs in user ID to make the resulting file name valid in shell commands.
                - Fixed a regression in the default value of --enable-sargphp.
+               - Increase the maximum size of a line read from access.log and detect longer lines that were silently splitted in previous versions.
 
 Jan-06-2010 Version 2.2.6.1
                - Remove unnecessary dependency on off_t.

diff --git a/include/info.h b/include/info.h
index 728f0f0c60e5b685aaad06665ee7ab930d142bb2..e0693c948ddf3ed8e3ae8db2abd285f944d4644c 100755 (executable)
--- a/include/info.h
+++ b/include/info.h
@@ -1,3 +1,3 @@
-#define VERSION PACKAGE_VERSION" Feb-02-2010"
+#define VERSION PACKAGE_VERSION" Feb-04-2010"
 #define PGM PACKAGE_NAME
 #define URL "http://sarg.sourceforge.net"

diff --git a/log.c b/log.c
index ffb4d963f0fbd097af9fabbff7ed1e63b722c766..366909bfc186ae92f0ef4313643c4afd7bca671e 100644 (file)
--- a/log.c
+++ b/log.c
@@ -98,8 +98,7 @@ int main(int argc,char *argv[])
    char tbuf2[128];
    char zip[20];
    char *str;
-   char bufz[MAXLEN];
-   char bufy[MAXLEN];
+   char bufz[2*MAXLEN];
    char tmp2[MAXLEN];
    char start_hour[128];
    char end_hour[128];
@@ -118,6 +117,7 @@ int main(int argc,char *argv[])
    int  exstring=0;
    int isa_ncols=0,isa_cols[ISACOL_Last];
    int from_stdin;
+   int blen;
    long totregsl=0;
    long totregsg=0;
    long totregsx=0;
@@ -731,6 +731,11 @@ int main(int argc,char *argv[])
       }
 
       while(fgets(bufz,sizeof(bufz),fp_in)!=NULL) {
+         blen=strlen(bufz);
+         if (blen>0 && bufz[blen-1]!='\r' && bufz[blen-1]!='\n' && !feof(fp_in)) {
+            fprintf(stderr,"SARG: line too long (more than %d bytes) in %s\n",sizeof(bufz)-1,arq);
+            exit(1);
+         }
 
          if (ilf==ILF_Unknown) {
             if(strncmp(bufz,"#Software: Mic",14) == 0) {
@@ -776,12 +781,10 @@ int main(int argc,char *argv[])
            fflush (stdout);
            OutputNonZero = REPORT_EVERY_X_LINES ;
          }
-         if(strlen(bufz) > MAXLEN-1) continue;
-         if(!bufz[0]) continue;
+         if(blen < 58) continue;
          if(strstr(bufz,"HTTP/0.0") != 0) continue;
          if(strstr(bufz,"logfile turned over") != 0) continue;
          if(bufz[0] == ' ') continue;
-         if(strlen(bufz) < 58) continue;
 
          // Record only hours usage which is required
          tt = (time_t) strtoul( bufz, NULL, 10 );
@@ -798,22 +801,20 @@ int main(int argc,char *argv[])
          // exclude_string
          exstring=0;
          if(ExcludeString[0] != '\0') {
-            strcpy(warea,bufz);
             getword_start(&gwarea,ExcludeString);
             while(strchr(gwarea.current,':') != 0) {
                if (getword_multisep(val1,sizeof(val1),&gwarea,':')<0) {
                   printf("SARG: Maybe you have a broken record or garbage in your exclusion string.\n");
                   exit(1);
                }
-               if((str=(char *) strstr(warea,val1)) != (char *) NULL )
+               if((str=(char *) strstr(bufz,val1)) != (char *) NULL )
                   exstring++;
             }
-            if((str=(char *) strstr(warea,gwarea.current)) != (char *) NULL )
+            if((str=(char *) strstr(bufz,gwarea.current)) != (char *) NULL )
                   exstring++;
          }
          if(exstring) continue;
 
-         strcpy(bufy,bufz);
          if ((str = strchr(bufz, '\n')) != NULL)
             *str = '\0';          /* strip \n */
 
@@ -824,7 +825,7 @@ int main(int argc,char *argv[])
          if (ilf==ILF_Squid || ilf==ILF_Common || ilf==ILF_Unknown) {
             getword_start(&gwarea,bufz);
             if (getword(data,sizeof(data),&gwarea,' ')<0) {
-               printf("SARG: Maybe you have a broken record or garbage in your access.log file.\n");
+               printf("SARG: Maybe you have a broken time in your access.log file.\n");
                exit(1);
             }
             if((str=(char *) strchr(data, '.')) != (char *) NULL ) {
@@ -874,33 +875,33 @@ int main(int argc,char *argv[])
 
             if(ilf==ILF_Unknown || ilf==ILF_Squid) {
                if (getword(elap,sizeof(elap),&gwarea,' ')<0) {
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken elapsed time in your %s file.\n",arq);
                   exit(1);
                }
                while(strcmp(elap,"") == 0 && gwarea.current[0] != '\0')
                   if (getword(elap,sizeof(elap),&gwarea,' ')<0) {
-                     printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                     printf("SARG: Maybe you have a broken elapsed time in your %s file.\n",arq);
                      exit(1);
                   }
                if(strlen(elap) < 1) continue;
                if (getword(ip,sizeof(ip),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken client IP address in your %s file.\n",arq);
                   exit(1);
                }
                if (getword(code,sizeof(code),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken result code in your %s file.\n",arq);
                   exit(1);
                }
                if (getword(tam,sizeof(tam),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken amount of data in your %s file.\n",arq);
                   exit(1);
                }
                if (getword(fun,sizeof(fun),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken request method in your %s file.\n",arq);
                   exit(1);
                }
                if (getword(url,sizeof(url),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken URI in your %s file.\n",arq);
                   exit(1);
                }
 //              while (strstr(bufz,"%20") != 0) {
@@ -908,7 +909,7 @@ int main(int argc,char *argv[])
 //                 strcat(url,warea);
 //              }
                if (getword(user,sizeof(user),&gwarea,' ')<0){
-                  printf("SARG: Maybe you have a broken record or garbage in your %s file.\n",arq);
+                  printf("SARG: Maybe you have a broken user ID in your %s file.\n",arq);
                   exit(1);
                }
                ilf=ILF_Squid;