[GNUnet-SVN] [taler-schemafuzz] branch master updated: Parsing stack traces

[gnunet]

This is an automated email from the git hooks/post-receive script.

erwan-ulrich pushed a commit to branch master
in repository schemafuzz.

The following commit(s) were added to refs/heads/master by this push:
     new f9d67f1  Parsing stack traces
f9d67f1 is described below

commit f9d67f1648c17c75169c0f7d9c76d492c84fbea4
Author: Feideus <address@hidden>
AuthorDate: Wed Jul 4 18:32:14 2018 +0200

    Parsing stack traces
---
 errorReports/core                    | Bin 0 -> 389120 bytes
 errorReports/stackTrace_test_c_crash |   3 +
 intersting_script.py                 | 168 -----------------------------------
 stackTraceCParser.sh                 |  49 ++++++++++
 test_c_crash.c                       |  22 +++++
 5 files changed, 74 insertions(+), 168 deletions(-)

diff --git a/errorReports/core b/errorReports/core
new file mode 100644
index 0000000..936af1d
Binary files /dev/null and b/errorReports/core differ
diff --git a/errorReports/stackTrace_test_c_crash 
b/errorReports/stackTrace_test_c_crash
new file mode 100644
index 0000000..3b968d4
--- /dev/null
+++ b/errorReports/stackTrace_test_c_crash
@@ -0,0 +1,3 @@
+#0  0x000000000040056f in tmpfun2 () at test_c_crash.c:21
+#1  0x0000000000400563 in tmpfun () at test_c_crash.c:16
+#2  0x000000000040054e in main (argc=1, argv=0x7ffd2395ebc8) at 
test_c_crash.c:10
diff --git a/intersting_script.py b/intersting_script.py
deleted file mode 100644
index d9a68d0..0000000
--- a/intersting_script.py
+++ /dev/null
@@ -1,168 +0,0 @@
-import argparse
-import subprocess
-import csv
-import os
-import sys
-import shutil
-import xlsxwriter
-import codecs
-import re
-reload(sys)
-sys.setdefaultencoding('utf8')
-global_keep_separate_output_files = False
-global_input_is_directory = False
-import xlrd
-import csv
-"""
-Simple function to convert xlsx to csv used by this program
-internally.
-"""
-def csv_from_excel():
-wb = xlrd.open_workbook('Crashes.xlsx')
-
2
-sh = wb.sheet_by_name('Sheet1')
-your_csv_file = open('Crashes.csv', 'wb')
-wr = csv.writer(your_csv_file, quoting=csv.QUOTE_ALL)
-for rownum in xrange(sh.nrows):
-wr.writerow(sh.row_values(rownum))
-your_csv_file.close()
-"""
-Simple function to verify args.
-- Checks that fuzzed program, input directory
-and output directories exists.
-"""
-def Validate_args(args):
-print ('\n')
-print ('Checking args')
-parser = argparse.ArgumentParser(description='Processes American
-Fuzzy Lop logs with GDB and creates a
-database of the results.')
-parser.add_argument('fuzzed', metavar='program',
-help='the fuzzed program')
-parser.add_argument('input', metavar='input',
-help='input directory of crashes')
-parser.add_argument('output', metavar='output',
-help='output directory to store analyzed crashes')
-parser.add_argument('keep', metavar='keep',
-help='if set to True keeps separate output files
-instead of deleting them on exit.')
-args = parser.parse_args()3
-#print(args)
-if (os.path.lexists(args.fuzzed) == False):
-print('The fuzzed program could not be found')
-sys.exit()
-if (os.path.lexists(args.input) == False):
-print('The input directory could not be found')
-sys.exit()
-if (os.path.isdir(args.input) == True):
-print(os.listdir(args.input))
-list = os.listdir(args.input)
-number_files = len(list)
-print number_files
-global_input_is_directory = True
-if (os.path.lexists(args.output) == False):
-print('The AFL\'s output file or directory could not be
-found')
-sys.exit()
-if (os.path.isdir(args.output) == False):
-print('The AFL\'s output is not directory')
-sys.exit()
-if(len(os.listdir(args.output)) >0):
-print('The AFL\'s output directory still has some files!
-Remove those.')4
-sys.exit()
-keep_files = args.keep[0]
-keep_files = keep_files.upper()
-if (keep_files == 'Y'):
-print('Leaving separate output files')
-global_keep_separate_output_files = True
-else:
-print('Not keeping separate output files')
-print ('\n')
-Validate_args(args)
-print ('\n')
-#Reads the number of files in the input directory
-list = os.listdir(args.input)
-number_files = len(list)
-number_files = int(number_files)
-path_to_input_files = os.getcwd() + '\\' + args.input + '\\'
-#Copies the given fuzzed program to the same directory as the input
-files in order to enable GDB to process it in batch run.
-shutil.copy(args.fuzzed, args.input + '/' + args.fuzzed)
-#Limits the files to be processed by GDB to the files that start with
-id_
-inputFilesList = [x for x in list if x.startswith('id:')]5
-#Processes every file in inputFilesList and runs GDB as a shell
-command for every input files
-amountOfFiles = len(inputFilesList)
-prevdir = os.getcwd()
-for i in range(amountOfFiles):
-ifile = open(args.input + '//run_gdb.bat', "wb+")
-os.chmod(args.input + '//run_gdb.bat', 0o777)
-filename=inputFilesList[i]
-#print(inputFilesList[i])
-parameters2 = "gdb --batch -ex 'r <"+filename+"'
-"+args.fuzzed+" -ex bt >output_"+filename+".txt"
-ifile.write(parameters2)
-ifile.close()
-print ('\nSending file: ' + filename + ' to GDB')
-os.chdir(args.input)
-subprocess.call('./run_gdb.bat', shell=True)
-os.chdir(prevdir)
-#print parameters2
-#Before writing to a database file every output file from GDB is
-moved to output directory
-source =os.getcwd() + '/' + args.input
-destination =
-os.getcwd() + '/' + args.output
-files = os.listdir(source)
-prevdir = os.getcwd()6
-os.chdir(args.input)
-for f in files:
-if (f.endswith("txt")):
-shutil.move(f, destination)
-os.chdir(prevdir)
-list = os.listdir(args.output)
-number_files = len(list)
-number_files = int(number_files)
-path_to_output_files = os.getcwd() + '/' + args.output + '/'
-# A workbook is created and a worksheet is added.
-workbook = xlsxwriter.Workbook('Crashes.xlsx')
-worksheet = workbook.add_worksheet()
-# Start from the first cell. Rows and columns are zero indexed.
-row = 0
-col = 0
-print ('\nStarting to write output files:')
-for i in range(len(list)):
-#ifile = open(path_to_output_files + list[i], "rb")
-ifile = codecs.open(path_to_output_files + list[i], "rb",
-encoding='utf8', errors = 'ignore')
-print ("\tWriting output file: " + list[i])
-reader = ifile.readlines()7
-#print reader
-x = i+1
-worksheet.write_string(row, col, 'crash dump #:' + str(x))
-row += 1
-for item in (reader):
-print item
-item = re.sub(r'\([^()]*\)', '', item)
-#worksheet.write_string(row, col, item)
-worksheet.write(row, col, item)
-row += 1
-worksheet.write_blank(row, col, item)
-workbook.close()
-numberOfInputFiles = str(len(inputFilesList))
-numberOfOutputFiles = str(len(os.listdir(path_to_output_files)))
-print ('\nProducing output files, please wait...')
-csv_from_excel()
-print "\n" + numberOfInputFiles +
-" input files processed and " +
-numberOfOutputFiles + " output files done!"
-#clean up
-if (keep_files == 'N'):
-print('\nRemoving separate output files')
-os.chdir(args.output)
-files = os.listdir(path_to_output_files)
-for f in files:
-if (f.endswith("txt") and len(f) >1):8
-os.remove(f)
\ No newline at end of file
diff --git a/stackTraceCParser.sh b/stackTraceCParser.sh
new file mode 100644
index 0000000..98d6700
--- /dev/null
+++ b/stackTraceCParser.sh
@@ -0,0 +1,49 @@
+#!/bin/bash
+
+isBinaryInDir=`ls | grep $1`;
+
+if [[ -n "$isBinaryInDir" ]]
+then
+    echo "chosen binary is : "$1;
+else
+    echo "couldnt find the binary in the current folder";
+    exit 0;
+fi
+
+binaryWithoutExtention=`echo $1 | cut -d '.' -f1`
+echo "saving result in : "$binaryWithoutExtention;
+
+ulimit -c 9999
+echo "--------------------------"
+./$1
+echo "--------------------------"
+mv ./core errorReports/
+
+res=`echo bt | gdb test_c_crash.exe errorReports/core | sed -e 's/(gdb) //' | 
grep \# | uniq > errorReports/stackTrace_$binaryWithoutExtention `
+RESULT=`cat errorReports/stackTrace_$binaryWithoutExtention | sed 's/.* in //' 
| cut -d "(" -f1`
+
+i=0
+echo $RESULT |
+while IFS= read -r line;
+do
+    array[i]=`echo $line`
+    i=i+1
+done
+
+echo "${array[0]}"
+echo "${array[1]}"
+echo "${array[2]}"
+#echo $res;
+#cat errorReports/stackTrace_$binaryWithoutExtention;
+
+
+
+
+
+
+
+
+
+
+#rm errorReports/core A LA FIN
+#rm stackTrace files depending on option ?
\ No newline at end of file
diff --git a/test_c_crash.c b/test_c_crash.c
new file mode 100644
index 0000000..55afa1c
--- /dev/null
+++ b/test_c_crash.c
@@ -0,0 +1,22 @@
+#include<stdio.h>
+#include<stdlib.h>
+
+void tmpfun();
+void tmpfun2();
+
+int main(int argc, char** argv)
+{
+       printf("crash test");
+       tmpfun();
+       return EXIT_SUCCESS;
+}
+
+void tmpfun()
+{
+    tmpfun2();
+}
+
+void tmpfun2()
+{
+    *(char *)0 = 0;
+}

-- 
To stop receiving notification emails like this one, please contact
address@hidden