Add IDA scripts for normalizing the JSON

scripting/ida-extraction-scripts/normalize-all.sh

+#!/bin/bash
+
+if [ $# -eq 0 ]; then
+  echo "No arguments supplied"
+  exit 1
+fi
+
+arch=$1
+basepath=$(pwd)
+workdir=$basepath/workdir-$arch
+
+normalized_dir=$workdir/ida-json-normalized
+
+if [ ! -d $normalized_dir ]; then
+  mkdir $normalized_dir;
+fi
+
+cd $workdir/ida-json
+
+for filename in *
+do
+  echo "Normalizing input: $filename"
+  $basepath/scripting/ida-extraction-scripts/normalize.py $filename $normalized_dir/$filename
+done;

scripting/ida-extraction-scripts/normalize.py

+#!/usr/bin/env python3
+
+import argparse
+import json
+
+from rangeset import RangeSet
+
+empty = (RangeSet(0,0)).difference(RangeSet(-1,1))
+
+def compute_coverage(function):
+
+    # Compute the coverage of the function starting from the info contained in
+    # the basic blocks
+    coverage = empty
+    for x in function["basic_blocks"]:
+        coverage = coverage | RangeSet(int(x["start"], 16), int(x["end"], 16))
+
+    # Write each interval to the coverage field
+        coverage_list = []
+    for interval in coverage:
+        coverage_list.append({'start': hex(interval[0]), 'end': hex(interval[1])})
+    function["coverage"] = coverage_list
+
+def main():
+    parser = argparse.ArgumentParser(description='My nice tool.')
+    parser.add_argument('ida', metavar='IDAFILE', help='IDA created file')
+    parser.add_argument('normalized', metavar='OUTPUT', help='File in output')
+    args = parser.parse_args()
+
+    # Json output preparation
+    normalized = []
+
+    with open(args.ida, "r") as input_file, open(args.normalized, "w") as output_file:
+
+        # Load the original json
+        data = json.load(input_file)
+
+        # Iterate over all the functions
+        for function in data:
+
+            # Create the new function element with the basic infos
+            new_function = {'entry_point': "bb." + function['name'],
+                            'entry_point_address': function['startAddr'],
+                            'instruction_number' : function['instruction_number'],
+                            'basic_blocks': []}
+
+            # Add the basic blocks
+            for basic_block in function['basicBlocks']:
+                new_function['basic_blocks'].append({'start': basic_block['startAddr'],
+                                                     'end': basic_block['endAddr']})
+
+            # Calculate the coverage using the info contained in the basic
+            # blocks, since the end address produced by IDA does not take into
+            # account gaps
+            compute_coverage(new_function)
+
+            # Append the newly created function to the output json
+            normalized.append(new_function)
+
+        # Write the json on the output file
+        json.dump(normalized, output_file, indent=2)
+
+if __name__ == "__main__":
+    main()