From 3d77ba93209652befde4e9d7e811d039171a081a Mon Sep 17 00:00:00 2001
From: Tobias Kessels <tobias.kessels@certbw.de>
Date: Wed, 19 Dec 2018 11:01:51 +0100
Subject: [PATCH] first version of scatterhash

---
 scatterhash.py | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)
 create mode 100755 scatterhash.py

diff --git a/scatterhash.py b/scatterhash.py
new file mode 100755
index 0000000..e95a577
--- /dev/null
+++ b/scatterhash.py
@@ -0,0 +1,49 @@
+#!/usr/bin/python3
+import sys
+import hashlib
+import os
+import numpy as np
+
+def even_select(N, M):
+    if M > N/2:
+        cut = np.zeros(N, dtype=int)
+        q, r = divmod(N, N-M)
+        indices = [q*i + min(i, r) for i in range(N-M)]
+        cut[indices] = True
+    else:
+        cut = np.ones(N, dtype=int)
+        q, r = divmod(N, M)
+        indices = [q*i + min(i, r) for i in range(M)]
+        cut[indices] = False
+
+    return cut
+
+def get_offsets(chunksize, spread):
+    selection=even_select(100,spread)
+    for i in range(0,100):
+        if selection[i]==0:
+            offset=int(chunksize*i)
+            yield offset
+
+def get_blocks(filename,spread,blocksize):
+    filesize=os.path.getsize(filename)
+    chunksize=filesize/100
+    with open(filename,'rb') as infile:
+        for of in get_offsets(chunksize,spread):
+            infile.seek(of)
+            tohashsize=chunksize
+            while tohashsize > 0:
+                yield infile.read(blocksize)
+                tohashsize-=h.block_size
+
+
+
+hashalgo="md5"
+filename=sys.argv[2]
+spread=int(sys.argv[1]) #percentage of hash
+h=hashlib.new(hashalgo)
+
+blocksize=h.block_size*4
+for block in get_blocks(filename,spread,blocksize):
+    h.update(block)
+print(h.hexdigest())