Add treehash512.py

Add script for quickly getting the treehash of a certain commit.

Author: laanwj

README.md

@@ -47,3 +47,17 @@ For example:
 unittest-statistics.py src/test/test_bitcoin wallet_tests
 ```
 
+treehash512
+--------------
+
+This script will show the SHA512 tree has for a certain commit, or HEAD
+by default.
+
+Usage:
+
+```bash
+treehash512.py [<commithash>]
+```
+
+This should match the Tree-SHA512 commit metadata field added by
+github-merge.

treehash512.py

@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+# Copyright (c) 2017 The Bitcoin Core developers
+# Distributed under the MIT software license, see the accompanying
+# file COPYING or http://www.opensource.org/licenses/mit-license.php.
+'''
+This script will show the SHA512 tree has for a certain commit, or HEAD
+by default.
+
+Usage:
+
+    treehash512.py [<commithash>]
+'''
+import os
+from sys import stdin,stdout,stderr
+import sys
+import argparse
+import hashlib
+import subprocess
+
+# External tools (can be overridden using environment)
+GIT = os.getenv('GIT','git')
+
+def tree_sha512sum(commit='HEAD'):
+    # request metadata for entire tree, recursively
+    files = []
+    blob_by_name = {}
+    for line in subprocess.check_output([GIT, 'ls-tree', '--full-tree', '-r', commit]).splitlines():
+        name_sep = line.index(b'\t')
+        metadata = line[:name_sep].split() # perms, 'blob', blobid
+        assert(metadata[1] == b'blob')
+        name = line[name_sep+1:]
+        files.append(name)
+        blob_by_name[name] = metadata[2]
+
+    files.sort()
+    # open connection to git-cat-file in batch mode to request data for all blobs
+    # this is much faster than launching it per file
+    p = subprocess.Popen([GIT, 'cat-file', '--batch'], stdout=subprocess.PIPE, stdin=subprocess.PIPE)
+    overall = hashlib.sha512()
+    for f in files:
+        blob = blob_by_name[f]
+        # request blob
+        p.stdin.write(blob + b'\n')
+        p.stdin.flush()
+        # read header: blob, "blob", size
+        reply = p.stdout.readline().split()
+        assert(reply[0] == blob and reply[1] == b'blob')
+        size = int(reply[2])
+        # hash the blob data
+        intern = hashlib.sha512()
+        ptr = 0
+        while ptr < size:
+            bs = min(65536, size - ptr)
+            piece = p.stdout.read(bs)
+            if len(piece) == bs:
+                intern.update(piece)
+            else:
+                raise IOError('Premature EOF reading git cat-file output')
+            ptr += bs
+        dig = intern.hexdigest()
+        assert(p.stdout.read(1) == b'\n') # ignore LF that follows blob data
+        # update overall hash with file hash
+        overall.update(dig.encode("utf-8"))
+        overall.update("  ".encode("utf-8"))
+        overall.update(f)
+        overall.update("\n".encode("utf-8"))
+    p.stdin.close()
+    if p.wait():
+        raise IOError('Non-zero return value executing git cat-file')
+    return overall.hexdigest()
+
+def main():
+    if len(sys.argv)>1:
+        commit = sys.argv[1]
+    else:
+        commit = 'HEAD'
+    print(tree_sha512sum(commit))
+
+if __name__ == '__main__':
+    main()
+