Merge #9373: Linearize script update (hash byte reversal and Python 3 support)

3c8f63b Make linearize scripts Python 3-compatible. (Doug)
d5aa198 Allow linearization scripts to support hash byte reversal (Doug)

4 files changed, 97 insertions, 37 deletions

diff --git a/contrib/linearize/README.md b/contrib/linearize/README.md
index 06f278f3b3..adc9a559cc 100644
--- a/contrib/linearize/README.md
+++ b/contrib/linearize/README.md
@@ -1,33 +1,48 @@
 # Linearize
-Construct a linear, no-fork, best version of the blockchain.
+Construct a linear, no-fork, best version of the Bitcoin blockchain. The scripts
+run using Python 3 but are compatible with Python 2.
 
 ## Step 1: Download hash list
 
     $ ./linearize-hashes.py linearize.cfg > hashlist.txt
 
 Required configuration file settings for linearize-hashes:
-* RPC: rpcuser, rpcpassword
+* RPC: `rpcuser`, `rpcpassword`
 
 Optional config file setting for linearize-hashes:
-* RPC: host, port
-* Block chain: min_height, max_height
+* RPC: `host`  (Default: `127.0.0.1`)
+* RPC: `port`  (Default: `8332`)
+* Blockchain: `min_height`, `max_height`
+* `rev_hash_bytes`: If true, the written block hash list will be
+byte-reversed. (In other words, the hash returned by getblockhash will have its
+bytes reversed.) False by default. Intended for generation of
+standalone hash lists but safe to use with linearize-data.py, which will output
+the same data no matter which byte format is chosen.
+
+The `linearize-hashes` script requires a connection, local or remote, to a
+JSON-RPC server. Running `bitcoind` or `bitcoin-qt -server` will be sufficient.
 
 ## Step 2: Copy local block data
 
     $ ./linearize-data.py linearize.cfg
 
 Required configuration file settings:
-* "input": bitcoind blocks/ directory containing blkNNNNN.dat
-* "hashlist": text file containing list of block hashes, linearized-hashes.py
-output.
-* "output_file": bootstrap.dat
+* `output_file`: The file that will contain the final blockchain.
       or
-* "output": output directory for linearized blocks/blkNNNNN.dat output
+* `output`: Output directory for linearized `blocks/blkNNNNN.dat` output.
 
 Optional config file setting for linearize-data:
-* "netmagic": network magic number
-* "max_out_sz": maximum output file size (default `1000*1000*1000`)
-* "split_timestamp": Split files when a new month is first seen, in addition to
-reaching a maximum file size.
-* "file_timestamp": Set each file's last-modified time to that of the
-most recent block in that file.
+* `file_timestamp`: Set each file's last-modified time to that of the most
+recent block in that file.
+* `genesis`: The hash of the genesis block in the blockchain.
+* `input`: bitcoind blocks/ directory containing blkNNNNN.dat
+* `hashlist`: text file containing list of block hashes created by
+linearize-hashes.py.
+* `max_out_sz`: Maximum size for files created by the `output_file` option.
+(Default: `1000*1000*1000 bytes`)
+* `netmagic`: Network magic number.
+* `rev_hash_bytes`: If true, the block hash list written by linearize-hashes.py
+will be byte-reversed when read by linearize-data.py. See the linearize-hashes
+entry for more information.
+* `split_timestamp`: Split blockchain files when a new month is first seen, in
+addition to reaching a maximum file size (`max_out_sz`).
diff --git a/contrib/linearize/example-linearize.cfg b/contrib/linearize/example-linearize.cfg
index 38da02e66c..cccdd79213 100644
--- a/contrib/linearize/example-linearize.cfg
+++ b/contrib/linearize/example-linearize.cfg
@@ -23,7 +23,9 @@ input=/home/example/.bitcoin/blocks
 
 output_file=/home/example/Downloads/bootstrap.dat
 hashlist=hashlist.txt
-split_year=1
 
 # Maxmimum size in bytes of out-of-order blocks cache in memory
 out_of_order_cache_sz = 100000000
+
+# Do we want the reverse the hash bytes coming from getblockhash?
+rev_hash_bytes = False
diff --git a/contrib/linearize/linearize-data.py b/contrib/linearize/linearize-data.py
index d0417748fc..043bf2e814 100755
--- a/contrib/linearize/linearize-data.py
+++ b/contrib/linearize/linearize-data.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 #
 # linearize-data.py: Construct a linear, no-fork version of the chain.
 #
@@ -8,23 +8,33 @@
 #
 
 from __future__ import print_function, division
+try: # Python 3
+    import http.client as httplib
+except ImportError: # Python 2
+    import httplib
 import json
 import struct
 import re
 import os
 import os.path
 import base64
-import httplib
 import sys
 import hashlib
 import datetime
 import time
 from collections import namedtuple
+from binascii import hexlify, unhexlify
 
 settings = {}
 
+##### Switch endian-ness #####
+def hex_switchEndian(s):
+	""" Switches the endianness of a hex string (in pairs of hex chars) """
+	pairList = [s[i:i+2].encode() for i in range(0, len(s), 2)]
+	return b''.join(pairList[::-1]).decode()
+
 def uint32(x):
-	return x & 0xffffffffL
+	return x & 0xffffffff
 
 def bytereverse(x):
 	return uint32(( ((x) << 24) | (((x) << 8) & 0x00ff0000) |
@@ -35,14 +45,14 @@ def bufreverse(in_buf):
 	for i in range(0, len(in_buf), 4):
 		word = struct.unpack('@I', in_buf[i:i+4])[0]
 		out_words.append(struct.pack('@I', bytereverse(word)))
-	return ''.join(out_words)
+	return b''.join(out_words)
 
 def wordreverse(in_buf):
 	out_words = []
 	for i in range(0, len(in_buf), 4):
 		out_words.append(in_buf[i:i+4])
 	out_words.reverse()
-	return ''.join(out_words)
+	return b''.join(out_words)
 
 def calc_hdr_hash(blk_hdr):
 	hash1 = hashlib.sha256()
@@ -59,7 +69,7 @@ def calc_hash_str(blk_hdr):
 	hash = calc_hdr_hash(blk_hdr)
 	hash = bufreverse(hash)
 	hash = wordreverse(hash)
-	hash_str = hash.encode('hex')
+	hash_str = hexlify(hash).decode('utf-8')
 	return hash_str
 
 def get_blk_dt(blk_hdr):
@@ -69,17 +79,21 @@ def get_blk_dt(blk_hdr):
 	dt_ym = datetime.datetime(dt.year, dt.month, 1)
 	return (dt_ym, nTime)
 
+# When getting the list of block hashes, undo any byte reversals.
 def get_block_hashes(settings):
 	blkindex = []
 	f = open(settings['hashlist'], "r")
 	for line in f:
 		line = line.rstrip()
+		if settings['rev_hash_bytes'] == 'true':
+			line = hex_switchEndian(line)
 		blkindex.append(line)
 
 	print("Read " + str(len(blkindex)) + " hashes")
 
 	return blkindex
 
+# The block map shouldn't give or receive byte-reversed hashes.
 def mkblockmap(blkindex):
 	blkmap = {}
 	for height,hash in enumerate(blkindex):
@@ -207,7 +221,7 @@ class BlockDataCopier:
 
 			inMagic = inhdr[:4]
 			if (inMagic != self.settings['netmagic']):
-				print("Invalid magic: " + inMagic.encode('hex'))
+				print("Invalid magic: " + hexlify(inMagic).decode('utf-8'))
 				return
 			inLenLE = inhdr[4:]
 			su = struct.unpack("<I", inLenLE)
@@ -265,6 +279,12 @@ if __name__ == '__main__':
 		settings[m.group(1)] = m.group(2)
 	f.close()
 
+	# Force hash byte format setting to be lowercase to make comparisons easier.
+	# Also place upfront in case any settings need to know about it.
+	if 'rev_hash_bytes' not in settings:
+		settings['rev_hash_bytes'] = 'false'
+	settings['rev_hash_bytes'] = settings['rev_hash_bytes'].lower()
+
 	if 'netmagic' not in settings:
 		settings['netmagic'] = 'f9beb4d9'
 	if 'genesis' not in settings:
@@ -278,14 +298,14 @@ if __name__ == '__main__':
 	if 'split_timestamp' not in settings:
 		settings['split_timestamp'] = 0
 	if 'max_out_sz' not in settings:
-		settings['max_out_sz'] = 1000L * 1000 * 1000
+		settings['max_out_sz'] = 1000 * 1000 * 1000
 	if 'out_of_order_cache_sz' not in settings:
 		settings['out_of_order_cache_sz'] = 100 * 1000 * 1000
 
-	settings['max_out_sz'] = long(settings['max_out_sz'])
+	settings['max_out_sz'] = int(settings['max_out_sz'])
 	settings['split_timestamp'] = int(settings['split_timestamp'])
 	settings['file_timestamp'] = int(settings['file_timestamp'])
-	settings['netmagic'] = settings['netmagic'].decode('hex')
+	settings['netmagic'] = unhexlify(settings['netmagic'].encode('utf-8'))
 	settings['out_of_order_cache_sz'] = int(settings['out_of_order_cache_sz'])
 
 	if 'output_file' not in settings and 'output' not in settings:
@@ -295,9 +315,8 @@ if __name__ == '__main__':
 	blkindex = get_block_hashes(settings)
 	blkmap = mkblockmap(blkindex)
 
+	# Block hash map won't be byte-reversed. Neither should the genesis hash.
 	if not settings['genesis'] in blkmap:
 		print("Genesis block not found in hashlist")
 	else:
 		BlockDataCopier(settings, blkindex, blkmap).run()
-
-
diff --git a/contrib/linearize/linearize-hashes.py b/contrib/linearize/linearize-hashes.py
index ba8d6f8a04..f749da5396 100755
--- a/contrib/linearize/linearize-hashes.py
+++ b/contrib/linearize/linearize-hashes.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 #
 # linearize-hashes.py:  List blocks in a linear, no-fork version of the chain.
 #
@@ -8,32 +8,47 @@
 #
 
 from __future__ import print_function
+try: # Python 3
+    import http.client as httplib
+except ImportError: # Python 2
+    import httplib
 import json
 import struct
 import re
 import base64
-import httplib
 import sys
 
 settings = {}
 
+##### Switch endian-ness #####
+def hex_switchEndian(s):
+	""" Switches the endianness of a hex string (in pairs of hex chars) """
+	pairList = [s[i:i+2].encode() for i in range(0, len(s), 2)]
+	return b''.join(pairList[::-1]).decode()
+
 class BitcoinRPC:
 	def __init__(self, host, port, username, password):
 		authpair = "%s:%s" % (username, password)
-		self.authhdr = "Basic %s" % (base64.b64encode(authpair))
-		self.conn = httplib.HTTPConnection(host, port, False, 30)
+		authpair = authpair.encode('utf-8')
+		self.authhdr = b"Basic " + base64.b64encode(authpair)
+		self.conn = httplib.HTTPConnection(host, port=port, timeout=30)
 
 	def execute(self, obj):
-		self.conn.request('POST', '/', json.dumps(obj),
-			{ 'Authorization' : self.authhdr,
-			  'Content-type' : 'application/json' })
+		try:
+			self.conn.request('POST', '/', json.dumps(obj),
+				{ 'Authorization' : self.authhdr,
+				  'Content-type' : 'application/json' })
+		except ConnectionRefusedError:
+			print('RPC connection refused. Check RPC settings and the server status.',
+			      file=sys.stderr)
+			return None
 
 		resp = self.conn.getresponse()
 		if resp is None:
 			print("JSON-RPC: no response", file=sys.stderr)
 			return None
 
-		body = resp.read()
+		body = resp.read().decode('utf-8')
 		resp_obj = json.loads(body)
 		return resp_obj
 
@@ -64,12 +79,17 @@ def get_block_hashes(settings, max_blocks_per_call=10000):
 			batch.append(rpc.build_request(x, 'getblockhash', [height + x]))
 
 		reply = rpc.execute(batch)
+		if reply is None:
+			print('Cannot continue. Program will halt.')
+			return None
 
 		for x,resp_obj in enumerate(reply):
 			if rpc.response_is_error(resp_obj):
 				print('JSON-RPC: error at height', height+x, ': ', resp_obj['error'], file=sys.stderr)
 				exit(1)
 			assert(resp_obj['id'] == x) # assume replies are in-sequence
+			if settings['rev_hash_bytes'] == 'true':
+				resp_obj['result'] = hex_switchEndian(resp_obj['result'])
 			print(resp_obj['result'])
 
 		height += num_blocks
@@ -101,6 +121,8 @@ if __name__ == '__main__':
 		settings['min_height'] = 0
 	if 'max_height' not in settings:
 		settings['max_height'] = 313000
+	if 'rev_hash_bytes' not in settings:
+		settings['rev_hash_bytes'] = 'false'
 	if 'rpcuser' not in settings or 'rpcpassword' not in settings:
 		print("Missing username and/or password in cfg file", file=stderr)
 		sys.exit(1)
@@ -109,5 +131,7 @@ if __name__ == '__main__':
 	settings['min_height'] = int(settings['min_height'])
 	settings['max_height'] = int(settings['max_height'])
 
-	get_block_hashes(settings)
+	# Force hash byte format setting to be lowercase to make comparisons easier.
+	settings['rev_hash_bytes'] = settings['rev_hash_bytes'].lower()
 
+	get_block_hashes(settings)