Skip to content

Commit

Permalink
contrib/linearize: split output files based on new-timestamp-year or …
Browse files Browse the repository at this point in the history
…max-file-size
  • Loading branch information
Jeff Garzik committed Aug 24, 2014
1 parent 476eb7e commit b4a72a7
Show file tree
Hide file tree
Showing 3 changed files with 39 additions and 9 deletions.
3 changes: 3 additions & 0 deletions contrib/linearize/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -26,4 +26,7 @@ output.

Optional config file setting for linearize-data:
* "netmagic": network magic number
* "max_out_sz": maximum output file size (default 1000*1000*1000)
* "split_year": Split files when a new year is first seen, in addition to
reaching a maximum file size.

1 change: 1 addition & 0 deletions contrib/linearize/example-linearize.cfg
Original file line number Diff line number Diff line change
Expand Up @@ -13,4 +13,5 @@ netmagic=f9beb4d9
input=/home/example/.bitcoin/blocks
output_file=/home/example/Downloads/bootstrap.dat
hashlist=hashlist.txt
split_year=1

44 changes: 35 additions & 9 deletions contrib/linearize/linearize-data.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,8 +14,7 @@
import httplib
import sys
import hashlib

MAX_OUT_SZ = 128 * 1024 * 1024
import datetime

settings = {}

Expand All @@ -41,9 +40,7 @@ def wordreverse(in_buf):
out_words.reverse()
return ''.join(out_words)

def calc_hdr_hash(rawblock):
blk_hdr = rawblock[:80]

def calc_hdr_hash(blk_hdr):
hash1 = hashlib.sha256()
hash1.update(blk_hdr)
hash1_o = hash1.digest()
Expand All @@ -54,13 +51,18 @@ def calc_hdr_hash(rawblock):

return hash2_o

def calc_hash_str(rawblock):
hash = calc_hdr_hash(rawblock)
def calc_hash_str(blk_hdr):
hash = calc_hdr_hash(blk_hdr)
hash = bufreverse(hash)
hash = wordreverse(hash)
hash_str = hash.encode('hex')
return hash_str

def get_blk_year(blk_hdr):
members = struct.unpack("<I", blk_hdr[68:68+4])
dt = datetime.datetime.fromtimestamp(members[0])
return dt.year

def get_block_hashes(settings):
blkindex = []
f = open(settings['hashlist'], "r")
Expand All @@ -86,9 +88,14 @@ def copydata(settings, blkindex, blkset):
outF = None
blkCount = 0

lastYear = 0
splitYear = False
fileOutput = True
maxOutSz = settings['max_out_sz']
if 'output' in settings:
fileOutput = False
if settings['split_year'] != 0:
splitYear = True

while True:
if not inF:
Expand All @@ -111,17 +118,30 @@ def copydata(settings, blkindex, blkset):
su = struct.unpack("<I", inLenLE)
inLen = su[0]
rawblock = inF.read(inLen)
blk_hdr = rawblock[:80]

hash_str = calc_hash_str(rawblock)
hash_str = calc_hash_str(blk_hdr)
if not hash_str in blkset:
print("Skipping unknown block " + hash_str)
continue

if not fileOutput and ((outsz + inLen) > MAX_OUT_SZ):
if not fileOutput and ((outsz + inLen) > maxOutSz):
outF.close()
outF = None
outFn = outFn + 1
outsz = 0

if splitYear:
blkYear = get_blk_year(blk_hdr)
if blkYear > lastYear:
print("New year " + str(blkYear) + " @ " + hash_str)
lastYear = blkYear
if outF:
outF.close()
outF = None
outFn = outFn + 1
outsz = 0

if not outF:
if fileOutput:
fname = settings['output_file']
Expand Down Expand Up @@ -164,7 +184,13 @@ def copydata(settings, blkindex, blkset):
settings['input'] = 'input'
if 'hashlist' not in settings:
settings['hashlist'] = 'hashlist.txt'
if 'split_year' not in settings:
settings['split_year'] = 0
if 'max_out_sz' not in settings:
settings['max_out_sz'] = 1000L * 1000 * 1000

settings['max_out_sz'] = long(settings['max_out_sz'])
settings['split_year'] = int(settings['split_year'])
settings['netmagic'] = settings['netmagic'].decode('hex')

if 'output_file' not in settings and 'output' not in settings:
Expand Down

0 comments on commit b4a72a7

Please sign in to comment.