mirror of
https://github.com/nodejs/node.git
synced 2024-11-24 20:29:23 +01:00
1e761654d3
Refs: https://en.wikipedia.org/wiki/Byte#Multiple-byte_units PR-URL: https://github.com/nodejs/node/pull/42587 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Darshan Sen <raisinten@gmail.com> Reviewed-By: Paolo Insogna <paolo@cowtech.it> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Mestery <mestery@protonmail.com>
34 lines
900 B
Python
34 lines
900 B
Python
#!/usr/bin/env python
|
|
|
|
import json
|
|
import struct
|
|
import sys
|
|
import zlib
|
|
|
|
try:
|
|
xrange # Python 2
|
|
PY2 = True
|
|
except NameError:
|
|
PY2 = False
|
|
xrange = range # Python 3
|
|
|
|
|
|
if __name__ == '__main__':
|
|
with open(sys.argv[1]) as fp:
|
|
obj = json.load(fp)
|
|
text = json.dumps(obj, separators=(',', ':')).encode('utf-8')
|
|
data = zlib.compress(text, zlib.Z_BEST_COMPRESSION)
|
|
|
|
# To make decompression a little easier, we prepend the compressed data
|
|
# with the size of the uncompressed data as a 24 bits BE unsigned integer.
|
|
assert len(text) < 1 << 24, 'Uncompressed JSON must be < 16 MiB.'
|
|
data = struct.pack('>I', len(text))[1:4] + data
|
|
|
|
step = 20
|
|
slices = (data[i:i+step] for i in xrange(0, len(data), step))
|
|
slices = [','.join(str(ord(c) if PY2 else c) for c in s) for s in slices]
|
|
text = ',\n'.join(slices)
|
|
|
|
with open(sys.argv[2], 'w') as fp:
|
|
fp.write(text)
|