Magic-Mirror/gecko-dev
1
0
Fork 0
mirror of https://github.com/mozilla/gecko-dev.git synced 2025-02-28 05:10:49 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Bug 1303210 - Show compressed size in find_dupes.py report, and sort its output. r=gps

Browse Source
This commit is contained in:
Mike Hommey 2016-09-16 08:36:55 +09:00
parent 97f635df6b
commit f3effba503
1 changed files with 18 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
toolkit/mozapps/installer/find-dupes.py
View File

@ -5,6 +5,7 @@
import sys import sys
import hashlib import hashlib
from mozpack.packager.unpack import UnpackFinder from mozpack.packager.unpack import UnpackFinder
from mozpack.files import DeflatedFile
from collections import OrderedDict from collections import OrderedDict
''' '''
@ -18,21 +19,31 @@ def find_dupes(source):
for p, f in UnpackFinder(source): for p, f in UnpackFinder(source):
content = f.open().read() content = f.open().read()
m = hashlib.md5(content).digest() m = hashlib.md5(content).digest()
if not m in md5s: if m not in md5s:
md5s[m] = (len(content), []) if isinstance(f, DeflatedFile):
md5s[m][1].append(p) compressed = f.file.compressed_size
else:
compressed = len(content)
md5s[m] = (len(content), compressed, [])
md5s[m][2].append(p)
total = 0 total = 0
total_compressed = 0
num_dupes = 0 num_dupes = 0
for m, (size, paths) in md5s.iteritems(): for m, (size, compressed, paths) in sorted(md5s.iteritems(),
key=lambda x: x[1][1]):
if len(paths) > 1: if len(paths) > 1:
print 'Duplicates %d bytes%s:' % (size, print 'Duplicates %d bytes%s%s:' % (size,
' (%d compressed)' % compressed if compressed != size else '',
' (%d times)' % (len(paths) - 1) if len(paths) > 2 else '') ' (%d times)' % (len(paths) - 1) if len(paths) > 2 else '')
print ''.join(' %s\n' % p for p in paths) print ''.join(' %s\n' % p for p in paths)
total += (len(paths) - 1) * size total += (len(paths) - 1) * size
total_compressed += (len(paths) - 1) * compressed
num_dupes += 1 num_dupes += 1
if num_dupes: if num_dupes:
print "WARNING: Found %d duplicated files taking %d bytes" % \ print "WARNING: Found %d duplicated files taking %d bytes (%s)" % \
(num_dupes, total) + " (uncompressed)" (num_dupes, total,
'%d compressed' % total_compressed if total_compressed != total
else 'uncompressed')
def main(): def main():