Do not make duplicates fatal while packaging

This commit is contained in:
NTD 2018-04-14 10:01:26 -04:00 committed by Roy Tam
commit ba496fd6f0
2 changed files with 15 additions and 102 deletions

View file

@ -4,15 +4,8 @@
import sys
import hashlib
import re
from mozbuild.preprocessor import Preprocessor
from mozbuild.util import DefinesAction
from mozpack.packager.unpack import UnpackFinder
from mozpack.files import DeflatedFile
from collections import OrderedDict
from StringIO import StringIO
import argparse
import buildconfig
'''
Find files duplicated in a given packaged directory, independently of its
@ -20,116 +13,36 @@ package format.
'''
def normalize_osx_path(p):
'''
Strips the first 3 elements of an OSX app path
>>> normalize_osx_path('Nightly.app/foo/bar/baz')
'baz'
'''
bits = p.split('/')
if len(bits) > 3 and bits[0].endswith('.app'):
return '/'.join(bits[3:])
return p
def normalize_l10n_path(p):
'''
Normalizes localized paths to en-US
>>> normalize_l10n_path('chrome/es-ES/locale/branding/brand.properties')
'chrome/en-US/locale/branding/brand.properties'
>>> normalize_l10n_path('chrome/fr/locale/fr/browser/aboutHome.dtd')
'chrome/en-US/locale/en-US/browser/aboutHome.dtd'
'''
# Keep a trailing slash here! e.g. locales like 'br' can transform
# 'chrome/br/locale/branding/' into 'chrome/en-US/locale/en-USanding/'
p = re.sub(r'chrome/(\S+)/locale/\1/',
'chrome/en-US/locale/en-US/',
p)
p = re.sub(r'chrome/(\S+)/locale/',
'chrome/en-US/locale/',
p)
return p
def normalize_path(p):
return normalize_osx_path(normalize_l10n_path(p))
def find_dupes(source, allowed_dupes, bail=True):
allowed_dupes = set(allowed_dupes)
def find_dupes(source):
md5s = OrderedDict()
for p, f in UnpackFinder(source):
content = f.open().read()
m = hashlib.md5(content).digest()
if m not in md5s:
if isinstance(f, DeflatedFile):
compressed = f.file.compressed_size
else:
compressed = len(content)
md5s[m] = (len(content), compressed, [])
md5s[m][2].append(p)
if not m in md5s:
md5s[m] = (len(content), [])
md5s[m][1].append(p)
total = 0
total_compressed = 0
num_dupes = 0
unexpected_dupes = []
for m, (size, compressed, paths) in sorted(md5s.iteritems(),
key=lambda x: x[1][1]):
for m, (size, paths) in md5s.iteritems():
if len(paths) > 1:
print 'Duplicates %d bytes%s%s:' % (size,
' (%d compressed)' % compressed if compressed != size else '',
print 'Duplicates %d bytes%s:' % (size,
' (%d times)' % (len(paths) - 1) if len(paths) > 2 else '')
print ''.join(' %s\n' % p for p in paths)
total += (len(paths) - 1) * size
total_compressed += (len(paths) - 1) * compressed
num_dupes += 1
unexpected_dupes.extend([p for p in paths if normalize_path(p) not in allowed_dupes])
if num_dupes:
print "WARNING: Found %d duplicated files taking %d bytes (%s)" % \
(num_dupes, total,
'%d compressed' % total_compressed if total_compressed != total
else 'uncompressed')
if unexpected_dupes:
errortype = "ERROR" if bail else "WARNING"
print "%s: The following duplicated files are not allowed:" % errortype
print "\n".join(unexpected_dupes)
if bail:
sys.exit(1)
print "WARNING: Found %d duplicated files taking %d bytes" % \
(num_dupes, total) + " (uncompressed)"
def main():
parser = argparse.ArgumentParser(description='Find duplicate files in directory.')
parser.add_argument('--warning', '-w', action='store_true',
help='Only warn about duplicates, do not exit with an error')
parser.add_argument('--file', '-f', action='append', dest='dupes_files', default=[],
help='Add exceptions to the duplicate list from this file')
parser.add_argument('-D', action=DefinesAction)
parser.add_argument('-U', action='append', default=[])
parser.add_argument('directory',
help='The directory to check for duplicates in')
if len(sys.argv) != 2:
import os
print >>sys.stderr, "Usage: %s directory" % \
os.path.basename(sys.argv[0])
sys.exit(1)
args = parser.parse_args()
allowed_dupes = []
for filename in args.dupes_files:
pp = Preprocessor()
pp.context.update(buildconfig.defines)
if args.D:
pp.context.update(args.D)
for undefine in args.U:
if undefine in pp.context:
del pp.context[undefine]
pp.out = StringIO()
pp.do_filter('substitution')
pp.do_include(filename)
allowed_dupes.extend([line.partition('#')[0].rstrip()
for line in pp.out.getvalue().splitlines()])
find_dupes(args.directory, bail=not args.warning, allowed_dupes=allowed_dupes)
find_dupes(sys.argv[1])
if __name__ == "__main__":
main()