from zlib import decompressobj, MAX_WBITS
from gzip import FCOMMENT, FEXTRA, FHCRC, FNAME, FTEXT
from urlparse import urlparse
-from binascii import a2b_hex
import os
from twisted.python import log, filepath
class ProxyFileStream(stream.SimpleStream):
"""Saves a stream to a file while providing a new stream."""
- def __init__(self, stream, outFile, decompress = None, decFile = None):
+ def __init__(self, stream, outFile, hash, decompress = None, decFile = None):
"""Initializes the proxy.
@type stream: C{twisted.web2.stream.IByteStream}
@param stream: the input stream to read from
@type outFile: C{twisted.python.filepath.FilePath}
@param outFile: the file to write to
+ @type hash: L{Hash.HashObject}
+ @param hash: the hash object to use for the file
@type decompress: C{string}
@param decompress: also decompress the file as this type
(currently only '.gz' and '.bz2' are supported)
"""
self.stream = stream
self.outFile = outFile.open('w')
+ self.hash = hash
+ self.hash.new()
self.gzfile = None
self.bz2file = None
if decompress == ".gz":
"""Close the output file."""
if not self.outFile.closed:
self.outFile.close()
+ self.hash.digest()
if self.gzfile:
data_dec = self.gzdec.flush()
self.gzfile.write(data_dec)
self.bz2file.close()
self.bz2file = None
- self.doneDefer.callback(1)
+ self.doneDefer.callback(self.hash)
def read(self):
"""Read some data from the stream."""
return data
self.outFile.write(data)
+ self.hash.update(data)
if self.gzfile:
if self.gzheader:
self.gzheader = False
def findHash(self, url):
site, baseDir, path = self.extractPath(url)
if site in self.apt_caches and baseDir in self.apt_caches[site]:
- d = self.apt_caches[site][baseDir].findHash(path)
- d.addCallback(self.translateHash)
- return d
+ return self.apt_caches[site][baseDir].findHash(path)
d = defer.Deferred()
d.errback(MirrorError("Site Not Found"))
return d
- def translateHash(self, (hash, size)):
- """Translate a hash from apt's hex encoding to a string."""
- if hash:
- hash = a2b_hex(hash)
- return (hash, size)
-
- def save_file(self, response, hash, size, url):
+ def save_file(self, response, hash, url):
"""Save a downloaded file to the cache and stream it."""
log.msg('Returning file: %s' % url)
else:
ext = None
decFile = None
-
+
orig_stream = response.stream
- response.stream = ProxyFileStream(orig_stream, destFile, ext, decFile)
+ response.stream = ProxyFileStream(orig_stream, destFile, hash, ext, decFile)
response.stream.doneDefer.addCallback(self.save_complete, url, destFile,
response.headers.getHeader('Last-Modified'),
ext, decFile)
response.stream.doneDefer.addErrback(self.save_error, url)
return response
- def save_complete(self, result, url, destFile, modtime = None, ext = None, decFile = None):
+ def save_complete(self, hash, url, destFile, modtime = None, ext = None, decFile = None):
"""Update the modification time and AptPackages."""
if modtime:
os.utime(destFile.path, (modtime, modtime))
if ext:
os.utime(decFile.path, (modtime, modtime))
-
- self.updatedFile(url, destFile.path)
- if ext:
- self.updatedFile(url[:-len(ext)], decFile.path)
+
+ result = hash.verify()
+ if result or result is None:
+ if result:
+ log.msg('Hashes match: %s' % url)
+ else:
+ log.msg('Hashed file to %s: %s' % (hash.hexdigest(), url))
+
+ self.updatedFile(url, destFile.path)
+ if ext:
+ self.updatedFile(url[:-len(ext)], decFile.path)
+ else:
+ log.msg("Hashes don't match %s != %s: %s" % (hash.hexexpected(), hash.hexdigest(), url))
def save_error(self, failure, url):
"""An error has occurred in downloadign or saving the file."""
self.failUnless(path == "/dists/unstable/Release", "no match: %s" % path)
def verifyHash(self, found_hash, path, true_hash):
- self.failUnless(found_hash[0] == true_hash,
- "%s hashes don't match: %s != %s" % (path, found_hash[0], true_hash))
+ self.failUnless(found_hash.hexexpected() == true_hash,
+ "%s hashes don't match: %s != %s" % (path, found_hash.hexexpected(), true_hash))
def test_findHash(self):
self.packagesFile = os.popen('ls -Sr /var/lib/apt/lists/ | grep -E "_main_.*Packages$" | tail -n 1').read().rstrip('\n')
idx_path = 'http://' + self.releaseFile.replace('_','/')[:-7] + 'main/binary-i386/Packages.bz2'
d = self.client.findHash(idx_path)
- d.addCallback(self.verifyHash, idx_path, a2b_hex(idx_hash))
+ d.addCallback(self.verifyHash, idx_path, idx_hash)
pkg_hash = os.popen('grep -A 30 -E "^Package: dpkg$" ' +
'/var/lib/apt/lists/' + self.packagesFile +
' | cut -d\ -f 2').read().rstrip('\n')
d = self.client.findHash(pkg_path)
- d.addCallback(self.verifyHash, pkg_path, a2b_hex(pkg_hash))
+ d.addCallback(self.verifyHash, pkg_path, pkg_hash)
src_dir = os.popen('grep -A 30 -E "^Package: dpkg$" ' +
'/var/lib/apt/lists/' + self.sourcesFile +
for i in range(len(src_hashes)):
src_path = 'http://' + self.releaseFile[:self.releaseFile.find('_dists_')+1].replace('_','/') + src_dir + '/' + src_paths[i]
d = self.client.findHash(src_path)
- d.addCallback(self.verifyHash, src_path, a2b_hex(src_hashes[i]))
+ d.addCallback(self.verifyHash, src_path, src_hashes[i])
idx_hash = os.popen('grep -A 3000 -E "^SHA1:" ' +
'/var/lib/apt/lists/' + self.releaseFile +
idx_path = 'http://' + self.releaseFile.replace('_','/')[:-7] + 'main/source/Sources.bz2'
d = self.client.findHash(idx_path)
- d.addCallback(self.verifyHash, idx_path, a2b_hex(idx_hash))
+ d.addCallback(self.verifyHash, idx_path, idx_hash)
d.addBoth(lastDefer.callback)
return lastDefer