# Copyright 2008 The Tor Project, Inc. See LICENSE for licensing information. import httplib import logging import os import Queue import random import sys import threading import time import traceback import urllib2 import thandy.util import thandy.socksurls import thandy.checkJson class BadCompoundData(thandy.DownloadError): """DOCDOC""" pass class DownloadManager: """Class to track a set of downloads and pass them out to worker threads. """ def __init__(self, n_threads=2): # Prevents concurrent modification to downloads and haveDownloaded self._lock = threading.RLock() # Map from resource relPath to job. self.downloads = {} # Map from resource relPath from True to objects that we have # managed to dowload. self.haveDownloaded = {} # Work queue of DownloadJobs that we intend to process once a thread # is free. self.downloadQueue = Queue.Queue() # DOCDOC self.resultQueue = Queue.Queue() # List of worker threads. self.threads = [ threading.Thread(target=self._thread, args=[idx]) for idx in xrange(n_threads) ] # Condition that gets triggered whenever a thread is finished doing # something. self.done = threading.Condition() for t in self.threads: t.setDaemon(True) # DOCDOC self.statusLog = DownloadStatusLog() #DOCDOC self._raiseMe = None def start(self): """Start all of this download manager's worker threads.""" for t in self.threads: t.start() def isCurrentlyDownloading(self, relPath): """Return true iff this download manager is currently downloading some copy of the resource at relPath.""" self._lock.acquire() try: return self.downloads.has_key(relPath) finally: self._lock.release() def isRedundant(self, relPath): """Return true iff we are currently downloading, or have downloaded, the resource at relPath.""" self._lock.acquire() try: return (self.downloads.has_key(relPath) or self.haveDownloaded.has_key(relPath)) finally: self._lock.release() def finished(self): """Return true iff we have no active or pending jobs.""" self._lock.acquire() try: return self.downloadQueue.empty() and len(self.downloads) == 0 finally: self._lock.release() def wait(self): """Pause until we have no active or pending jobs.""" while not self.finished(): self.done.acquire() self.done.wait() self.done.release() if self._raiseMe: raise self._raiseMe try: while True: item = self.resultQueue.get(block=False) item() except Queue.Empty: pass def addDownloadJob(self, job): """Add another DownloadJob to the end of the work queue.""" job.setDownloadStatusLog(self.statusLog) rp = job.getRelativePath() self._lock.acquire() self.downloads[rp] = job self._lock.release() self.downloadQueue.put(job) def getRetryTime(self, mirrorList, relPath): """Given a mirrorlist and a filename relative to the repository root, return the next time at which we are willing to retry fetching that file from any mirror, or 0 if we are willing to try immediately. """ readyAt = None for m in mirrorsThatSupport(mirrorList, relPath): r = self.statusLog.getDelayTime(m['urlbase']) if readyAt == None or r < readyAt: readyAt = r if readyAt != None: return readyAt else: return 0 def downloadFailed(self, mirror, relpath): """Callback: invoked when a download fails.""" pass def _thread(self, idx): # Run in the background per thread. idx is the number of the thread. while True: job = self.downloadQueue.get() # Grab job from queue. rp = job.getRelativePath() success = False try: logging.info("start %s in Thread %s", rp, idx) failure = job.download() # Execute the download. logging.info("end %s in Thread %s", rp, idx) finally: self._lock.acquire() try: del self.downloads[rp] if success: # If we downloaded correctly, say so. self.haveDownloaded[rp] = True finally: self._lock.release() if failure == None: self.statusLog.succeeded(job.getMirror(), job.getRelativePath()) self.resultQueue.put(job._success) else: self.statusLog.failed(failure) self.resultQueue.put(job._failure) self.done.acquire() self.done.notify() self.done.release() class DownloadFailure: """Helper class: represents a failure to download an item from a mirror. Broadly speaking, these errors are possible: A - The whole internet is down for us, either because our network connection sucks, our proxy is down, or whatever. - A particular mirror is down or nonfunctional. For these, use DownloadFailure.connectionFailed(): we can't easily tell a mirror failure from a network failure when we have only a single mirror to look at. B - The mirror is giving us errors we don't understand. - A particular mirror is missing a file we need. - A particular mirror served us something that was allegedly a file we need, but that file was no good. For these, use DownloadFailure.mirrorFailed(): Whether the mirror is broken or we're misunderstanding it, don't try to use it for a while. C - We finished a partial download and it was no good, but we can't tell who was at fault, because we don't know which part was corrupt. Use DownloadFailure.badCompoundFile(). We don't know who to blame, so we treat the whole network as having gone bibbledy. """ def __init__(self, urlbase, relPath, networkError=False): self._urlbase = urlbase self._relPath = relPath self._network = networkError self._when = time.time() @staticmethod def badCompoundFile(relpath): return DownloadFailure(None, relpath) @staticmethod def mirrorFailed(urlbase, relpath): return DownloadFailure(urlbase, relpath) @staticmethod def connectionFailed(urlbase): return DownloadFailure(urlbase, None, True) S = thandy.checkJson _FAIL_SCHEMA = S.Struct([S.Int(), thandy.formats.TIME_SCHEMA], allowMore=True) _STATUS_LOG_SCHEMA = S.Obj( v=S.Int(), mirrorFailures=S.DictOf(S.AnyStr(), _FAIL_SCHEMA), networkFailures=_FAIL_SCHEMA) del S class DownloadStatusLog: """Tracks when we can retry downloading from various mirrors. Currently, we treat every failure as affecting a mirror, the network, or both. When a mirror or the network fails, we back off for a while before we attempt it again. For each failure with no intervening success, we back off longer. """ # XXXX get smarter. # XXXX make this persistent. def __init__(self, mirrorFailures={}, networkFailures=[0,0]): self._lock = threading.RLock() # Map from urlbase to [ nFailures, lastFailureTime ] self._mirrorFailures = dict(mirrorFailures) # [ nFailures, lastFailureTime ] for the network as a while. self._netFailure = list(networkFailures) def _getDelay(self, isMirror, failureCount): """Return how long we should wait since the 'failureCount'th consecutive failure of something before we try it again. If isMirror, it's a mirror. Otherwise, it's the network.""" if isMirror: DELAYS = [ 0, 300, 300, 600, 900, 8400, 8400, 9000 ] else: DELAYS = [ 0, 10, 30, 30, 60, 300, 600, 1800, 3600, 7200 ] if failureCount < len(DELAYS): return DELAYS[failureCount] else: return DELAYS[-1] def toJSON(self): """Return an object suitable for encoding with json to represent the state of this DownloadStatusLog.""" def formatEnt(e): return [ e[0], thandy.formats.formatTime(e[1]) ] return { 'v': 1, 'networkFailures' : formatEnt(self._netFailure), 'mirrorFailures' : dict((k, formatEnt(v)) for k, v in self._mirrorFailures.iteritems()) } @staticmethod def fromJSON(obj): _STATUS_LOG_SCHEMA.checkMatch(obj) def parseEnt(e): return [ e[0], thandy.formats.parseTime(e[1]) ] return DownloadStatusLog( dict((k, parseEnt(v)) for k,v in obj['mirrorFailures'].iteritems()), parseEnt(obj['networkFailures'])) def failed(self, failure): """Note that 'failure', a DownloadFailure object, has occurred.""" self._lock.acquire() try: when = long(failure._when) # If there's a mirror to blame, blame it. if failure._urlbase != None: s = self._mirrorFailures.setdefault(failure._urlbase, [0, 0]) # XXXX This "+ 5" business is a hack. The idea is to keep # multiple failure within 5 seconds from counting as one, # since it's common for us to launch multiple downloads # simultaneously. If we launch 3, and they all fail because # the network is down, we want that to count as 1 failure, # not 3. if s[1] + 5 < when: s[0] += 1 s[1] = when # If there is no mirror to blame, or we suspect a network error, # blame the network too. if failure._urlbase == None or failure._network: s = self._netFailure # see note above. if s[1] + 5 < when: s[0] += 1 s[1] = when finally: self._lock.release() def succeeded(self, urlbase, url): """Note that we have successfully fetched url from the mirror at urlbase.""" self._lock.acquire() try: try: del self._mirrorFailures[urlbase] except KeyError: pass self._netFailure = [0, 0] finally: self._lock.release() def canRetry(self, urlbase=None, now=None): """Return True iff we are willing to retry the mirror at urlbase.""" if now == None: now = time.time() d = self.getDelayTime(urlbase) return d <= now def getDelayTime(self, urlbase=None): """Return the time after which we're willing to retry fetching from urlbase. 0 also means "we're ready now". """ self._lock.acquire() try: readyAt = 0 if urlbase: status = self._mirrorFailures.get(urlbase, (0,0)) if status[1] > readyAt: readyAt = status[1] + self._getDelay(True, status[0]) if self._netFailure[1] > readyAt: readyAt = (self._netFailure[1] + self._getDelay(False, self._netFailure[0])) return readyAt finally: self._lock.release() class DownloadJob: """Abstract base class. Represents a thing to be downloaded, and the knowledge of how to download it.""" def __init__(self, targetPath, tmpPath, wantHash=None, repoFile=None, useTor=False): """Create a new DownloadJob. When it is finally downloaded, store it in targetPath. Store partial results in tmpPath; if there is already a file in tmpPath, assume that it is an incomplete download. If wantHash, reject the file unless the hash is as given. If useTor, use a socks connection. If repoFile, use that RepositoryFile to validate the downloaded data.""" self._destPath = targetPath self._tmpPath = tmpPath self._wantHash = wantHash self._repoFile = repoFile self._useTor = useTor self._success = lambda : None self._failure = lambda : None def setCallbacks(self, success, failure): """Make sure that 'success' gets called if this download succeeds, and 'failure' gets called when it fails. Both are called from the downloader thread, so make sure to be threadsafe.""" self._success = success self._failure = failure def getURL(self): """Abstract implementation helper. Returns the URL that the _download function downloads from.""" raise NotImplemented() def getMirror(self): """Return a string identifying the mirror that's doing the downloading, if we know it.""" return None def getRelativePath(self): """Abstract. Returns a string representing this download, to keep two downloads of the same object from running at once. In Thandy, this is usually a relative path of a downloaded object within the repository. """ raise NotImplemented() def haveStalledFile(self): """Return true iff we have an existing incomplete download stored in the temporary file. """ return os.path.exists(self._tmpPath) def download(self): """Main interface function: Start the download, and return when complete. Return None on success, and a DownloadFailure on failure. """ try: self._download() return None except BadCompoundData, err: logging.warn("Download failed: %s", err) # No way to apportion the blame. return DownloadFailure.badCompoundFile(self.getRelativePath()) except (urllib2.HTTPError, thandy.DownloadError), err: # looks like we may have irreconcilable differences with a # particular mirror. logging.warn("Download failed: %s", err) return DownloadFailure.mirrorFailed(self.getMirror(), self.getRelativePath()) except (OSError, httplib.error, IOError, urllib2.URLError), err: logging.warn("Download failed: %s", err) # Could be the mirror; could be the network. Hard to say. return DownloadFailure.connectionFailed(self.getMirror()) except: tp, val, tb = sys.exc_info() logging.exception("Internal error during download: %s", val) # We have an exception! Treat it like a network error, I guess. return DownloadFailure.connectionFailed(None) def setDownloadStatusLog(self, log): """Base our URL-picking decisions on the DownloadStatusLog in 'log'. The caller is still reposnsible for invoking the logs failed() or succeeded methods. XXXX is that bad API design?""" pass def _checkTmpFile(self): """Helper: check whether the downloaded temporary file matches the hash and/or format we need.""" if self._wantHash and not self._repoFile: gotHash = thandy.formats.getFileDigest(self._tmpPath) if gotHash != self._wantHash: raise thandy.FormatException("File hash was not as expected.") elif self._repoFile: self._repoFile.checkFile(self._tmpPath, self._wantHash) def _removeTmpFile(self): """Helper: remove the temporary file so that we do not get stuck in a downloading-it-forever loop.""" os.unlink(self._tmpPath) def _download(self): # Implementation function. Unlike download(), can throw exceptions. f_in = f_out = None haveStalled = self.haveStalledFile() if haveStalled and self._wantHash: try: self._checkTmpFile() except thandy.Exception: pass else: # What luck! This stalled file was what we wanted. # (This happens mostly when we have an internal error.) thandy.util.ensureParentDir(self._destPath) thandy.util.moveFile(self._tmpPath, self._destPath) return try: url = self.getURL() logging.info("Downloading %s", url) if haveStalled: have_length = os.stat(self._tmpPath).st_size logging.info("Have stalled file for %s with %s bytes", url, have_length) else: have_length = None try: f_in = getConnection(url, self._useTor, have_length) except urllib2.HTTPError, err: if err.code == 416: # We asked for a range that couldn't be satisfied. # Usually, this means that the server thinks the file # is shorter than we think it is. We need to start over. self._removeTmpFile() raise logging.info("Connected to %s", url) gotRange = f_in.info().get("Content-Range") expectLength = f_in.info().get("Content-Length", "???") if gotRange: if gotRange.startswith("bytes %s-"%have_length): logging.info("Resuming download from %s"%url) f_out = open(self._tmpPath, 'ab') else: raise thandy.DownloadError("Got an unexpected range %s" %gotRange) else: f_out = open(self._tmpPath, 'wb') total = 0 while True: c = f_in.read(1024) if not c: break f_out.write(c) total += len(c) logging.debug("Got %s/%s bytes from %s", total, expectLength, url) finally: if f_in is not None: f_in.close() if f_out is not None: f_out.close() try: self._checkTmpFile() except (thandy.FormatException, thandy.DownloadError), err: self._removeTmpFile() if haveStalled: raise BadCompoundData(err) else: raise thandy.util.ensureParentDir(self._destPath) thandy.util.moveFile(self._tmpPath, self._destPath) class SimpleDownloadJob(DownloadJob): """Testing subtype of DownloadJob: just downloads a URL and writes it to disk.""" def __init__(self, targetPath, url, wantHash=None, supportedURLTypes=None, useTor=False): DownloadJob.__init__(self, targetPath, targetPath+".tmp", wantHash=wantHash, useTor=useTor) self._url = url def getURL(self): return self._url def getRelativePath(self): return self._url def mirrorsThatSupport(mirrorList, relPath, urlTypes=None, statusLog=None): """Generator: yields all the mirrors from mirrorList that let us fetch from relPath, whose URL type is in urlTypes (if present), and who are not marked as failed-too-recently in statusLog (if present).""" now = time.time() for m in mirrorList['mirrors']: if urlTypes != None: urltype = urllib2.splittype(m['urlbase'])[0] if urltype.lower() not in urlTypes: continue if statusLog != None and not statusLog.canRetry(m['urlbase'], now): continue for c in m['contents']: if thandy.formats.rolePathMatches(c, relPath): yield m break class ThandyDownloadJob(DownloadJob): """Thandy's subtype of DownloadJob: knows about mirrors, weighting, and Thandy's directory structure.""" def __init__(self, relPath, destPath, mirrorList, wantHash=None, supportedURLTypes=None, useTor=None, repoFile=None, downloadStatusLog=None): DownloadJob.__init__(self, destPath, None, wantHash=wantHash, useTor=useTor, repoFile=repoFile) self._mirrorList = mirrorList self._relPath = relPath tmppath = thandy.util.userFilename("tmp") if relPath.startswith("/"): relPath = relPath[1:] self._tmpPath = os.path.join(tmppath, relPath) d = os.path.dirname(self._tmpPath) if not os.path.exists(d): os.makedirs(d, 0700) self._supportedURLTypes = None if self._supportedURLTypes is None and useTor: self._supportedURLTypes = [ "http", "https" ] self._usingMirror = None #DOCDOC self._downloadStatusLog = downloadStatusLog def setDownloadStatusLog(self, log): self._downloadStatusLog = log def getURL(self): usable = [] for m in mirrorsThatSupport(self._mirrorList, self._relPath, self._supportedURLTypes, self._downloadStatusLog): usable.append( (m['weight'], m) ) try: mirror = thandy.util.randChooseWeighted(usable) except IndexError: raise thandy.DownloadError("No mirror supports download.") self._usingMirror = mirror['urlbase'] if m['urlbase'][-1] == '/' and self._relPath[0] == '/': return m['urlbase'] + self._relPath[1:] else: return m['urlbase'] + self._relPath def getRelativePath(self): return self._relPath def getMirror(self): return self._usingMirror _socks_opener = thandy.socksurls.build_socks_opener() def getConnection(url, useTor, have_length=None): """Open a connection to 'url'. We already have received have_length bytes of the file we're trying to fetch, so resume if possible. """ headers = {} urltype = urllib2.splittype(url)[0] is_http = urltype in ["http", "https"] if have_length is not None and is_http: headers['Range'] = "bytes=%s-"%have_length req = urllib2.Request(url, headers=headers) if useTor: conn = _socks_opener.open(req) else: conn = urllib2.urlopen(req) return conn if __name__ == '__main__': # Trivial CLI to test out downloading. import getopt options, args = getopt.getopt(sys.argv[1:], "", ["use-tor", "socksport=", "threads="]) useTor = False socksPort = 9050 nThreads = 2 for o,v in options: if o == "--use-tor": useTor = True elif o == "--socksport": socksPort = int(v) elif o == "--threads": nThreads = int(v) logging.basicConfig(level=logging.DEBUG) if useTor: thandy.socksurls.setSocksProxy("127.0.0.1", socksPort) manager = DownloadManager(nThreads) for url in args: fn = urllib2.splithost(urllib2.splittype(url)[1])[1] fn = os.path.split(fn)[1] job = SimpleDownloadJob(fn, url, useTor=useTor) manager.addDownloadJob(job) manager.start() manager.wait()