diff --git a/scrapy/trunk/scrapy/contrib/pipeline/media.py b/scrapy/trunk/scrapy/contrib/pipeline/media.py
new file mode 100644
index 000000000..df12233e2
--- /dev/null
+++ b/scrapy/trunk/scrapy/contrib/pipeline/media.py
@@ -0,0 +1,119 @@
+from __future__ import with_statement
+
+from twisted.internet import defer
+from twisted.python import failure
+from pprint import pprint
+
+from scrapy.core import log
+from scrapy.http import Request
+from scrapy.core.engine import scrapyengine
+from scrapy.spider import spiders
+from scrapy.core.exceptions import DropItem, NotConfigured
+from scrapy.core.exceptions import HttpException
+from scrapy.stats import stats
+from scrapy.utils.misc import chain_deferred, mustbe_deferred
+from scrapy.conf import settings
+
+class DomainInfo(object):
+    def __init__(self, domain):
+        self.domain = domain
+        self.spider = spiders.fromdomain(domain)
+        self.downloading = {}
+        self.downloaded = {}
+        self.waiting = {}
+
+
+class MediaPipeline(object):
+    def __init__(self):
+        self.cache = {}
+
+    def process_item(self, domain, response, item):
+        urls = self.get_urls_from_item(item)
+        info = self.cache[domain]
+
+        ulist = []
+        for url in urls:
+            dfd = self._enqueue(url, info)
+            dfd.addCallbacks(
+                    callback=self.new_item_media,
+                    callbackArgs=(item,),
+                    errback=self.failed_item_media,
+                    errbackArgs=(item,),
+                    )
+            ulist.append(dfd)
+
+        dlist = defer.DeferredList(ulist, consumeErrors=False)
+        dlist.addBoth(self._item_completed, item)
+        return dlist
+
+    def _enqueue(self, url, info):
+        wad = defer.Deferred()
+        wad.addBoth(lambda _: pprint(_))
+
+        request = Request(url=url)
+        fp = request.fingerprint()
+        waiting = info.waiting.setdefault(fp, []).append(wad)
+
+        if fp in info.downloading:
+            return
+
+        dwld = scrapyengine.schedule(request, info.spider, priority=0)
+        dwld.addCallbacks(self.media_downloaded, self.media_failure)
+        dwld.addBoth(self._download_finished, info, fp)
+        info.downloading[fp] = (request, dwld)
+        return wad
+
+    def _download_finished(self, result, info, fp):
+        del info.downloading[fp]
+        info.downloaded[fp] = result # cache result
+        waiting = info.waiting[fp]
+
+        isfail = isinstance(result, failure.Failure)
+        for wad in waiting:
+            tocall = wad.errback if isfail else wad.callback
+            pprint(tocall)
+            tocall(result)
+        return result
+
+    def _item_completed(self, result, item):
+        return self.item_completed(item) #dummy as hell
+
+    def open_domain(self, domain):
+        self.cache[domain] = DomainInfo(domain)
+
+    def close_domain(self, domain):
+        print '##### closing'
+        del self.cache[domain]
+
+    ### Overradiable Interface
+
+    def get_urls_from_item(self, item):
+        print '##### get_urls_from_item'
+        return item.image_urls
+
+    def media_to_download(self, url):
+        print '##### media_to_download'
+        pass
+
+    def media_downloaded(self, response):
+        print '##### media_downloaded'
+        pass
+
+    def media_failure(self, _failure):
+        print '##### media_failure'
+        return _failure
+
+    def new_item_media(self, result, item):
+        print '##### new_item_media'
+        pass
+
+    def failed_item_media(self, _failure, item):
+        print '##### failed_item_media'
+        pass
+
+    def item_completed(self, item):
+        print '##### item_completed'
+        return item
+
+
+