Merge branch 'develop' for 0.5.0 release

CHANGES.rst

@@ -1,3 +1,33 @@
+pywb 0.5.0 changelist
+~~~~~~~~~~~~~~~~~~~~~
+
+* Catch live rewrite errors and display more friendly pywb error message.
+
+* LiveRewriteHandler and WBHandler refactoring: LiveRewriteHandler now supports a root search page html template.
+
+* Proxy mode option: 'unaltered_replay' to proxy archival data with no modifications (no banner, no server or client side rewriting).
+
+* Fix client side rewriting (wombat.js) for proxy mode: only rewrite https -> http in absolute urls.
+
+* Fixes to memento timemap/timegate to work with framed replay mode.
+
+* Support for a fallback handler which will be called from a replay handler instead of a 404 response.
+
+  The handler, specified via the ``fallback`` option, can be the name of any other replay handler. Typically, it can be used with a live rewrite handler to fetch missing content from live instead of showing a 404.
+
+* Live Rewrite can now be included as a 'collection type' in a pywb deployment by setting index path to ``$liveweb``.
+
+* ``live-rewrite-server`` has optional ``--proxy host:port`` param to specify a loading live web data through an HTTP/S proxy, such as for use with a recording proxy.
+
+* wombat: add document.cookie -> document.WB_wombat_cookie rewriting to check and rewrite Path= to archival url
+
+* Better parent relative '../' path rewriting, resolved to correct absolute urls when rewritten. Additional testing for parent relative urls.
+
+* New 'proxy_options' block, including 'use_default_coll' to allow defaulting to first collection w/o proxy auth.
+
+* Improved support for proxy mode, allow different collections to be selected via proxy auth
+
+
 pywb 0.4.7 changelist
 ~~~~~~~~~~~~~~~~~~~~~
 

README.rst

@@ -1,4 +1,4 @@
-PyWb 0.4.7
+PyWb 0.5.0
 ==========
 
 .. image:: https://travis-ci.org/ikreymer/pywb.png?branch=master
@@ -11,9 +11,25 @@ pywb is a python implementation of web archival replay tools, sometimes also kno
 
 pywb allows high-quality replay (browsing) of archived web data stored in standardized `ARC <http://en.wikipedia.org/wiki/ARC_(file_format)>`_ and `WARC <http://en.wikipedia.org/wiki/Web_ARChive>`_.
 
-*For an example of deployed service using pywb, please see the https://webrecorder.io project*
 
-pywb Tools
+Usage Examples
+-----------------------------
+
+This README contains a basic overview of using pywb. After reading this intro, consider also taking a look at these seperate projects:
+
+* `pywb-webrecorder <https://github.com/ikreymer/pywb-webrecorder>`_ demonstrates a way to use pywb and warcprox to record web content while browsing.
+
+* `pywb-samples <https://github.com/ikreymer/pywb-samples>`_ provides additional archive samples with difficult-to-replay content.
+
+
+The following deployed applications use pywb:
+
+* https://perma.cc embeds pywb as part of a larger `open source application <https://github.com/harvard-lil/perma>`_ to provide web archive replay for law libraries.
+
+* https://webrecorder.io uses pywb and builds upon pywb-webrecorder to create a hosted web recording and replay system.
+
+
+pywb Tools Overview
 -----------------------------
 
 In addition to the standard wayback machine (explained further below), pywb tool suite includes a 
@@ -72,7 +88,7 @@ This process can be done by running the ``cdx-indexer`` script and only needs to
 
 Given an archive of warcs at ``myarchive/warcs``
 
-1. Create a dir for indexs, .eg. ``myarchive/cdx``
+1. Create a dir for indexes, .eg. ``myarchive/cdx``
 
 2. Run ``cdx-indexer --sort myarchive/cdx myarchive/warcs`` to generate .cdx files for each
    warc/arc file in ``myarchive/warcs``

config.yaml

@@ -91,6 +91,12 @@ static_routes:
 # Enable simple http proxy mode
 enable_http_proxy: true
 
+# Additional proxy options (defaults)
+#proxy_options:
+#    use_default_coll: true
+#
+#    unaltered_replay: false
+
 # enable cdx server api for querying cdx directly (experimental)
 enable_cdx_api: true
 

pywb/apps/live_rewrite_server.py

@@ -2,15 +2,36 @@ from pywb.framework.wsgi_wrappers import init_app, start_wsgi_server
 
 from pywb.webapp.live_rewrite_handler import create_live_rewriter_app
 
+from argparse import ArgumentParser
+
+
 #=================================================================
-# init cdx server app
+# init rewrite server app
 #=================================================================
 
-application = init_app(create_live_rewriter_app, load_yaml=False)
+def create_app():
+    parser = ArgumentParser(description='Live Rewrite Server')
+
+    parser.add_argument('-x', '--proxy',
+                        action='store',
+                        help='Specify host:port to use as HTTP/S proxy')
+
+    result, unknown = parser.parse_known_args()
+
+    config = dict(proxyhostport=result.proxy, framed_replay=True)
+
+    app = init_app(create_live_rewriter_app, load_yaml=False,
+                   config=config)
+
+    return app
+
+
+application = create_app()
 
 
 def main():  # pragma: no cover
     start_wsgi_server(application, 'Live Rewriter App', default_port=8090)
 
+
 if __name__ == "__main__":
     main()

pywb/cdx/cdxserver.py

@@ -58,7 +58,7 @@ class BaseCDXServer(object):
                 return self.load_cdx(**fuzzy_query_params)
 
         msg = 'No Captures found for: ' + query.url
-        raise NotFoundException(msg)
+        raise NotFoundException(msg, url=query.url)
 
     def _calc_search_keys(self, query):
         return calc_search_range(url=query.url,

pywb/framework/archivalrouter.py

@@ -29,16 +29,48 @@ class ArchivalRouter(object):
         self.error_view = kwargs.get('error_view')
 
     def __call__(self, env):
+        request_uri = env['REL_REQUEST_URI']
+
         for route in self.routes:
-            result = route(env, self.abs_path)
+            matcher, coll = route.is_handling(request_uri)
-            if result:
+            if matcher:
-                return result
+                wbrequest = self.parse_request(route, env, matcher,
+                                               coll, request_uri,
+                                               use_abs_prefix=self.abs_path)
+
+                return route.handler(wbrequest)
 
         # Default Home Page
-        if env['REL_REQUEST_URI'] in ['/', '/index.html', '/index.htm']:
+        if request_uri in ['/', '/index.html', '/index.htm']:
             return self.render_home_page(env)
 
-        return self.fallback(env, self.routes) if self.fallback else None
+        return self.fallback(env, self) if self.fallback else None
+
+    def parse_request(self, route, env, matcher, coll, request_uri,
+                      use_abs_prefix=False):
+        matched_str = matcher.group(0)
+        if matched_str:
+            rel_prefix = env['SCRIPT_NAME'] + '/' + matched_str + '/'
+            # remove the '/' + rel_prefix part of uri
+            wb_url_str = request_uri[len(matched_str) + 2:]
+        else:
+            rel_prefix = env['SCRIPT_NAME'] + '/'
+            # the request_uri is the wb_url, since no coll
+            wb_url_str = request_uri[1:]
+
+        wbrequest = route.request_class(env,
+                              request_uri=request_uri,
+                              wb_url_str=wb_url_str,
+                              rel_prefix=rel_prefix,
+                              coll=coll,
+                              use_abs_prefix=use_abs_prefix,
+                              wburl_class=route.handler.get_wburl_type(),
+                              urlrewriter_class=UrlRewriter)
+
+        # Allow for applying of additional filters
+        route.apply_filters(wbrequest, matcher)
+
+        return wbrequest
 
     def render_home_page(self, env):
         # render the homepage!
@@ -73,45 +105,15 @@ class Route(object):
         self.coll_group = coll_group
         self._custom_init(config)
 
-    def __call__(self, env, use_abs_prefix):
+    def is_handling(self, request_uri):
-        wbrequest = self.parse_request(env, use_abs_prefix)
-        return self.handler(wbrequest) if wbrequest else None
-
-    def parse_request(self, env, use_abs_prefix, request_uri=None):
-        if not request_uri:
-            request_uri = env['REL_REQUEST_URI']
-
         matcher = self.regex.match(request_uri[1:])
         if not matcher:
-            return None
+            return None, None
-
-        matched_str = matcher.group(0)
-        if matched_str:
-            rel_prefix = env['SCRIPT_NAME'] + '/' + matched_str + '/'
-            # remove the '/' + rel_prefix part of uri
-            wb_url_str = request_uri[len(matched_str) + 2:]
-        else:
-            rel_prefix = env['SCRIPT_NAME'] + '/'
-            # the request_uri is the wb_url, since no coll
-            wb_url_str = request_uri[1:]
 
         coll = matcher.group(self.coll_group)
+        return matcher, coll
 
-        wbrequest = self.request_class(env,
+    def apply_filters(self, wbrequest, matcher):
-                              request_uri=request_uri,
-                              wb_url_str=wb_url_str,
-                              rel_prefix=rel_prefix,
-                              coll=coll,
-                              use_abs_prefix=use_abs_prefix,
-                              wburl_class=self.handler.get_wburl_type(),
-                              urlrewriter_class=UrlRewriter)
-
-        # Allow for applying of additional filters
-        self._apply_filters(wbrequest, matcher)
-
-        return wbrequest
-
-    def _apply_filters(self, wbrequest, matcher):
         for filter in self.filters:
             last_grp = len(matcher.groups())
             filter_str = filter.format(matcher.group(last_grp))
@@ -136,9 +138,11 @@ class ReferRedirect:
         else:
             self.match_prefixs = [match_prefixs]
 
-    def __call__(self, env, routes):
+    def __call__(self, env, the_router):
         referrer = env.get('HTTP_REFERER')
 
+        routes = the_router.routes
+
         # ensure there is a referrer
         if referrer is None:
             return None
@@ -166,17 +170,15 @@ class ReferRedirect:
         ref_request = None
 
         for route in routes:
-            ref_request = route.parse_request(env, False, request_uri=path)
+            matcher, coll = route.is_handling(path)
-            if ref_request:
+            if matcher:
+                ref_request = the_router.parse_request(route, env,
+                                                       matcher, coll, path)
                 ref_route = route
                 break
 
-        # must have matched one of the routes
+        # must have matched one of the routes with a urlrewriter
-        if not ref_request:
+        if not ref_request or not ref_request.urlrewriter:
-            return None
-
-        # must have a rewriter
-        if not ref_request.urlrewriter:
             return None
 
         rewriter = ref_request.urlrewriter

pywb/framework/memento.py

@@ -11,15 +11,13 @@ LINK_FORMAT = 'application/link-format'
 #=================================================================
 class MementoReqMixin(object):
     def _parse_extra(self):
-        self.is_timegate = False
-
         if not self.wb_url:
             return
 
         if self.wb_url.type != self.wb_url.LATEST_REPLAY:
             return
 
-        self.is_timegate = True
+        self.options['is_timegate'] = True
 
         accept_datetime = self.env.get('HTTP_ACCEPT_DATETIME')
         if not accept_datetime:
@@ -48,7 +46,7 @@ class MementoRespMixin(object):
         if not wbrequest or not wbrequest.wb_url:
             return
 
-        is_timegate = wbrequest.is_timegate
+        is_timegate = wbrequest.options.get('is_timegate', False)
 
         if is_timegate:
             self.status_headers.headers.append(('Vary', 'accept-datetime'))
@@ -59,7 +57,7 @@ class MementoRespMixin(object):
             is_memento = False
 
         # otherwise, if in proxy mode, then always a memento
-        elif wbrequest.is_proxy:
+        elif wbrequest.options['is_proxy']:
             is_memento = True
 
         # otherwise only for replay
@@ -80,7 +78,7 @@ class MementoRespMixin(object):
             link.append(self.make_link(req_url, 'original'))
 
         # for now, include timemap only in non-proxy mode
-        if not wbrequest.is_proxy and (is_memento or is_timegate):
+        if not wbrequest.options['is_proxy'] and (is_memento or is_timegate):
             link.append(self.make_timemap_link(wbrequest))
 
         if is_memento and not is_timegate:
@@ -117,6 +115,7 @@ def make_memento_link(cdx, prefix, datetime=None, rel='memento', end=',\n'):
     memento = '<{0}>; rel="{1}"; datetime="{2}"' + end
 
     string = WbUrl.to_wburl_str(url=cdx['original'],
+                                mod='mp_',
                                 timestamp=cdx['timestamp'],
                                 type=WbUrl.REPLAY)
 
@@ -140,7 +139,8 @@ def make_timemap(wbrequest, cdx_lines):
     # timemap link
     timemap = ('<{0}>; rel="self"; ' +
                'type="application/link-format"; from="{1}",\n')
-    yield timemap.format(prefix + wbrequest.wb_url.to_str(), from_date)
+    yield timemap.format(prefix + wbrequest.wb_url.to_str(),
+                         from_date)
 
     # original link
     original = '<{0}>; rel="original",\n'
@@ -148,7 +148,7 @@ def make_timemap(wbrequest, cdx_lines):
 
     # timegate link
     timegate = '<{0}>; rel="timegate",\n'
-    yield timegate.format(prefix + url)
+    yield timegate.format(prefix + 'mp_/' + url)
 
     # first memento link
     yield make_memento_link(first_cdx, prefix,

pywb/framework/proxy.py

@@ -1,8 +1,11 @@
 from wbrequestresponse import WbResponse, WbRequest
 from archivalrouter import ArchivalRouter
+
 import urlparse
+import base64
 
 from pywb.rewrite.url_rewriter import HttpsUrlRewriter
+from pywb.utils.statusandheaders import StatusAndHeaders
 
 
 #=================================================================
@@ -15,10 +18,7 @@ class ProxyArchivalRouter(ArchivalRouter):
     """
     def __init__(self, routes, **kwargs):
         super(ProxyArchivalRouter, self).__init__(routes, **kwargs)
-        request_class = routes[0].request_class
+        self.proxy = ProxyRouter(routes, **kwargs)
-        self.proxy = ProxyRouter(routes[0].handler,
-                                 request_class=request_class,
-                                 **kwargs)
 
     def __call__(self, env):
         response = self.proxy(env)
@@ -43,12 +43,23 @@ class ProxyRouter(object):
     See: http://www.mementoweb.org/guide/rfc/#Pattern1.3
     for more details.
     """
-    def __init__(self, handler, **kwargs):
+
-        self.handler = handler
+    def __init__(self, routes, **kwargs):
+        self.routes = routes
         self.hostpaths = kwargs.get('hostpaths')
 
         self.error_view = kwargs.get('error_view')
-        self.request_class = kwargs.get('request_class')
+
+        proxy_options = kwargs.get('config', {})
+        if proxy_options:
+            proxy_options = proxy_options.get('proxy_options', {})
+
+        self.auth_msg = proxy_options.get('auth_msg',
+        'Please enter name of a collection to use for proxy mode')
+
+        self.use_default_coll = proxy_options.get('use_default_coll', True)
+
+        self.unaltered = proxy_options.get('unaltered_replay', False)
 
     def __call__(self, env):
         url = env['REL_REQUEST_URI']
@@ -59,16 +70,57 @@ class ProxyRouter(object):
         if not url.startswith('http://'):
             return None
 
-        wbrequest = self.request_class(env,
+        proxy_auth = env.get('HTTP_PROXY_AUTHORIZATION')
+
+        route = None
+        coll = None
+        matcher = None
+
+        if proxy_auth:
+            proxy_coll = self.read_basic_auth_coll(proxy_auth)
+
+            if not proxy_coll:
+                return self.proxy_auth_coll_response()
+
+            proxy_coll = '/' + proxy_coll + '/'
+
+            for r in self.routes:
+                matcher, c = r.is_handling(proxy_coll)
+                if matcher:
+                    route = r
+                    coll = c
+                    break
+
+            if not route:
+                return self.proxy_auth_coll_response()
+
+        # if 'use_default_coll' or only one collection, use that
+        # for proxy mode
+        elif self.use_default_coll or len(self.routes) == 1:
+            route = self.routes[0]
+            coll = self.routes[0].regex.pattern
+
+        # otherwise, require proxy auth 407 to select collection
+        else:
+            return self.proxy_auth_coll_response()
+
+        wbrequest = route.request_class(env,
                               request_uri=url,
                               wb_url_str=url,
+                              coll=coll,
                               host_prefix=self.hostpaths[0],
-                              wburl_class=self.handler.get_wburl_type(),
+                              wburl_class=route.handler.get_wburl_type(),
                               urlrewriter_class=HttpsUrlRewriter,
                               use_abs_prefix=False,
                               is_proxy=True)
 
-        return self.handler(wbrequest)
+        if matcher:
+            route.apply_filters(wbrequest, matcher)
+
+        if self.unaltered:
+            wbrequest.wb_url.mod = 'id_'
+
+        return route.handler(wbrequest)
 
     # Proxy Auto-Config (PAC) script for the proxy
     def make_pac_response(self, env):
@@ -97,3 +149,27 @@ class ProxyRouter(object):
         content_type = 'application/x-ns-proxy-autoconfig'
 
         return WbResponse.text_response(buff, content_type=content_type)
+
+    def proxy_auth_coll_response(self):
+        proxy_msg = 'Basic realm="{0}"'.format(self.auth_msg)
+
+        headers = [('Content-Type', 'text/plain'),
+                   ('Proxy-Authenticate', proxy_msg)]
+
+        status_headers = StatusAndHeaders('407 Proxy Authentication', headers)
+
+        value = self.auth_msg
+
+        return WbResponse(status_headers, value=[value])
+
+    @staticmethod
+    def read_basic_auth_coll(value):
+        parts = value.split(' ')
+        if parts[0].lower() != 'basic':
+            return ''
+
+        if len(parts) != 2:
+            return ''
+
+        user_pass = base64.b64decode(parts[1])
+        return user_pass.split(':')[0]

pywb/framework/test/test_archivalrouter.py

@@ -1,7 +1,7 @@
 """
 # Test WbRequest parsed via a Route
 # route with relative path, print resulting wbrequest
->>> print_req(Route('web', WbUrlHandler())({'REL_REQUEST_URI': '/web/test.example.com', 'SCRIPT_NAME': ''}, False))
+>>> _test_route_req(Route('web', WbUrlHandler()), {'REL_REQUEST_URI': '/web/test.example.com', 'SCRIPT_NAME': ''})
 {'coll': 'web',
  'request_uri': '/web/test.example.com',
  'wb_prefix': '/web/',
@@ -9,21 +9,21 @@
 
 
 # route with absolute path, running at script /my_pywb, print resultingwbrequest
->>> print_req(Route('web', WbUrlHandler())({'REL_REQUEST_URI': '/web/2013im_/test.example.com', 'SCRIPT_NAME': '/my_pywb', 'HTTP_HOST': 'localhost:8081', 'wsgi.url_scheme': 'https'}, True))
+>>> _test_route_req(Route('web', WbUrlHandler()), {'REL_REQUEST_URI': '/web/2013im_/test.example.com', 'SCRIPT_NAME': '/my_pywb', 'HTTP_HOST': 'localhost:8081', 'wsgi.url_scheme': 'https'}, True)
 {'coll': 'web',
  'request_uri': '/web/2013im_/test.example.com',
  'wb_prefix': 'https://localhost:8081/my_pywb/web/',
  'wb_url': ('replay', '2013', 'im_', 'http://test.example.com', '2013im_/http://test.example.com')}
 
 # route with no collection
->>> print_req(Route('', BaseHandler())({'REL_REQUEST_URI': 'http://example.com', 'SCRIPT_NAME': '/pywb'}, False))
+>>> _test_route_req(Route('', BaseHandler()), {'REL_REQUEST_URI': 'http://example.com', 'SCRIPT_NAME': '/pywb'})
 {'coll': '',
  'request_uri': 'http://example.com',
  'wb_prefix': '/pywb/',
  'wb_url': None}
 
 # not matching route -- skipped
->>> Route('web', BaseHandler())({'REL_REQUEST_URI': '/other/test.example.com', 'SCRIPT_NAME': ''}, False)
+>>> _test_route_req(Route('web', BaseHandler()), {'REL_REQUEST_URI': '/other/test.example.com', 'SCRIPT_NAME': ''})
 
 
 # Referer Redirect Test
@@ -84,11 +84,18 @@ False
 
 """
 
-from pywb.framework.archivalrouter import Route, ReferRedirect
+from pywb.framework.archivalrouter import Route, ReferRedirect, ArchivalRouter
 from pywb.framework.basehandlers import BaseHandler, WbUrlHandler
 import pprint
 
-def print_req(req):
+def _test_route_req(route, env, abs_path=False):
+    matcher, coll = route.is_handling(env['REL_REQUEST_URI'])
+    if not matcher:
+        return
+
+    the_router = ArchivalRouter([route], abs_path=abs_path)
+    req = the_router.parse_request(route, env, matcher, coll, env['REL_REQUEST_URI'], abs_path)
+
     varlist = vars(req)
     the_dict = dict((k, varlist[k]) for k in ('request_uri', 'wb_prefix', 'wb_url', 'coll'))
     pprint.pprint(the_dict)
@@ -102,9 +109,11 @@ def _test_redir(match_host, request_uri, referrer, script_name = '', coll = 'col
 
     routes = [Route(coll, WbUrlHandler())]
 
+    the_router = ArchivalRouter(routes)
+
     redir = ReferRedirect(match_host)
     #req = WbRequest.from_uri(request_uri, env)
-    rep = redir(env, routes)
+    rep = redir(env, the_router)
     if not rep:
         return False
 

pywb/framework/test/test_wsgi_wrapper.py

@@ -22,7 +22,7 @@ class TestCustomErrApp:
 
 
 def initer(app_class):
-    def init():
+    def init(config=None):
         return app_class()
     return init
 

pywb/framework/wbrequestresponse.py

@@ -78,12 +78,11 @@ class WbRequest(object):
 
         self.referrer = env.get('HTTP_REFERER')
 
-        self.is_ajax = self._is_ajax()
+        self.options = dict()
+        self.options['is_ajax'] = self._is_ajax()
+        self.options['is_proxy'] = is_proxy
 
         self.query_filter = []
-
-        self.is_proxy = is_proxy
-
         self.custom_params = {}
 
         # PERF

pywb/framework/wsgi_wrappers.py

@@ -112,7 +112,7 @@ DEFAULT_CONFIG_FILE = 'config.yaml'
 
 
 #=================================================================
-def init_app(init_func, load_yaml=True, config_file=None):
+def init_app(init_func, load_yaml=True, config_file=None, config={}):
     logging.basicConfig(format='%(asctime)s: [%(levelname)s]: %(message)s',
                         level=logging.DEBUG)
     logging.debug('')
@@ -129,9 +129,7 @@ def init_app(init_func, load_yaml=True, config_file=None):
 
             config = load_yaml_config(config_file)
 
-            wb_router = init_func(config)
+        wb_router = init_func(config)
-        else:
-            wb_router = init_func()
     except:
         msg = '*** pywb app init FAILED config from "%s"!\n'
         logging.exception(msg, init_func.__name__)
@@ -146,17 +144,8 @@ def init_app(init_func, load_yaml=True, config_file=None):
 #=================================================================
 def start_wsgi_server(the_app, name, default_port=None):  # pragma: no cover
     from wsgiref.simple_server import make_server
-    from optparse import OptionParser
 
-    opt = OptionParser('%prog [OPTIONS]')
+    port = the_app.port
-    opt.add_option('-p', '--port', type='int', default=None)
-
-    options, args = opt.parse_args()
-
-    port = options.port
-
-    if not port:
-        port = the_app.port
 
     if not port:
         if default_port:

pywb/rewrite/header_rewriter.py

@@ -31,7 +31,7 @@ class HeaderRewriter:
         'xml':  ['/xml', '+xml', '.xml', '.rss'],
     }
 
-    PROXY_HEADERS = ['content-type', 'content-disposition']
+    PROXY_HEADERS = ['content-type', 'content-disposition', 'content-range']
 
     URL_REWRITE_HEADERS = ['location', 'content-location', 'content-base']
 

pywb/rewrite/regex_rewriters.py

@@ -126,6 +126,7 @@ class JSLinkAndLocationRewriter(JSLinkOnlyRewriter):
              (r'(?<!/)\blocation\b', RegexRewriter.add_prefix(prefix), 0),
              (r'(?<=document\.)domain', RegexRewriter.add_prefix(prefix), 0),
              (r'(?<=document\.)referrer', RegexRewriter.add_prefix(prefix), 0),
+             (r'(?<=document\.)cookie', RegexRewriter.add_prefix(prefix), 0),
 
             #todo: move to mixin?
              (r'(?:[\s=(){]|^)(top)(?:[\s!=}()]|$)',

pywb/rewrite/rewrite_content.py

@@ -160,7 +160,7 @@ class RewriteContent:
                                   first_buff=first_buff)
 
     @staticmethod
-    def _decode_buff(buff, stream, encoding): # pragma: no coverage
+    def _decode_buff(buff, stream, encoding):  # pragma: no coverage
         try:
             buff = buff.decode(encoding)
         except UnicodeDecodeError, e:

pywb/rewrite/rewrite_live.py

@@ -5,6 +5,7 @@ Fetch a url from live web and apply rewriting rules
 import requests
 import datetime
 import mimetypes
+import logging
 
 from urlparse import urlsplit
 
@@ -19,24 +20,13 @@ from pywb.rewrite.rewrite_content import RewriteContent
 
 #=================================================================
 class LiveRewriter(object):
-    PROXY_HEADER_LIST = [('HTTP_USER_AGENT', 'User-Agent'),
+    def __init__(self, defmod='', default_proxy=None):
-                         ('HTTP_ACCEPT', 'Accept'),
-                         ('HTTP_ACCEPT_LANGUAGE', 'Accept-Language'),
-                         ('HTTP_ACCEPT_CHARSET', 'Accept-Charset'),
-                         ('HTTP_ACCEPT_ENCODING', 'Accept-Encoding'),
-                         ('HTTP_RANGE', 'Range'),
-                         ('HTTP_CACHE_CONTROL', 'Cache-Control'),
-                         ('HTTP_X_REQUESTED_WITH', 'X-Requested-With'),
-                         ('HTTP_X_CSRF_TOKEN', 'X-CSRF-Token'),
-                         ('HTTP_PE_TOKEN', 'PE-Token'),
-                         ('HTTP_COOKIE', 'Cookie'),
-                         ('CONTENT_TYPE', 'Content-Type'),
-                         ('CONTENT_LENGTH', 'Content-Length'),
-                         ('REL_REFERER', 'Referer'),
-                        ]
-
-    def __init__(self, defmod=''):
         self.rewriter = RewriteContent(defmod=defmod)
+        self.default_proxy = default_proxy
+        if self.default_proxy:
+            logging.debug('Live Rewrite via proxy ' + self.default_proxy)
+        else:
+            logging.debug('Live Rewrite Direct (no proxy)')
 
     def fetch_local_file(self, uri):
         fh = open(uri)
@@ -89,6 +79,10 @@ class LiveRewriter(object):
         method = 'GET'
         data = None
 
+        if not proxies and self.default_proxy:
+            proxies = {'http': self.default_proxy,
+                       'https': self.default_proxy}
+
         if env is not None:
             method = env['REQUEST_METHOD'].upper()
             input_ = env['wsgi.input']
@@ -156,7 +150,8 @@ class LiveRewriter(object):
                'timestamp': timestamp,
                'original': url,
                'statuscode': status_headers.get_statuscode(),
-               'mimetype': status_headers.get_header('Content-Type')
+               'mimetype': status_headers.get_header('Content-Type'),
+               'is_live': True,
               }
 
         result = (self.rewriter.

pywb/rewrite/test/test_regex_rewriters.py

@@ -53,6 +53,10 @@ r"""
 >>> _test_js('cool_Location = "//example.com/abc.html" //comment')
 'cool_Location = "/web/20131010em_/http://example.com/abc.html" //comment'
 
+# document.cookie test
+>>> _test_js('document.cookie = "a=b; Path=/"')
+'document.WB_wombat_cookie = "a=b; Path=/"'
+
 
 #=================================================================
 # XML Rewriting
@@ -86,6 +90,12 @@ r"""
 >>> _test_css("background: url(file.jpeg)")
 'background: url(/web/20131010em_/http://example.com/file.jpeg)'
 
+>>> _test_css("background:#abc url('/static/styles/../images/layout/logo.png')")
+"background:#abc url('/web/20131010em_/http://example.com/static/images/layout/logo.png')"
+
+>>> _test_css("background:#000 url('/static/styles/../../images/layout/logo.png')")
+"background:#000 url('/web/20131010em_/http://example.com/images/layout/logo.png')"
+
 >>> _test_css("background: url('')")
 "background: url('')"
 

pywb/rewrite/test/test_rewrite_live.py

@@ -83,7 +83,8 @@ def test_example_domain_specific_3():
     assert '/* Bootloader.configurePage' in buff
 
 def test_wombat_top():
-    status_headers, buff = get_rewritten('https://assets-cdn.github.com/assets/github-0f06d0f46fe7bcfbf31f2380f23aec15ba21b8ec.js', urlrewriter)
+    #status_headers, buff = get_rewritten('https://assets-cdn.github.com/assets/github-0f06d0f46fe7bcfbf31f2380f23aec15ba21b8ec.js', urlrewriter)
+    status_headers, buff = get_rewritten(get_test_dir() + 'text_content/toptest.js', urlrewriter)
 
     assert 'WB_wombat_top!==window' in buff
 

pywb/rewrite/test/test_url_rewriter.py

@@ -1,4 +1,21 @@
 """
+# urljoin tests
+
+>>> UrlRewriter.urljoin('http://example.com/test/', '../file.html')
+'http://example.com/file.html'
+
+>>> UrlRewriter.urljoin('http://example.com/test/', '../path/../../../file.html')
+'http://example.com/file.html'
+
+>>> UrlRewriter.urljoin('http://example.com/test/', '/../file.html')
+'http://example.com/file.html'
+
+>>> UrlRewriter.urljoin('http://example.com/', '/abc/../../file.html')
+'http://example.com/file.html'
+
+>>> UrlRewriter.urljoin('http://example.com/path/more/', 'abc/../../file.html')
+'http://example.com/path/file.html'
+
 # UrlRewriter tests
 >>> do_rewrite('other.html', '20131010/http://example.com/path/page.html', 'https://web.archive.org/web/')
 'https://web.archive.org/web/20131010/http://example.com/path/other.html'

pywb/rewrite/url_rewriter.py

@@ -57,7 +57,7 @@ class UrlRewriter(object):
         else:
             # optimize: join if not absolute url, otherwise just use that
             if not is_abs:
-                new_url = urlparse.urljoin(wburl.url, url).replace('../', '')
+                new_url = self.urljoin(wburl.url, url)
             else:
                 new_url = url
 
@@ -81,8 +81,6 @@ class UrlRewriter(object):
         if new_url.startswith(self.prefix):
             new_url = new_url[len(self.prefix):]
 
-        #new_wburl = copy.copy(self.wburl)
-        #new_wburl.url = new_url
         new_wburl = WbUrl(new_url)
         return UrlRewriter(new_wburl, self.prefix)
 
@@ -92,6 +90,39 @@ class UrlRewriter(object):
     def __repr__(self):
         return "UrlRewriter('{0}', '{1}')".format(self.wburl, self.prefix)
 
+    @staticmethod
+    def urljoin(orig_url, url):
+        new_url = urlparse.urljoin(orig_url, url)
+        if '../' not in new_url:
+            return new_url
+
+        parts = urlparse.urlsplit(new_url)
+        scheme, netloc, path, query, frag = parts
+
+        path_parts = path.split('/')
+        i = 0
+        n = len(path_parts) - 1
+        while i < n:
+            if path_parts[i] == '..':
+                del path_parts[i]
+                n -= 1
+                if i > 0:
+                    del path_parts[i - 1]
+                    n -= 1
+                    i -= 1
+            else:
+                i += 1
+
+        if path_parts == ['']:
+            path = '/'
+        else:
+            path = '/'.join(path_parts)
+
+        parts = (scheme, netloc, path, query, frag)
+
+        new_url = urlparse.urlunsplit(parts)
+        return new_url
+
 
 #=================================================================
 class HttpsUrlRewriter(object):

pywb/static/wb.js

@@ -19,6 +19,12 @@ This file is part of pywb.
 
 _wb_js = (function() {
 
+
+var labels = {LOADING_MSG: "Loading...",
+              REPLAY_MSG: "This is an <b>archived</b> page from ",
+              LIVE_MSG: "This is a <b>live</b> page loaded on "};
+
+
 function init_banner() {
     var PLAIN_BANNER_ID = "_wb_plain_banner";
     var FRAME_BANNER_ID = "_wb_frame_top_banner";
@@ -40,19 +46,33 @@ function init_banner() {
 
     var banner = document.getElementById(bid);
     
-    if (!banner) {
+    if (banner) {
-        banner = document.createElement("wb_div");
+        return;
-        banner.setAttribute("id", bid);
-        banner.setAttribute("lang", "en");
-
-        text = "This is an archived page ";
-        if (wbinfo && wbinfo.capture_str) {
-            text += " from <b id='_wb_capture_info'>" + wbinfo.capture_str + "</b>";
-        }
-        banner.innerHTML = text;
-
-        document.body.insertBefore(banner, document.body.firstChild);
     }
+        
+    banner = document.createElement("wb_div");
+    banner.setAttribute("id", bid);
+    banner.setAttribute("lang", "en");
+
+    var text;
+
+    if (wbinfo.is_frame) {
+        text = labels.LOADING_MSG;
+    } else if (wbinfo.is_live) {
+        text = labels.LIVE_MSG;
+    } else {
+        text = labels.REPLAY_MSG;
+    }
+    
+    text = "<span id='_wb_label'>" + text + "</span>";
+
+    var capture_str = (wbinfo ? wbinfo.capture_str : "");
+
+    text += "<b id='_wb_capture_info'>" + capture_str + "</b>";
+    
+    banner.innerHTML = text;
+
+    document.body.insertBefore(banner, document.body.firstChild);
 }
 
 function add_event(name, func, object) {
@@ -105,7 +125,10 @@ function notify_top(event) {
     }
 
     if (window.top.update_wb_url) {
-        window.top.update_wb_url(window.WB_wombat_location.href, wbinfo.timestamp, wbinfo.capture_str);
+        window.top.update_wb_url(window.WB_wombat_location.href,
+                                 wbinfo.timestamp,
+                                 wbinfo.capture_str,
+                                 wbinfo.is_live);
     }
 }
 
@@ -126,4 +149,6 @@ if (wbinfo.is_frame_mp && wbinfo.canon_url &&
     window.location.replace(wbinfo.canon_url);
 }
 
+return {'labels': labels};
+
 })();

pywb/static/wombat.js

@@ -121,6 +121,15 @@ WB_wombat_init = (function() {
             return url;
         }
 
+        // proxy mode: If no wb_replay_prefix, only rewrite https:// -> http://
+        if (!wb_replay_prefix) {
+            if (starts_with(url, HTTPS_PREFIX)) {
+                return HTTP_PREFIX + url.substr(HTTPS_PREFIX.length);
+            } else {
+                return url;
+            }
+        }
+        
         // just in case wombat reference made it into url!
         url = url.replace("WB_wombat_", "");
 
@@ -181,6 +190,11 @@ WB_wombat_init = (function() {
             return "";
         }
         
+        // proxy mode: no extraction needed
+        if (!wb_replay_prefix) {
+            return href;
+        }
+                
         href = href.toString();
 
         var index = href.indexOf("/http", 1);
@@ -639,6 +653,31 @@ WB_wombat_init = (function() {
         }
     }
     
+    function init_cookies_override()
+    {
+        var cookie_path_regex = /\bPath=\'?\"?([^;'"\s]+)/i;
+        
+        var get_cookie = function() {
+            return document.cookie;
+        }
+        
+        var set_cookie = function(value) {
+            var matched = value.match(cookie_path_regex);
+                        
+            // if has cookie path, rewrite and replace
+            if (matched) {
+                var rewritten = rewrite_url(matched[1]);
+                value = value.replace(matched[1], rewritten);
+            }
+            
+            document.cookie = value;
+        }
+        
+        def_prop(document, "WB_wombat_cookie", document.cookie,
+                set_cookie,
+                get_cookie);
+    }
+    
     //============================================
     function init_write_override()
     {
@@ -658,20 +697,22 @@ WB_wombat_init = (function() {
     //============================================
     function wombat_init(replay_prefix, capture_date, orig_scheme, orig_host, timestamp) {
         wb_replay_prefix = replay_prefix;
-
-        wb_replay_date_prefix = replay_prefix + capture_date + "em_/";
         
-        if (capture_date.length > 0) {
+        if (wb_replay_prefix) {
-            wb_capture_date_part = "/" + capture_date + "/";
+            wb_replay_date_prefix = replay_prefix + capture_date + "em_/";
-        } else {
+            
-            wb_capture_date_part = "";
+            if (capture_date.length > 0) {
+                wb_capture_date_part = "/" + capture_date + "/";
+            } else {
+                wb_capture_date_part = "";
+            }
+            
+            wb_orig_scheme = orig_scheme + '://';
+    
+            wb_orig_host = wb_orig_scheme + orig_host;
+            
+            init_bad_prefixes(replay_prefix);
         }
-        
-        wb_orig_scheme = orig_scheme + '://';
-
-        wb_orig_host = wb_orig_scheme + orig_host;
-        
-        init_bad_prefixes(replay_prefix);
 
         // Location
         var wombat_location = new WombatLocation(window.self.location);
@@ -747,6 +788,9 @@ WB_wombat_init = (function() {
         // Ajax
         init_ajax_rewrite();
         init_worker_override();
+        
+        // Cookies
+        init_cookies_override();
 
         // DOM
         init_dom_override();

pywb/ui/frame_insert.html

@@ -3,7 +3,6 @@
 <!-- Start WB Insert -->
 <script>
   wbinfo = {}
-//  wbinfo.capture_str = "{{ timestamp | format_ts }}";
   wbinfo.capture_str = " ";
   wbinfo.is_embed = false;
   wbinfo.prefix = "{{ wbrequest.wb_prefix }}";
@@ -33,7 +32,7 @@ function make_inner_url(url, ts)
     }
 }
 
-function push_state(url, timestamp, capture_str) {
+function push_state(url, timestamp, capture_str, is_live) {
     var curr_href = null;
     
     if (window.frames[0].WB_wombat_location) {
@@ -41,7 +40,7 @@ function push_state(url, timestamp, capture_str) {
     }
         
     if (url != curr_href) {
-        update_status(capture_str);
+        update_status(capture_str, is_live);
         return;
     }
     
@@ -55,14 +54,15 @@ function push_state(url, timestamp, capture_str) {
     state.inner_url = make_inner_url(url, state.timestamp);
     state.url = url;
     state.capture_str = capture_str;
+    state.is_live = is_live;
     
     window.history.replaceState(state, "", state.outer_url);
 
-    update_status(state.capture_str);
+    update_status(state.capture_str, is_live);
 }
 
 function pop_state(state) {
-    update_status(state.capture_str);
+    update_status(state.capture_str, state.is_live);
     
     window.frames[0].src = state.outer_url;
 }
@@ -81,10 +81,19 @@ function extract_ts(url)
     return url.substring(inx + 1);
 }
 
-function update_status(str) {
+function update_status(str, is_live) {
-    var elem = document.getElementById("_wb_capture_info");
+    var capture_info = document.getElementById("_wb_capture_info");
-    if (elem) {
+    if (capture_info) {
-        elem.innerHTML = str;
+        capture_info.innerHTML = str;
+    }
+
+    var label = document.getElementById("_wb_label");
+    if (label) {
+        if (is_live) {
+            label.innerHTML = _wb_js.labels.LIVE_MSG;
+        } else {
+            label.innerHTML = _wb_js.labels.REPLAY_MSG;
+        }
     }
 }
 

pywb/ui/head_insert.html

@@ -18,6 +18,7 @@
   wbinfo.is_embed = {{"true" if wbrequest.wb_url.is_embed else "false"}};
   wbinfo.is_frame_mp = {{"true" if wbrequest.wb_url.mod == 'mp_' else "false"}}
   wbinfo.canon_url = "{{ canon_url }}";
+  wbinfo.is_live = {{ "true" if cdx.is_live else "false" }};
 </script>
 <script src='{{ wbrequest.host_prefix }}/{{ static_path }}/wb.js'> </script>
 <link rel='stylesheet' href='{{ wbrequest.host_prefix }}/{{ static_path }}/wb.css'/>

pywb/ui/index.html

@@ -9,3 +9,14 @@ The following archive collections are available:
   {% endif %} 
 {% endfor %}
 </ul>
+
+Other endpoints in this deployment:
+
+<ul>
+{% for route in routes %}
+  {% if not route | is_wb_handler %}
+  <li><b>{{ '/' + route.path }}</b> - {{ route | string }}</li>
+  {% endif %} 
+{% endfor %}
+</ul>
+

pywb/warc/archiveiterator.py

@@ -215,7 +215,8 @@ def create_record_iter(arcv_iter, options):
                  not append_post):
                 continue
 
-            elif (not include_all and record.content_type == 'application/warc-fields'):
+            elif (not include_all and
+                  record.content_type == 'application/warc-fields'):
                 continue
 
             entry = parse_warc_record(record)
@@ -226,7 +227,8 @@ def create_record_iter(arcv_iter, options):
             continue
 
         if entry.url and not entry.key:
-            entry.key = canonicalize(entry.url, options.get('surt_ordered', True))
+            entry.key = canonicalize(entry.url,
+                                     options.get('surt_ordered', True))
 
         compute_digest = False
 

pywb/warc/cdxindexer.py

@@ -77,6 +77,10 @@ class SortedCDXWriter(CDXWriter):
         return False
 
 
+#=================================================================
+ALLOWED_EXT = ('.arc', '.arc.gz', '.warc', '.warc.gz')
+
+
 #=================================================================
 def iter_file_or_dir(inputs):
     for input_ in inputs:
@@ -84,12 +88,13 @@ def iter_file_or_dir(inputs):
             yield input_, os.path.basename(input_)
         else:
             for filename in os.listdir(input_):
-                yield os.path.join(input_, filename), filename
+                if filename.endswith(ALLOWED_EXT):
+                    yield os.path.join(input_, filename), filename
 
 
 #=================================================================
 def remove_ext(filename):
-    for ext in ('.arc', '.arc.gz', '.warc', '.warc.gz'):
+    for ext in ALLOWED_EXT:
         if filename.endswith(ext):
             filename = filename[:-len(ext)]
             break

pywb/webapp/cdx_api_handler.py

@@ -25,7 +25,7 @@ class CDXAPIHandler(BaseHandler):
         return WbResponse.text_stream(cdx_iter)
 
     def __str__(self):
-        return 'CDX Handler: ' + str(self.index_handler)
+        return 'CDX Index Handler'
 
     @staticmethod
     def extract_params_from_wsgi_env(env):

pywb/webapp/handlers.py

@@ -8,37 +8,23 @@ from pywb.utils.loaders import BlockLoader
 from pywb.framework.basehandlers import BaseHandler, WbUrlHandler
 from pywb.framework.wbrequestresponse import WbResponse
 
+from pywb.warc.recordloader import ArcWarcRecordLoader
+from pywb.warc.resolvingloader import ResolvingLoader
+
+from views import J2TemplateView, add_env_globals
+from replay_views import ReplayView
+
 
 #=================================================================
-# Standard WB Handler
+class SearchPageWbUrlHandler(WbUrlHandler):
-#=================================================================
+    """
-class WBHandler(WbUrlHandler):
+    Loads a default search page html template to be shown when
-    def __init__(self, index_reader, replay,
+    the wb_url is empty
-                 search_view=None, config=None):
+    """
-
+    def __init__(self, config):
-        self.index_reader = index_reader
+        self.search_view = (J2TemplateView.
-
+                            create_template(config.get('search_html'),
-        self.replay = replay
+                           'Search Page'))
-
-        self.search_view = search_view
-
-    def __call__(self, wbrequest):
-        if wbrequest.wb_url_str == '/':
-            return self.render_search_page(wbrequest)
-
-        with PerfTimer(wbrequest.env.get('X_PERF'), 'query') as t:
-            response = self.index_reader.load_for_request(wbrequest)
-
-        if isinstance(response, WbResponse):
-            return response
-
-        cdx_lines = response[0]
-        cdx_callback = response[1]
-
-        with PerfTimer(wbrequest.env.get('X_PERF'), 'replay') as t:
-            return self.replay(wbrequest,
-                               cdx_lines,
-                               cdx_callback)
 
     def render_search_page(self, wbrequest, **kwargs):
         if self.search_view:
@@ -48,6 +34,67 @@ class WBHandler(WbUrlHandler):
         else:
             return WbResponse.text_response('No Lookup Url Specified')
 
+
+#=================================================================
+# Standard WB Handler
+#=================================================================
+class WBHandler(SearchPageWbUrlHandler):
+    def __init__(self, query_handler, config=None):
+        super(WBHandler, self).__init__(config)
+
+        self.index_reader = query_handler
+
+        cookie_maker = config.get('cookie_maker')
+        record_loader = ArcWarcRecordLoader(cookie_maker=cookie_maker)
+
+        paths = config.get('archive_paths')
+
+        resolving_loader = ResolvingLoader(paths=paths,
+                                           record_loader=record_loader)
+
+        template_globals = config.get('template_globals')
+        if template_globals:
+            add_env_globals(template_globals)
+
+        self.replay = ReplayView(resolving_loader, config)
+
+        self.fallback_handler = None
+        self.fallback_name = config.get('fallback')
+
+    def resolve_refs(self, handler_dict):
+        if self.fallback_name:
+            self.fallback_handler = handler_dict.get(self.fallback_name)
+
+    def __call__(self, wbrequest):
+        if wbrequest.wb_url_str == '/':
+            return self.render_search_page(wbrequest)
+
+        try:
+            with PerfTimer(wbrequest.env.get('X_PERF'), 'query') as t:
+                response = self.index_reader.load_for_request(wbrequest)
+        except NotFoundException as nfe:
+            return self.handle_not_found(wbrequest, nfe)
+
+        if isinstance(response, WbResponse):
+            return response
+
+        cdx_lines, cdx_callback = response
+        return self.handle_replay(wbrequest, cdx_lines, cdx_callback)
+
+    def handle_replay(self, wbrequest, cdx_lines, cdx_callback):
+        with PerfTimer(wbrequest.env.get('X_PERF'), 'replay') as t:
+            return self.replay(wbrequest,
+                               cdx_lines,
+                               cdx_callback)
+
+    def handle_not_found(self, wbrequest, nfe):
+        if (not self.fallback_handler or
+            wbrequest.wb_url.is_query() or
+            wbrequest.wb_url.is_identity):
+            raise
+
+        return self.fallback_handler(wbrequest)
+
     def __str__(self):
         return 'Web Archive Replay Handler'
 

pywb/webapp/live_rewrite_handler.py

@@ -2,23 +2,44 @@ from pywb.framework.basehandlers import WbUrlHandler
 from pywb.framework.wbrequestresponse import WbResponse
 from pywb.framework.archivalrouter import ArchivalRouter, Route
 
-from handlers import StaticHandler
+from handlers import StaticHandler, SearchPageWbUrlHandler
 
 from replay_views import RewriteLiveView
 
+from pywb.utils.wbexception import WbException
+
 
 #=================================================================
-class RewriteHandler(WbUrlHandler):
+class LiveResourceException(WbException):
-    def __init__(self, config=dict(framed_replay=True)):
+    def status(self):
+        return '400 Bad Live Resource'
+
+
+#=================================================================
+class RewriteHandler(SearchPageWbUrlHandler):
+    def __init__(self, config):
+        super(RewriteHandler, self).__init__(config)
         self.rewrite_view = RewriteLiveView(config)
 
     def __call__(self, wbrequest):
-        return self.rewrite_view(wbrequest)
+        if wbrequest.wb_url_str == '/':
+            return self.render_search_page(wbrequest)
+
+        try:
+            return self.rewrite_view(wbrequest)
+
+        except Exception as exc:
+            url = wbrequest.wb_url.url
+            msg = 'Could not load the url from the live web: ' + url
+            raise LiveResourceException(msg=msg, url=url)
+
+    def __str__(self):
+        return 'Live Web Rewrite Handler'
 
 
 #=================================================================
-def create_live_rewriter_app():
+def create_live_rewriter_app(config={}):
-    routes = [Route('rewrite', RewriteHandler()),
+    routes = [Route('rewrite', RewriteHandler(config)),
               Route('static/default', StaticHandler('pywb/static/'))
              ]
 

pywb/webapp/pywb_init.py

@@ -6,13 +6,10 @@ from pywb.framework.wbrequestresponse import WbRequest
 from pywb.framework.memento import MementoRequest
 from pywb.framework.basehandlers import BaseHandler
 
-from pywb.warc.recordloader import ArcWarcRecordLoader
+from views import J2TemplateView
-from pywb.warc.resolvingloader import ResolvingLoader
-
-from views import J2TemplateView, add_env_globals
 from views import J2HtmlCapturesView, HeadInsertView
 
-from replay_views import ReplayView
+from live_rewrite_handler import RewriteHandler
 
 from query_handler import QueryHandler
 from handlers import WBHandler
@@ -62,31 +59,10 @@ class DictChain:
 
 #=================================================================
 def create_wb_handler(query_handler, config):
-
-    cookie_maker = config.get('cookie_maker')
-    record_loader = ArcWarcRecordLoader(cookie_maker=cookie_maker)
-
-    paths = config.get('archive_paths')
-
-    resolving_loader = ResolvingLoader(paths=paths,
-                                       record_loader=record_loader)
-
-    template_globals = config.get('template_globals')
-    if template_globals:
-        add_env_globals(template_globals)
-
-    replayer = ReplayView(resolving_loader, config)
-
-    search_view = (J2TemplateView.
-                   create_template(config.get('search_html'),
-                                   'Search Page'))
-
     wb_handler_class = config.get('wb_handler_class', WBHandler)
 
     wb_handler = wb_handler_class(
         query_handler,
-        replayer,
-        search_view=search_view,
         config=config,
     )
 
@@ -94,23 +70,33 @@ def create_wb_handler(query_handler, config):
 
 
 #=================================================================
-def init_collection(value, config):
+def create_live_handler(config):
+    live_handler = RewriteHandler(config)
+    return live_handler
+
+
+#=================================================================
+def init_route_config(value, config):
     if isinstance(value, str):
-        value = {'index_paths': value}
+        value = dict(index_paths=value)
 
     route_config = DictChain(value, config)
+    return route_config
 
+
+#=================================================================
+def init_collection(route_config):
     ds_rules_file = route_config.get('domain_specific_rules', None)
 
     html_view = (J2HtmlCapturesView.
-                 create_template(config.get('query_html'),
+                 create_template(route_config.get('query_html'),
                                  'Captures Page'))
 
     query_handler = QueryHandler.init_from_config(route_config,
                                                   ds_rules_file,
                                                   html_view)
 
-    return route_config, query_handler
+    return query_handler
 
 
 #=================================================================
@@ -139,8 +125,8 @@ def create_cdx_server_app(passed_config):
     routes = []
 
     for name, value in collections.iteritems():
-        result = init_collection(value, config)
+        route_config = init_route_config(value, config)
-        route_config, query_handler = result
+        query_handler = init_collection(route_config)
 
         cdx_api_suffix = route_config.get('enable_cdx_api', True)
 
@@ -173,23 +159,32 @@ def create_wb_router(passed_config={}):
     else:
         request_class = WbRequest
 
-    #if config.get('use_lxml_parser', False):
+    # store live and replay handlers
-    #    use_lxml_parser()
+    handler_dict = {}
 
     for name, value in collections.iteritems():
-
         if isinstance(value, BaseHandler):
+            handler_dict[name] = value
             routes.append(Route(name, value))
             continue
 
-        result = init_collection(value, config)
+        route_config = init_route_config(value, config)
-        route_config, query_handler = result
+
+        if route_config.get('index_paths') == '$liveweb':
+            live = create_live_handler(route_config)
+            handler_dict[name] = live
+            routes.append(Route(name, live))
+            continue
+
+        query_handler = init_collection(route_config)
 
         wb_handler = create_wb_handler(
             query_handler=query_handler,
-            config=route_config
+            config=route_config,
         )
 
+        handler_dict[name] = wb_handler
+
         logging.debug('Adding Collection: ' + name)
 
         route_class = route_config.get('route_class', Route)
@@ -215,6 +210,12 @@ def create_wb_router(passed_config={}):
     for static_name, static_path in static_routes.iteritems():
         routes.append(Route(static_name, StaticHandler(static_path)))
 
+    # resolve any cross handler references
+    for route in routes:
+        if hasattr(route.handler, 'resolve_refs'):
+            route.handler.resolve_refs(handler_dict)
+
+
     # Check for new proxy mode!
     if config.get('enable_http_proxy', False):
         router = ProxyArchivalRouter
@@ -237,5 +238,7 @@ def create_wb_router(passed_config={}):
                                                  'Home Page'),
 
         error_view=J2TemplateView.create_template(config.get('error_html'),
-                                                 'Error Page')
+                                                 'Error Page'),
+
+        config=config
     )

pywb/webapp/replay_views.py

@@ -65,14 +65,15 @@ class BaseContentView(object):
     def __call__(self, wbrequest, *args):
         # render top level frame if in frame mode
         # (not supported in proxy mode)
-        if (self.is_frame_mode and
+        if (self.is_frame_mode and wbrequest.wb_url and
-            not wbrequest.is_proxy and
+            not wbrequest.wb_url.mod and
-            not wbrequest.wb_url.mod):
+            not wbrequest.options['is_proxy'] and
+            not wbrequest.options.get('is_timegate', False)):
 
             embed_url = wbrequest.wb_url.to_str(mod=self._mp_mod)
             timestamp = datetime_to_timestamp(datetime.datetime.utcnow())
             url = wbrequest.wb_url.url
-            ctype='text/html'
+            ctype = 'text/html'
 
             return self.frame_insert_view.render_response(embed_url=embed_url,
                                                           wbrequest=wbrequest,
@@ -88,7 +89,9 @@ class RewriteLiveView(BaseContentView):
     def __init__(self, config):
         super(RewriteLiveView, self).__init__(config)
 
-        self.rewriter = LiveRewriter(defmod=self._mp_mod)
+        default_proxy = config.get('proxyhostport')
+        self.rewriter = LiveRewriter(defmod=self._mp_mod,
+                                     default_proxy=default_proxy)
 
     def render_content(self, wbrequest, *args):
         head_insert_func = self.head_insert_view.create_insert_func(wbrequest)
@@ -257,12 +260,10 @@ class ReplayView(BaseContentView):
         return content
 
     def _redirect_if_needed(self, wbrequest, cdx):
-        if wbrequest.is_proxy:
+        if wbrequest.options['is_proxy']:
             return None
 
-        # todo: generalize this?
+        redir_needed = (wbrequest.options.get('is_timegate', False))
-        redir_needed = (hasattr(wbrequest, 'is_timegate') and
-                        wbrequest.is_timegate)
 
         if not redir_needed and self.redir_to_exact:
             redir_needed = (cdx['timestamp'] != wbrequest.wb_url.timestamp)

pywb/webapp/views.py

@@ -2,8 +2,6 @@ from pywb.utils.timeutils import timestamp_to_datetime
 from pywb.framework.wbrequestresponse import WbResponse
 from pywb.framework.memento import make_timemap, LINK_FORMAT
 
-from handlers import WBHandler
-
 import urlparse
 import logging
 
@@ -62,7 +60,8 @@ def is_wb_handler(obj):
     if not hasattr(obj, 'handler'):
         return False
 
-    return isinstance(obj.handler, WBHandler)
+    #return isinstance(obj.handler, WBHandler)
+    return obj.handler.__class__.__name__ == "WBHandler"
 
 
 #=================================================================

sample_archive/text_content/toptest.js

@@ -0,0 +1 @@
+!function(){top!==window&&(alert("For security reasons, framing is not allowed."),top.location.replace(document.location))}

setup.py

@@ -34,7 +34,7 @@ class PyTest(TestCommand):
 
 setup(
     name='pywb',
-    version='0.4.7',
+    version='0.5.0',
     url='https://github.com/ikreymer/pywb',
     author='Ilya Kreymer',
     author_email='ikreymer@gmail.com',

tests/test_config.yaml

@@ -17,6 +17,10 @@ collections:
         index_paths: './sample_archive/cdx/'
         filters: ['filename:dupe*']
     
+    pywb-filt-2:
+        index_paths: './sample_archive/cdx/'
+        filters: ['!filename:dupe*']
+
     pywb-nonframe:
         index_paths: './sample_archive/cdx/'
         framed_replay: false
@@ -26,6 +30,14 @@ collections:
         index_paths: './sample_archive/non-surt-cdx/'
         surt_ordered: false
 
+    # live collection
+    live: $liveweb
+
+    # coll with fallback
+    pywb-fallback:
+        index_paths: ./sample_archive/cdx/
+        fallback: live
+
 
 # indicate if cdx files are sorted by SURT keys -- eg: com,example)/
 # SURT keys are recommended for future indices, but non-SURT cdxs
@@ -94,6 +106,12 @@ static_routes:
 # Enable simple http proxy mode
 enable_http_proxy: true
 
+# Additional proxy options (defaults)
+proxy_options:
+    use_default_coll: true
+
+    unaltered_replay: false
+
 # enable cdx server api for querying cdx directly (experimental)
 #enable_cdx_api: True
 # or specify suffix

tests/test_config_memento.yaml

@@ -14,7 +14,12 @@ enable_memento: true
 # Enable simple http proxy mode
 enable_http_proxy: true
 
+# test unaltered replay for proxy as well
+proxy_options:
+    unaltered_replay: true
+
 # enable cdx server api for timemap
 enable_cdx_api: true
 
-
+# test memento with framed replay
+framed_replay: true

tests/test_integration.py

@@ -1,5 +1,6 @@
 from pytest import raises
 import webtest
+import base64
 from pywb.webapp.pywb_init import create_wb_router
 from pywb.framework.wsgi_wrappers import init_app
 from pywb.cdx.cdxobject import CDXObject
@@ -258,6 +259,14 @@ class TestWb:
         resp = self.testapp.get('/pywb/20140603030351mp_/http://example.com?example=3', status = 503)
         assert resp.status_int == 503
 
+    def test_live_frame(self):
+        resp = self.testapp.get('/live/mp_/http://example.com/?test=test')
+        assert resp.status_int == 200
+
+    def test_live_fallback(self):
+        resp = self.testapp.get('/pywb-fallback/mp_/http://example.com/?test=test')
+        assert resp.status_int == 200
+
     def test_post_1(self):
         resp = self.testapp.post('/pywb/mp_/httpbin.org/post', {'foo': 'bar', 'test': 'abc'})
 
@@ -317,6 +326,50 @@ class TestWb:
         assert 'Sun, Jan 26 2014 20:11:27' in resp.body
         assert 'wb.js' in resp.body
 
+    def test_proxy_replay_auth_filtered(self):
+        headers = [('Proxy-Authorization', 'Basic ' + base64.b64encode('pywb-filt-2:'))]
+        resp = self.testapp.get('/x-ignore-this-x', headers = headers,
+                                extra_environ = dict(REQUEST_URI = 'http://www.iana.org/', SCRIPT_NAME = ''))
+
+        self._assert_basic_html(resp)
+
+        assert 'Sun, Jan 26 2014 20:06:24' in resp.body
+        assert 'wb.js' in resp.body
+
+    def test_proxy_replay_auth(self):
+        headers = [('Proxy-Authorization', 'Basic ' + base64.b64encode('pywb'))]
+        resp = self.testapp.get('/x-ignore-this-x', headers = headers,
+                                extra_environ = dict(REQUEST_URI = 'http://www.iana.org/', SCRIPT_NAME = ''))
+
+        self._assert_basic_html(resp)
+
+        assert 'Mon, Jan 27 2014 17:12:38' in resp.body
+        assert 'wb.js' in resp.body
+
+    def test_proxy_replay_auth_no_coll(self):
+        headers = [('Proxy-Authorization', 'Basic ' + base64.b64encode('no-such-coll'))]
+        resp = self.testapp.get('/x-ignore-this-x', headers = headers,
+                                extra_environ = dict(REQUEST_URI = 'http://www.iana.org/', SCRIPT_NAME = ''),
+                                status=407)
+
+        assert resp.status_int == 407
+
+    def test_proxy_replay_auth_invalid_1(self):
+        headers = [('Proxy-Authorization', 'abc' + base64.b64encode('no-such-coll'))]
+        resp = self.testapp.get('/x-ignore-this-x', headers = headers,
+                                extra_environ = dict(REQUEST_URI = 'http://www.iana.org/', SCRIPT_NAME = ''),
+                                status=407)
+
+        assert resp.status_int == 407
+
+    def test_proxy_replay_auth_invalid_2(self):
+        headers = [('Proxy-Authorization', 'basic')]
+        resp = self.testapp.get('/x-ignore-this-x', headers = headers,
+                                extra_environ = dict(REQUEST_URI = 'http://www.iana.org/', SCRIPT_NAME = ''),
+                                status=407)
+
+        assert resp.status_int == 407
+
     def test_proxy_pac(self):
         resp = self.testapp.get('/proxy.pac', extra_environ = dict(SERVER_NAME='pywb-proxy', SERVER_PORT='8080'))
         assert resp.content_type == 'application/x-ns-proxy-autoconfig'

tests/test_live_rewriter.py

@@ -4,7 +4,8 @@ import webtest
 
 class TestLiveRewriter:
     def setup(self):
-        self.app = init_app(create_live_rewriter_app, load_yaml=False)
+        self.app = init_app(create_live_rewriter_app, load_yaml=False,
+                            config=dict(framed_replay=True))
         self.testapp = webtest.TestApp(self.app)
 
     def test_live_rewrite_1(self):
@@ -22,4 +23,12 @@ class TestLiveRewriter:
         assert '<iframe ' in resp.body
         assert 'src="/rewrite/mp_/http://example.com/"' in resp.body
 
+    def test_live_invalid(self):
+        resp = self.testapp.get('/rewrite/mp_/http://abcdef', status=400)
+        assert resp.status_int == 400
+
+    def test_live_invalid_2(self):
+        resp = self.testapp.get('/rewrite/mp_/@#$@#$', status=400)
+        assert resp.status_int == 400
+
 

tests/test_memento.py

@@ -34,7 +34,7 @@ class TestWb:
         """
         TimeGate with no Accept-Datetime header
         """
-        resp = self.testapp.get('/pywb/http://www.iana.org/_css/2013.1/screen.css')
+        resp = self.testapp.get('/pywb/mp_/http://www.iana.org/_css/2013.1/screen.css')
 
         assert resp.status_int == 302
 
@@ -46,7 +46,7 @@ class TestWb:
 
         assert MEMENTO_DATETIME not in resp.headers
 
-        assert '/pywb/20140127171239/http://www.iana.org/_css/2013.1/screen.css' in resp.headers['Location']
+        assert '/pywb/20140127171239mp_/http://www.iana.org/_css/2013.1/screen.css' in resp.headers['Location']
 
 
     def test_timegate_accept_datetime(self):
@@ -54,7 +54,7 @@ class TestWb:
         TimeGate with Accept-Datetime header
         """
         headers = {ACCEPT_DATETIME: 'Sun, 26 Jan 2014 20:08:04'}
-        resp = self.testapp.get('/pywb/http://www.iana.org/_css/2013.1/screen.css', headers=headers)
+        resp = self.testapp.get('/pywb/mp_/http://www.iana.org/_css/2013.1/screen.css', headers=headers)
 
         assert resp.status_int == 302
 
@@ -67,7 +67,7 @@ class TestWb:
 
         assert MEMENTO_DATETIME not in resp.headers
 
-        assert '/pywb/20140126200804/http://www.iana.org/_css/2013.1/screen.css' in resp.headers['Location']
+        assert '/pywb/20140126200804mp_/http://www.iana.org/_css/2013.1/screen.css' in resp.headers['Location']
 
 
     def test_non_timegate_intermediate_redir(self):
@@ -76,7 +76,7 @@ class TestWb:
         """
         headers = {ACCEPT_DATETIME: 'Sun, 26 Jan 2014 20:08:04'}
         # not a timegate, partial timestamp /2014/ present
-        resp = self.testapp.get('/pywb/2014/http://www.iana.org/_css/2013.1/screen.css', headers=headers)
+        resp = self.testapp.get('/pywb/2014mp_/http://www.iana.org/_css/2013.1/screen.css', headers=headers)
 
         assert resp.status_int == 302
 
@@ -90,14 +90,14 @@ class TestWb:
 
 
         # redirect to latest, not negotiation via Accept-Datetime
-        assert '/pywb/20140127171239/' in resp.headers['Location']
+        assert '/pywb/20140127171239mp_/' in resp.headers['Location']
 
 
     def test_memento_url(self):
         """
         Memento response, 200 capture
         """
-        resp = self.testapp.get('/pywb/20140126200804/http://www.iana.org/_css/2013.1/screen.css')
+        resp = self.testapp.get('/pywb/20140126200804mp_/http://www.iana.org/_css/2013.1/screen.css')
 
         assert resp.status_int == 200
 
@@ -105,7 +105,7 @@ class TestWb:
 
         links = self.get_links(resp)
         assert '<http://www.iana.org/_css/2013.1/screen.css>; rel="original"' in links
-        assert '<http://localhost:80/pywb/http://www.iana.org/_css/2013.1/screen.css>; rel="timegate"' in links
+        assert '<http://localhost:80/pywb/mp_/http://www.iana.org/_css/2013.1/screen.css>; rel="timegate"' in links
         assert self.make_timemap_link('http://www.iana.org/_css/2013.1/screen.css') in links
 
         assert resp.headers[MEMENTO_DATETIME] == 'Sun, 26 Jan 2014 20:08:04 GMT'
@@ -115,7 +115,7 @@ class TestWb:
         """
         Memento (capture) of a 302 response
         """
-        resp = self.testapp.get('/pywb/20140128051539/http://www.iana.org/domains/example')
+        resp = self.testapp.get('/pywb/20140128051539mp_/http://www.iana.org/domains/example')
 
         assert resp.status_int == 302
 
@@ -123,7 +123,7 @@ class TestWb:
 
         links = self.get_links(resp)
         assert '<http://www.iana.org/domains/example>; rel="original"' in links
-        assert '<http://localhost:80/pywb/http://www.iana.org/domains/example>; rel="timegate"' in links
+        assert '<http://localhost:80/pywb/mp_/http://www.iana.org/domains/example>; rel="timegate"' in links
         assert self.make_timemap_link('http://www.iana.org/domains/example') in links
 
         assert resp.headers[MEMENTO_DATETIME] == 'Tue, 28 Jan 2014 05:15:39 GMT'
@@ -147,12 +147,12 @@ rel="self"; type="application/link-format"; from="Fri, 03 Jan 2014 03:03:21 GMT"
 
         assert lines[1] == '<http://example.com?example=1>; rel="original",'
 
-        assert lines[2] == '<http://localhost:80/pywb/http://example.com?example=1>; rel="timegate",'
+        assert lines[2] == '<http://localhost:80/pywb/mp_/http://example.com?example=1>; rel="timegate",'
 
-        assert lines[3] == '<http://localhost:80/pywb/20140103030321/http://example.com?example=1>; \
+        assert lines[3] == '<http://localhost:80/pywb/20140103030321mp_/http://example.com?example=1>; \
 rel="memento"; datetime="Fri, 03 Jan 2014 03:03:21 GMT",'
 
-        assert lines[4] == '<http://localhost:80/pywb/20140103030341/http://example.com?example=1>; \
+        assert lines[4] == '<http://localhost:80/pywb/20140103030341mp_/http://example.com?example=1>; \
 rel="memento"; datetime="Fri, 03 Jan 2014 03:03:41 GMT"'
 
     def test_timemap_2(self):