3 - serialization code (download)
4 - linking of downloads to searches (expose opaque struct SearchResult;
5 allow starting download based on search result (new API):
6 => have meta data for instant completion!
8 => linking of download to search
9 => expose link to search result in download events (including search result's
11 - generate SUSPEND events (publish, unindex, search, download)
12 - fix 'start_time' time stamps (search, publish, unindex, download)
13 - actually call 'sync' functions (publish, unindex, search, download)
15 => refactor fs.c to join common code segments!
16 => document directory structure
17 => ensure all files & dirs are cleaned up! (at least during 'clean' runs)
18 - persistence testing (publish, unindex, search, download):
20 => schedule suspending tasks DURING event handler => good coverage!
21 - gnunet-service-fs (hot-path routing, load-based routing, nitpicks)
22 - [gnunet-service-fs.c:208]: member 'LocalGetContext::results_bf_size' is never used
23 - [gnunet-service-fs.c:501]: member 'PendingRequest::used_pids_size' is never used
24 - [gnunet-service-fs.c:654]: member 'ConnectedPeer::last_client_replies' is never used
25 - [gnunet-service-fs.c:669]: member 'ConnectedPeer::avg_delay' is never used
26 - [gnunet-service-fs.c:675]: member 'ConnectedPeer::avg_priority' is never used
27 - [gnunet-service-fs.c:688]: member 'ConnectedPeer::pending_requests' is never used
28 - [gnunet-service-fs.c:694]: member 'ConnectedPeer::last_p2p_replies_woff' is never used
29 - [gnunet-service-fs.c:700]: member 'ConnectedPeer::last_client_replies_woff' is never used
31 + active reply route caching design & implementation of service; gap extension!
33 - on-demand encoding => move logic to block-library!?
34 - peer selection => how to consider latency/bw/etc?
35 - content transmission => how often the same block?
36 - how to select delay before next migration?
39 - integrate with FS or not? (peer list, index/on-demand encoding, block code,
40 inbound priority assignment; all would be easier with tight integration!)
42 - good to have for transport/DV evaluation!
44 - write DV API (need to move declarations from dv_api.c to gnunet_dv_service.h!)
45 - implement DV service
46 - implement DV library (looks done)
47 - implement DV transport plugin
49 - implement performance tests (needs tbench)
51 - needs more testing (especially F2F topology) & transport blacklisting
53 - only connect() sockets that are ready (select()) [Nils]
54 [On W32, we need to select after calling socket before doing connect etc.]
56 - use g_main_context_set_poll_func to integrate GTK with GNUnet Scheduler!? (YUCK!)
57 - OR: add scheduler API to enable integration with GTK main loop instead of doing our own select
58 - use g_main_context_pending, g_main_context_query / g_main_context_check / g_main_context_dispatch
59 and NEVER g_main_loop_run (can this be done? might be the clean way to do this! But how
60 to integrate this with "gtk_main"? Docu says:
61 "It's OK to use the GLib main loop directly instead of gtk_main(), though it involves
62 slightly more typing. See GMainLoop in the GLib documentation."
63 => so maybe it "just works"?
65 - design & implement new setup tool
69 - good to have for DV/DHT evaluation!
71 - implement DHT service
73 - implement performance tests
77 - need to get rid of synchronous API for service starts (cause all kinds of problems)
78 [=> eliminate for need to tell ARM about service starts most of the time!] [Safey]
79 - better tracking of which config changes actually need to cause process restarts by ARM.
80 - listen for requests to discover dependencies between services (and avoid
81 having to explicitly program start requests)
82 - better crash management (attach debugging support, capture and analyze
83 debug output, detect random vs. deterministic crashes)
86 - datastore reservation (publishing)
87 - location URIs (publish, search, download)
88 - utilize in-line files in meta data always (including in search results or
89 when download is triggered manually and for probes); currently the data is
90 only used when users do a general 'recursive' download
91 - non-anonymous FS service (needs DHT)
92 + DHT integration for search
93 + CS-DHT-functions (DHT-put of LOC)
94 + P2P-functions (DHT-get)
95 - collection API & tests
96 + gnunet-pseudonym (collection support)
97 - implement FS performance tests
103 - improved content selection (not just 'get_random')
106 * Determine RC bugs and fix those!
108 - modify configuration to allow controlling connections for non-local starts
109 - testbed creation with topology (needs working F2F topology)
111 - implement testcases for distributed testing
112 - test basic peer re-configure
113 - test topology creation
114 - test churn generation
115 - consider changing API for peer-group termination to
116 call continuation when done
118 - finalize API design
121 - integration with transport service
122 * MYSQL database backends: [CG]
126 - reconstruct IBLOCKS from DBLOCKS if possible (during download; see FIXME in fs_download)
131 - expand bibliography
132 - convert documentation pages to books
133 - update books (especially for developers)
134 - create good Drupal theme for GNUnet
135 - make a NICE download page and figure out how to enable developers to publish TGZs nicely
136 - port "contact" page
137 - add content type for "todo" items?
138 * POSTGRES database backends: [CG]
141 * Determine RC bugs and fix those!
145 - SMTP transport backend
146 - HTTPS transport backend
147 + improved HTTPS support in MHD
149 - Implement method of learning our external addresses from
150 other peers; need some kind of threshold-based
151 scheme, limiting both the total number of addresses that we accept
152 this way as well as requiring multiple confirmations; also, we
153 should possibly try to confirm that the given address works for
154 us ourselves (loopback-style) before adding it to the list
155 + we may be able to simplify WELCOME messages (no need to add
156 addresses there anymore, but may help to learn them there anyway...).
157 + we probably want some kind of voting/counting for learning IP addresses
158 (maybe including IP addresses in ads proportional to how often others
159 report them? we at least need some protection against >64k HELLOs!),
160 + provide a way to give the user a list of "learned" IP addresses and
161 a way to easily "veto" addresses off the list!
162 => If MiM attacker uses vetoed address, blacklist the specific IP for
163 the presumed neighbour!
164 - implement gnunet-transport (transport configurator / tester)
165 - UPnP-based IP detection
166 (Note: build library always, build service when libxml2/etc. are available)
168 - Remove KBlocks in gnunet-unindex (see discussion with Kenneth Almquist on gnunet-devs in 9/2009)
170 - expire 'ancient' HELLOs (those without valid addresses AND that
171 we have not 'used' (for their public keys) in a while; need a way
172 to track actual 'use')
173 - make sue we also trigger notifications whenever HELLOs expire
180 - should use hash map to look up sessions
182 - should use BIO instead of mmap
184 - need to periodically probe latency/transport cost changes & possibly switch transport
185 - should use hash map to look up Neighbours (service AND plugins!)
187 - check for duplicates on insertion (currently, same content is frequently
188 stored again [seen with KBLOCKS and SBLOCKS]!)
190 - merge multiple HELLOs of the same peer in the transmission queue
191 (theoretically reduces overhead; bounds message queue size)
192 - merge multiple iteration requests over "all" peers in the queue
193 (theoretically reduces overhead; bounds messgae queue size)
195 - use different queue prioritization for probe-downloads vs. normal downloads (!?)
199 - repeatedly resolve hostname and look up interfaces to determine our own IP
200 - [./transport/plugin_transport_tcp.c:391]: (style) struct or union member 'Plugin::address_update_task' is never used (related to issue above)
202 - [./transport/gnunet-service-transport.c:173]: (style) struct or union member 'TransportPlugin::rebuild' is never used (related to TCP not refreshing external addresses?)
204 - testcase would be nice...