![]() |
|
Register | FAQ | The Twelve Commandments | Members List | Calendar | Arcade | Find the Best VPN | Today's Posts | Search |
Development Open Discussion Anything else about the Phex development |
![]() |
| LinkBack | Thread Tools | Display Modes |
| |||
![]() Hi all, I've trying to find specific information about the features that Phex uses but I haven't been able to find anything. * Which gnutella version is phex implementing. As I found there are different specification. * Which search mecanism it uses. Looking at the source code I've found it uses Dynamic Queries, I just want to make sure here. And is it the only mechanism used? * I've been looking for a way to do a hashcode search programatically using the phex source code. I haven't found anything, can anyone point me in the right direction. * It would be great to have a firewall to firewall transfer, which I see is in the roadmap (http://www.phex.org/wiki/index.php/Roadmap). The question here is if there is an estimated time for when the development on this feature will start and how long will it take? I see there's also a feature to support the limewire's mojito DHT which may help with the firewall to firewall transfer. Is there any estimated time for this as well? * Is there a way that when bootstraping the phex client, it will find the clients in the local area network. I suppose it would by doing a multicast or broadcast in the network to identify which phex peers are on the same network. This will help in doing downloads faster if the file being wanted to download is within the same network. * I have found some topics related to this. How would be the best way to manage fake results when these fakes constantly change their host or their sha1 hashcodes? * Does phex have a limit handling large files? * Are there any plans on implementing UPnP and NatPMP?? |
| |||||
![]() Quote:
Quote:
Quote:
Check out ResearchSettings.startSearch() Quote:
Quote:
Yes: 2^63 - 1 bytes ~ 8 exabytes Yes, once we found someone who would like to help. |
| ||||
![]() Hi, Welcome to the Gnutella Forums! I’ll directly come to your questions: * Phex implements Gnutella 0.6 with almost all extensions except the DHT. * The search mechanism is dynamic querying on the requester side. Also it supports Query Routing (via QRT) on the forwarding part. * Hash-searching can only be done efficiently using the DHT which isn’t yet implemented. So you can only search for the filename and then match the returned hashes. * The RoadMap progressing slowly right now – job vs. hoppy. * Currently Phex gets all download-sources via the download mesh. I don’t know if there are additions for local broadcasts. But it would be nice, I think. * For fake files you can either check for some common names (i.e. weed out all *.qt files for they can phone home), check for the host and ban it (only works if the fake files are intended to be downloaded and not just an attack on the network structure) or implement something like credence for Phex: Credence: Thwarting P2P Pollution * I don’t know a limit for the file size. * See RoadMap info… For what do you want to use it? EDIT: Posting collision ![]()
__________________ ![]() -> put this banner into your own signature! <- -- Erst im Spiel lebt der Mensch. Nur ludantaj homoj vivas. GnuFU.net - Gnutella For Users Draketo.de - Shortstories, Poems, Music and strange Ideas. Last edited by arne_bab; April 8th, 2010 at 11:13 PM. |
| ||||||||
![]() Thanks GregorK and arne_bab for the quick response. I have more questions regarding the answers Quote:
Quote:
Quote:
Quote:
Quote:
I guess this have something to do with multicast as well, which I imagine would be the mechanism of the LAN discovery. As a resume, we would be glad to help. I think there are two main topics here: mojito DHT and multicast. About mojito DHT, it will be a task that will take time. Having this we can use it as base for the firewall-to-firewall. About multicast, it will help with the LAN discovery and with the upnp and natpmp. Which I guess will be less time consuming than DHT Quote:
Quote:
Quote:
Actually I've done that ![]() We want to use it as a gnutella API. Basically for download audio files and do keyword searches. |
| ||||
![]() The download mesh is an extension to http used by Gnutella. Basically all downloaders and uploaders exchange the sources they know (just like a completely decentral BitTorrent server). The headers are X-Alt (IPs+ports of other down- and uploaders) and X-NAlt (non working IPs+ports for preventing spam and getting offline nodes out of the mesh). As soon as you've seen a download going from 5 sources to 400 sources in half a minute, you know its efficiency ![]() Quote:
This had the potential of stopping spam once and for all, so I thought it might interest you. Quote:
Quote:
If I can do anything to help, please tell me. I’m no Java coder, but I have some background knowledge from supporting Phex users in here and discussing in the_gdf (the Gnutella development forum). The current discussion in the GDF might also be interesting to you: More intelligent searches via a GGEP extension: "a AND (b OR c)" → the_gdf : The Gnutella Developer Forum (GDF) → Yahoo! Groups
__________________ ![]() -> put this banner into your own signature! <- -- Erst im Spiel lebt der Mensch. Nur ludantaj homoj vivas. GnuFU.net - Gnutella For Users Draketo.de - Shortstories, Poems, Music and strange Ideas. |
![]() |
| |