| *** tpb has joined #tp | 00:00 | |
| *** ChanServ sets mode: +o tpb | 00:00 | |
| *** stupidincarnate has quit IRC | 00:18 | |
| *** alanp has quit IRC | 01:23 | |
| *** tuna-fish has quit IRC | 01:56 | |
| *** Epyon has quit IRC | 02:53 | |
| *** Epyon has joined #tp | 03:10 | |
| *** StupidIncarnate has joined #tp | 03:54 | |
| *** glew has quit IRC | 04:04 | |
| *** Tanoku_ has quit IRC | 04:20 | |
| *** Tanoku_ has joined #tp | 04:20 | |
| StupidIncarnate | mithro around? | 06:10 |
|---|---|---|
| mithro | yes | 06:10 |
| StupidIncarnate | What's the guideline for using other images for the clients? Just that they're free of license kind of thing? | 06:13 |
| mithro | what do you mean? | 06:13 |
| mithro | most of the images are determined by the server in tp04 | 06:13 |
| StupidIncarnate | tpweb determines them based on the object types | 06:14 |
| StupidIncarnate | is that something that needs to change then? | 06:14 |
| StupidIncarnate | I mean like the image icons for planets and whatnot | 06:14 |
| StupidIncarnate | actually, I should think about it more. Nevermind | 06:18 |
| *** llnz has joined #tp | 06:22 | |
| *** tuna-fish has joined #tp | 06:23 | |
| mithro | hey llnz | 06:31 |
| llnz | hi mithro | 06:32 |
| mithro | had a chance to look into that tpserver-cpp bug? | 06:32 |
| llnz | yes, note added to bug report | 06:33 |
| llnz | going to try to solve it soon | 06:33 |
| *** StupidIncarnate has quit IRC | 07:31 | |
| mithro | we really should do a a GSoC announcement sometime | 07:56 |
| *** Epyon has quit IRC | 08:36 | |
| llnz | mithro: yeah | 08:47 |
| *** null_000 has joined #tp | 09:01 | |
| null_000 | hey people | 09:02 |
| null_000 | (and bots) | 09:02 |
| *** cahirwpz has joined #tp | 09:09 | |
| mithro | howdy null_000 and cahirwpz | 09:20 |
| cahirwpz | hi mithro :) | 09:20 |
| null_000 | g'day mithro | 09:21 |
| mithro | cahirwpz: how goes the research | 09:22 |
| mithro | ? | 09:22 |
| cahirwpz | I've been busy lately - only came to conclusion that I need to start from tp.netlib.{common,server,twisted} and tp.server.server | 09:24 |
| cahirwpz | code there is quite straightforward, but I still need to read more about the protocol | 09:25 |
| *** Agon-phone has joined #tp | 09:49 | |
| llnz | odd | 09:52 |
| llnz | can't find the source of the bug | 09:52 |
| * llnz wanders off | 09:57 | |
| llnz | later all | 09:57 |
| *** llnz has quit IRC | 09:57 | |
| *** Epyon has joined #tp | 10:06 | |
| mithro | cahirwpz: the design of tpserver-py is relatively simple | 10:11 |
| mithro | it's really just a thin translation layer between Thousand Parsec frames and the database | 10:11 |
| *** null_000 has quit IRC | 10:19 | |
| *** Agon-laptop has joined #tp | 10:27 | |
| *** Agon-phone has quit IRC | 10:28 | |
| cahirwpz | mithro: we don't have any tests for tpserver-py, do we ? | 10:29 |
| mithro | none at all sadly | 10:30 |
| mithro | tpserver-py was originally not suppose to stick around :P | 10:30 |
| cahirwpz | heh | 10:30 |
| mithro | but it's kinda cool and it has an interesting design compared to tpserver-cpp | 10:31 |
| cahirwpz | did you know that I and epyon know each other? | 10:33 |
| cahirwpz | we met like 6 years ago at the university :) | 10:35 |
| mithro | cahirwpz: no, we did not | 10:35 |
| mithro | we won't hold it against you :P | 10:35 |
| cahirwpz | it's like tp servers takeover ;P | 10:36 |
| mithro | cahirwpz: tests are a good thing to add as you go :) | 10:39 |
| mithro | don't let my bad habits rub off :) | 10:40 |
| *** verhoevenv has joined #tp | 10:40 | |
| cahirwpz | are there any timeouts defined for TP protocol - like how long are we going to wait for single frame arrival ? | 10:44 |
| mithro | cahirwpz: no | 10:44 |
| *** jmtan has joined #tp | 11:39 | |
| *** Epyon has quit IRC | 11:51 | |
| *** verhoevenv has quit IRC | 12:06 | |
| *** Epyon has joined #tp | 12:37 | |
| *** null_000 has joined #tp | 12:57 | |
| mithro | null_000: still stuck? | 13:21 |
| null_000 | yes and no | 13:23 |
| null_000 | didn't fix the bug... but I made all the python helper functions I will need | 13:24 |
| null_000 | so the time wasn't wasted | 13:24 |
| null_000 | and I found a new but... probably in daneel-ai | 13:25 |
| null_000 | issue 96 | 13:27 |
| cahirwpz | mithro: ok, I wrote a simple test client using twisted | 13:43 |
| mithro | test client? | 13:43 |
| cahirwpz | i'm using libtpproto2-py library to generate messages | 13:43 |
| cahirwpz | objects.Connect( 1, "tptests-py client" ) - generates a packet | 13:44 |
| Epyon | cahirwpz, how about setting a common testing framework for both servers? | 13:44 |
| Epyon | *evil smile* | 13:44 |
| cahirwpz | history likes to make loops ;-) | 13:44 |
| * Epyon smiles | 13:45 | |
| cahirwpz | mithro: i suppose that pack() method on generated packet should be used to serialize it | 13:45 |
| mithro | cahirwpz: in theory yes | 13:46 |
| *** verhoevenv has joined #tp | 13:46 | |
| cahirwpz | mithro: generated packet seems ok - 5450303300000001000000030000000000000011747074657374732d707920636c69656e74 | 13:48 |
| mithro | cahirwpz: well that is good to hear | 13:48 |
| cahirwpz | mithro: well... not quite... server crashes after receiving it | 13:49 |
| mithro | tpserver-cpp for tpserver-py? | 13:49 |
| cahirwpz | py | 13:49 |
| mithro | and which protocol version are you using? | 13:49 |
| cahirwpz | tp03 | 13:49 |
| Epyon | tried sending it to -cpp? | 13:50 |
| cahirwpz | Unknown packet type 1937011501 | 13:50 |
| cahirwpz | nope | 13:50 |
| mithro | sounds like the packet isn't quite right, you compared it to the one from libtpproto-py? | 13:51 |
| cahirwpz | don't know yet how to construct a similar packet using libtpproto-py | 13:52 |
| *** tansell-laptop has quit IRC | 13:54 | |
| cahirwpz | mithro: can you give me a clue ? | 13:54 |
| mithro | should be in the same way in fact | 13:54 |
| cahirwpz | ok - i see... module tp.netlib.objects | 13:57 |
| cahirwpz | how about serializing? | 13:57 |
| cahirwpz | nvm - converting to string works | 13:59 |
| cahirwpz | ok - now I get an answer from tpserver-py | 13:59 |
| cahirwpz | proper request looks as following: 545030330000000100000000000000190000001557656c636f6d6520746f2070792d73657276657221 | 14:00 |
| cahirwpz | sorry - this was a response | 14:00 |
| cahirwpz | request is 5450303300000001000000030000001500000011747074657374732d707920636c69656e74 | 14:00 |
| cahirwpz | mithro: correct me if I'm wrong - payload length is missing in packet generated from libtpproto2-py ? | 14:02 |
| mithro | cahirwpz: I'm afraid I can't decode packets like that in my head | 14:02 |
| mithro | plus I'm not sure the format you are using there, it's certainly not format that libtpproto-py prints out | 14:03 |
| cahirwpz | this is encoded packet printed out using following method: print packet.encode("hex") | 14:07 |
| cahirwpz | what does exactly following message mean "Skipping Header because it's pre-written." | 14:07 |
| mithro | cahirwpz: if you print using hexbyte in xstruct | 14:08 |
| mithro | it'll be easier to read | 14:08 |
| cahirwpz | any example how to use it ? | 14:09 |
| mithro | print xstruct.hexbyte(str) | 14:09 |
| cahirwpz | thx | 14:09 |
| cahirwpz | wrong packet: TP03\x00\x00\x00\x01\x00\x00\x00\x03\x00\x00\x00\x00\x00\x00\x00\x11tptests\x2dpy client | 14:11 |
| cahirwpz | good packet: TP03\x00\x00\x00\x01\x00\x00\x00\x03\x00\x00\x00\x15\x00\x00\x00\x11tptests\x2dpy client | 14:12 |
| mithro | cahirwpz: yes, looks like length is not being set correctly | 14:12 |
| cahirwpz | and what about the message about header ? | 14:14 |
| mithro | that message is because header is hard-coded rather then being auto-generated via the xml | 14:15 |
| mithro | looks like we are missing a test which checks for the length | 14:16 |
| cahirwpz | I suppose this is caused by serializing header and payload at the same time in libtpproto2-py... | 14:25 |
| cahirwpz | quick fix for that would be: | 14:28 |
| cahirwpz | binary = ''.join( [binary[:12], struct.pack( "!L", len(binary) - 16 ), binary[16:]] ) | 14:28 |
| *** Epyon has quit IRC | 14:29 | |
| mithro | cahirwpz: I'm not sure why it length is not being set | 14:30 |
| mithro | o.length seems to be right | 14:31 |
| cahirwpz | hmm... are you able to tell what's the length of payload before you serialize a packet ? | 14:34 |
| mithro | yes | 14:35 |
| mithro | ~seen cherez | 14:36 |
| tpb | mithro: cherez was last seen in #tp 35 weeks, 6 days, 12 hours, 2 minutes, and 27 seconds ago: <cherez> llnz: Nothing, I just noticed you were on CC to e-mails I get from Google. | 14:36 |
| *** Agon-phone has joined #tp | 14:38 | |
| *** Agon-laptop has quit IRC | 14:38 | |
| mithro | ahh found the problem | 14:40 |
| cahirwpz | mithro: how about decoding packet received from a server using libtpproto2-py ? | 14:41 |
| mithro | cahirwpz: give me a second to fix your first bug :) | 14:43 |
| cahirwpz | it's not mine - I did not introduce it ;P | 14:43 |
| mithro | http://notinventedhe.re/on/2010-4-26 | 14:46 |
| tpb | Title: Not Invented Here by Bill Barnes and Paul Southworth (at notinventedhe.re) | 14:46 |
| cahirwpz | :) | 14:47 |
| cahirwpz | drop me a line when I can pull your fix from the repo | 14:48 |
| mithro | cahirwpz: just about to push it now | 14:53 |
| CIA-29 | mithro libtpproto2-py * r746779517928 /tp/netlib/parser.py: Make the print a logging.debug message. | 14:57 |
| CIA-29 | mithro libtpproto2-py * r8f9fc7cfed4e /tp/netlib/structures/Structure.py: | 14:57 |
| CIA-29 | Use single underscore rather then double underscore. | 14:57 |
| CIA-29 | Allows overriding by subclasses. | 14:57 |
| CIA-29 | mithro libtpproto2-py * r22c1da2b57a2 / (3 files in 2 dirs): Fixing the length of a packed output. | 14:57 |
| mithro | cahirwpz: give that a go? | 14:57 |
| mithro | cahirwpz: I'm heading to bed now | 15:08 |
| cahirwpz | ok | 15:09 |
| cahirwpz | it works :) | 15:10 |
| mithro | yay! | 15:12 |
| cahirwpz | how about decoding ? | 15:13 |
| mithro | so it looks like constructing packets off the wire might not yet be totally implemented. | 15:13 |
| mithro | once you have the packet type it's pretty easy | 15:13 |
| cahirwpz | ok - I'll look into that | 15:14 |
| mithro | look at the unpack method | 15:19 |
| *** Agon-phone has quit IRC | 16:32 | |
| *** null_000 has quit IRC | 16:55 | |
| *** Epyon has joined #tp | 16:56 | |
| *** Agon-phone has joined #tp | 17:17 | |
| *** Agon-phone has quit IRC | 18:36 | |
| *** Epyon has quit IRC | 18:42 | |
| *** Noughmad has joined #tp | 19:15 | |
| *** JLP has quit IRC | 19:44 | |
| *** llnz has joined #tp | 20:18 | |
| *** llnz has joined #tp | 20:18 | |
| llnz | morning all | 20:18 |
| *** llnz2 has joined #tp | 20:24 | |
| *** llnz has quit IRC | 20:24 | |
| *** llnz2 is now known as llnz | 20:25 | |
| llnz | less than graceful transition to wireless | 20:25 |
| *** alanp has joined #tp | 20:41 | |
| *** cahirwpz has quit IRC | 21:01 | |
| *** DTRemenak has quit IRC | 21:40 | |
| *** DTRemenak has joined #tp | 22:03 | |
| *** jmtan has quit IRC | 22:27 | |
| *** jmtan has joined #tp | 22:27 | |
| *** Noughmad has quit IRC | 22:38 | |
| *** Tanoku_ has quit IRC | 22:52 | |
| *** glew has joined #tp | 23:28 | |
| *** mithro has quit IRC | 23:42 | |
| llnz | bbl | 23:50 |
| *** llnz has quit IRC | 23:50 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!