--- Day changed Sat Jun 11 2016
00:42 -!- molz [~molly@unaffiliated/molly] has quit [Ping timeout: 244 seconds]
03:52 < JM-IRCRelay> [AlexCato] currenlty working on fixing #567 ( https://github.com/JoinMarket-Org/joinmarket/issues/567 ), i found the logic error and know how to fix it arithmetically, but my python skills seem to be failing and I have no clue what could be the cause. Consider those python lines:
03:52 < JM-IRCRelay> [AlexCato]             print('Maxorder = ' + str(balance)) # works, current balance for the order is displayed correctly
03:52 < JM-IRCRelay> [AlexCato]             print('cjfee = ' + str(thecjfee[oid + delta])) # works, gives me the cj-fee for that balance (relative fee)
03:53 < JM-IRCRelay> [AlexCato]             cjfee_for_maxorder = (thecjfee[oid + delta]) * balance # i assume this works
03:53 < JM-IRCRelay> [AlexCato]             print('cjfee for max = ' + str(cjfee_for_maxorder)) # INFINITE loop displaying the cjfee
03:53 < JM-IRCRelay> [AlexCato] what the...?
03:55 < JM-IRCRelay> [AlexCato] if I work arithmetically with the "cjfee_for_maxorder", it crashes with a "MemoryError", so i guess its doing some inifinite stuff there as well
03:55 < waxwing> not sure which bit of code that's in, but what data type is thecjfee[oid+delta] ?
03:55 < waxwing> because if it's not a numerical value but a list, or tuple, then *balance will create a huge list/tuple
03:56 < waxwing> example: ([2])*4 -> [2, 2, 2, 2]
03:56 < JM-IRCRelay> [AlexCato] oid and delta are integers, the cj fee is this:
03:56 < JM-IRCRelay> [AlexCato]         thecjfee = cjfee[::-1]
03:56 < JM-IRCRelay> [AlexCato] and that indeed is a list
03:57 < waxwing> could you link the line on github?
03:57 < JM-IRCRelay> [AlexCato] the prints are my own while debugging, the rest is here:
03:58 < JM-IRCRelay> [AlexCato] https://github.com/JoinMarket-Org/joinmarket/blob/master/yield-generator-mixdepth.py#L127
03:58 < JM-IRCRelay> [AlexCato] the other line is 111
03:59 < waxwing> at first glance, since you're indexing into the list, it should be OK: if thecjfee[oid+delta] = 7, say then you're doing (7)*balance and that should spit out a number, not a tuple or list.
04:10 < JM-IRCRelay> [AlexCato] when i check the output of the above print lines i pasted into this irc chat, it seems fine as well
04:10 < JM-IRCRelay> [AlexCato] Maxorder = 80303658
04:10 < JM-IRCRelay> [AlexCato] cjfee = 0.00013
04:10 < JM-IRCRelay> [AlexCato] those multiplied and printed out = infloop
04:11 < JM-IRCRelay> [AlexCato] if one of those was a list, i should see it in the printed lines above, right?
04:11 < waxwing> maybe gist or pastebin your code? also when you say infloop, is it printing something out or just hanging?
04:12 < JM-IRCRelay> [AlexCato] it prints the same number over and over again, but its too fast for me to see what it is. Even ctrl-c doesnt work, have to close the terminal to get it to stop
04:12 < JM-IRCRelay> [AlexCato] will pastebin, sec
04:15 < JM-IRCRelay> [AlexCato] http://pastebin.com/G5CVdyuq
04:15 < JM-IRCRelay> [AlexCato] line 120-124 are the debugging lines
04:31 < waxwing> sorry afk but i notice thecjfee[(oid + delta)] * balance <- inner () , why? usually that should fail
04:32 < JM-IRCRelay> [AlexCato] removing the () delivers the same result, it was without the () at first and i just played around a little to see if that has any effect
04:33 < waxwing> yeah you're right, it's interpreting it as arithmetic and ignoring it. so it shouldn't fail,  assuming the index is valid.
04:34 < JM-IRCRelay> [AlexCato] and no worries, this isnt time sensitive. Just trying to do something useful while i wait for bitcoin core to load its blocks on the new windows VM to improve docs... that takes a while these days ;)
04:35 < waxwing> which line does it loop on? you can stick in a time.sleep(1) to allow you to keep control of the terminal, absent running in debugger
04:35 < JM-IRCRelay> [AlexCato] Line 123 loops
04:36 < waxwing> so does it output 'cjfee for max' only once and then output a long list of numbers? or?
04:36 < JM-IRCRelay> [AlexCato] yup
04:37 < waxwing> so my suggestion was right, but i can't see any logic to it from that code.
04:37 < waxwing> you can check if the variable is of type list before that line
04:37 < waxwing> isinstance
04:38 < waxwing> if isinstance(x, list) etc.
04:38 < JM-IRCRelay> [AlexCato] on it
04:41 < waxwing> hmm. i don't know this script. do we know what datatype 'balance' is there?
04:42 < waxwing> yeah it must be a number from what follows
04:43 < JM-IRCRelay> [AlexCato] so neither cjfee_for_maxorder nor balance is a list. Is there a reverse-way, like cjfee_for_maxorder.gettype() or something similar?
04:43 < waxwing> so i take it from that, the output does not look like 'cjfee for max: [x, x, x,... ' ? it literally just repeatedly prints a number after that string?
04:44 < JM-IRCRelay> [AlexCato] yes, repeatedly prints a number. Its the same one from what i can tell
04:44 < waxwing> so e.g. 'cjfee for max: 3 3 3 3...' ?
04:44 < JM-IRCRelay> [AlexCato] time.sleep doesnt help me either here
04:45 < waxwing> this is really bizarre. what exactly does it print? maybe you can use log.debug() to get it into a file, instead of print()?
04:45 < JM-IRCRelay> [AlexCato] cant see the "cjfee for max", because its scrolling down too fast.
04:45 < JM-IRCRelay> [AlexCato] can do.
04:46 < waxwing> ah, i wonder if it's this:
04:47 < waxwing> if thecjfee[x] is a byte/string instead of a number, you can get this effect: if thecjfee[0]='a' then (thecjfee[0])*10 = 'aaaaaaaaaa'
04:47 < JM-IRCRelay> [AlexCato] nothing in the logs. Just spams the terminal even with log.debug, but the log is empty at the point of the loop
04:49 < JM-IRCRelay> [AlexCato] Line 31 solves that mystery
04:50 < JM-IRCRelay> [AlexCato] why would anyone want these be strings
04:50 < waxwing> because they're to be published
04:50 < waxwing> but one can discuss where they should be string and where not etc.
04:52 < JM-IRCRelay> [AlexCato] Line99 would have been a good hint, but somehow i didnt see the typecast
04:52 < JM-IRCRelay> [AlexCato] thanks for your help, rest should be easy now
04:53 < waxwing> python's duck-typing causes a lot of unobvious behaviour :)
05:03 -!- moli [~molly@unaffiliated/molly] has joined #joinmarket
06:10 < GithubBot5678> [joinmarket] AlexCato opened pull request #569: yieldgens: fix issue 567 for max_size orders (develop...fix_issue_567) https://git.io/voZbu
06:33 -!- RedEmerald [~RedEmeral@c-73-231-129-86.hsd1.ca.comcast.net] has quit [Ping timeout: 260 seconds]
07:18 < JM-IRCRelay> [AlexCato] travis doesnt seem to like the new message channel ;)
07:18 < JM-IRCRelay> [AlexCato] E       TypeError: Can't instantiate abstract class DummyMessageChannel with abstract methods _announce_orders, _privmsg, _pubmsg, run, shutdown
07:59 < waxwing> alexcato, don't worry, i know what that is, fixed it the other day in a PR. it has no relation to your PR (which unfortunately isn't even covered in the build test. busy now, sorry.
08:09 < JM-IRCRelay> [AlexCato] yup, no hurries. This just fixes a very rare edge case anyways which barely ever should happen in real time
08:36 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Ping timeout: 244 seconds]
08:55 -!- imposter [uid57046@gateway/web/irccloud.com/x-aqqgtdjhrddysnzt] has quit [Quit: Connection closed for inactivity]
10:15 < GithubBot5678> [joinmarket] AdamISZ opened pull request #571: fix tests for MC refactoring (develop...fix_tests) https://git.io/vonfR
10:25 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has joined #joinmarket
10:26 < GithubBot5678> [joinmarket] AdamISZ pushed 2 new commits to develop: https://git.io/vonfb
10:26 < GithubBot5678> joinmarket/develop 80774b0 Adam Gibson: fix tests for MC refactoring
10:26 < GithubBot5678> joinmarket/develop 5e6cdf1 Adam Gibson: Merge pull request #571 from AdamISZ/fix_tests...
10:30 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 272 seconds]
10:30 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
10:35 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 260 seconds]
10:36 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
10:36 -!- proslogion [~proslogio@90.210.173.173] has joined #joinmarket
10:38 < proslogion> gmaxwell: if fans ask for your autograph in the future, make sure to create a pair with appended unique junk strings that can collide in the first 64 bits of their SHA256 digests :)
10:40 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 264 seconds]
10:51 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:02 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 276 seconds]
11:02 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:03 < waxwing> belcher_: and others interested in getting on top of this, here are some thoughts/questions on the wallet sync issue: http://0bin.net/paste/gCSoYSCCEcOsCYmo#3ZjlEicKTeFAK+y9Br1ZH40GWgmMqgR2YAHe1wNbVef
11:06 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 246 seconds]
11:07 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:26 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 276 seconds]
11:26 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:34 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 272 seconds]
11:34 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:39 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 258 seconds]
11:40 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
11:48 -!- viasil [~viasil@198.12.75.25] has quit [Ping timeout: 260 seconds]
11:49 -!- viasil [~viasil@198.12.75.25] has joined #joinmarket
12:39 -!- nkuttler [~nkuttler@unaffiliated/nkuttler] has left #joinmarket []
12:53 < gmaxwell> proslogion: lol
12:58 -!- proslogion [~proslogio@90.210.173.173] has left #joinmarket ["离开"]
15:18 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket
16:46 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has quit [Quit: Leaving.]
17:34 -!- belcher [~user@unaffiliated/belcher] has joined #joinmarket
18:08 -!- Ineedhelp [6c0c59f4@gateway/web/freenode/ip.108.12.89.244] has joined #joinmarket
18:09 -!- Ineedhelp [6c0c59f4@gateway/web/freenode/ip.108.12.89.244] has quit [Client Quit]
18:34 -!- dingus [~grubles@unaffiliated/grubles] has joined #joinmarket
19:08 -!- belcher [~user@unaffiliated/belcher] has quit [Quit: Leaving]
21:10 -!- RedEmerald [~RedEmeral@c-73-231-129-86.hsd1.ca.comcast.net] has joined #joinmarket
21:33 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has joined #joinmarket
22:41 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has quit [Quit: Leaving.]