--- Day changed Thu Nov 29 2018
00:18 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 250 seconds]     
00:21 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 252 seconds]     
00:22 -!- belcher [~belcher@unaffiliated/belcher] has joined #joinmarket     
00:24 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
00:31 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has quit [Ping timeout: 250 seconds]     
00:48 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 250 seconds]     
00:50 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
01:13 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has joined #joinmarket     
02:51 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 246 seconds]     
02:53 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket     
03:25 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has quit [Ping timeout: 250 seconds]     
03:31 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has joined #joinmarket     
03:38 < waxwing> thinking of adding utxo freeze and coin control features. shouldn't be too hard. well, i guess it's one of a million things but worth mentioning it.
04:06 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has quit [Ping timeout: 250 seconds]     
05:28 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 268 seconds]     
05:31 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket     
05:56 -!- takamatsu [~takamatsu@unaffiliated/takamatsu] has joined #joinmarket     
06:00 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
06:06 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
06:17 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
06:21 < waxwing> lol test_commands is totally borked and has been for like a year. it silently stops after the first couple of messages. fixing up now, it might even test something useful in future :)
06:22 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
06:27 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 250 seconds]     
06:28 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
06:50 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 250 seconds]     
06:57 < waxwing> with builtins, why does 'isinstance(x, str)' return False when type(x) is 'str'?
06:57 < waxwing> i guess this is probably the  same question in python3 if i get the idea right
06:58 < waxwing> hmm googling/stackexchanging suggests it's the builtins/future thing, right
07:04 -!- undeath [~undeath@hashcat/team/undeath] has joined #joinmarket     
07:06 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
07:16 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 240 seconds]     
07:27 -!- GitHub155 [GitHub155@gateway/service/github.com/x-qzcwjoeymbveyhwe] has joined #joinmarket     
07:27 < GitHub155> [joinmarket-clientserver] AdamISZ opened pull request #242: Fix bugs in bigstring and test_commands (master...fix-commands) https://git.io/fpV9q
07:27 -!- GitHub155 [GitHub155@gateway/service/github.com/x-qzcwjoeymbveyhwe] has left #joinmarket []     
07:27 < waxwing> re: the above undeath , Lightsword does this change make sense to you? https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/242/files#diff-56fdf590f4f84809d0e041750eab32d4R46
07:28 < waxwing> context: https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/242#issuecomment-442874968
07:28 < undeath> "{}.{}".format(name, counter)
07:28 < waxwing> basically, it doesn't allow a unicode key in the k-v pairs sent in the amp protocol
07:29 < waxwing> undeath, would that avoid the problem that was there before? the "%s.%d" % (name, counter) gave the unicode key error
07:29 < waxwing> i guess i can just try it :)
07:30 < undeath> i guess the problem is %s
07:30 < undeath> format should take care of whatever you throw at it
07:31 < waxwing> undeath, no seems not.
07:31 < undeath> really?
07:31 < waxwing> the builtins thing seems to be quite tricky for this stuff, since 'str' isn't quite what it was.
07:31 < waxwing> well, unless i made a mistake, that reintroduces the error
07:32 < undeath> if it breaks format that's pretty stupid
07:32 < waxwing> strings["{}.{}".format(name, counter)] = nextChunk
07:32 < waxwing> copy-paste ^
07:32 < undeath> that is the code I imagined
07:33 < undeath> oh, I think you may need strings[u"{}.{}".format(name, counter)] = nextChunk
07:33 < undeath> or b""?
07:33 < waxwing> undeath, oh? but i specifically *don't* want unicode
07:34 < undeath> well, then probably b""?
07:34 < waxwing> oh, b, yeah could work, i can try it, just grabbign a drink
07:34 < undeath> not even sure if bytes have format though
07:34 < undeath> no, they don't
07:35 < undeath> so, probably "{}.{}".format(name, counter).encode('ascii')
07:36 < undeath> where format is just the more py3-version of "" % (vars…)
07:37 < undeath> I tried to avoid concatenation using + if not explicitly dealing with strings only
07:37 < undeath> *I try
07:38 < undeath> trying to remember where to cast what aso is annoying and error-prone
07:47 < undeath> there is nothing wrong with your fix I guess
07:47 < undeath> it's just a matter of style
07:48 < undeath> I don't know anything about the command tests, I trust you known what you're doing there :)
07:52 < waxwing> yeah i agree better not to concat.
07:52 < waxwing> and i just tested, that works, unsurprisingly (your last suggestion), so i'll do that.
07:53 < waxwing> re: knowing what i'm doing, well no, the problem is just the broken test design. it's something i did right at the start, i just kinda found the first thing from googling that *seemed* to allow a twisted test to work, but it really doesn't.
07:53 < undeath> :D     
07:53 < waxwing> because failures seem to just get swallowed. i think there's like 3 tests that are like that (have that timeout)
07:53 < undeath> uh     
07:54 < undeath> that's not a good test then :/
07:54 < waxwing> yes. i can't say anything definitive, other than that i just don't really understand the right way to carry it out. the fix to make it better might be quite simple, just don't know.
07:54 < waxwing> at least locally you can test it and see what happens (should use -s). so at least it's not 100% worthless.
07:55 < undeath> i'll try to look into it
07:55 < undeath> well, I personally don't look at the output of successful tests usually
07:56 < waxwing> of course
07:57 < waxwing> i had hoped i was perfectly clear that i was saying it was not acceptable. but actually fixing it as another matter.
07:57 < undeath> yes, understood :)
07:57 < waxwing> updated to your syntax. i guess not much point in not merging it now, or ..?
07:58 < undeath> go ahead
07:59 < undeath> the quoting style is a bit weird, but that's caused by myself
07:59 < waxwing> oh? what's that?
07:59 < undeath> well, it's using double quotes for the first string and single quotes for the second one
08:00 < undeath> but that's a mess thoughout the whole codebase anyway
08:00 < waxwing> yes no consistency there atm
08:01 < undeath> when I checked stackoverflow there was no real agreement on when to use what
08:02 < undeath> a somewhat bigger minoriy favoured using double quotes for human-readable information and single quotes for internal stuff like identifiers, fmt str data, …
08:02 < undeath> *minority
08:03 < undeath> so it's mostly a matter of personal preference (or ideally project guidelines)
08:03 < undeath> not that it's really important
08:16 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
08:54 -!- GitHub101 [GitHub101@gateway/service/github.com/x-fsbrmbqsipzimmar] has joined #joinmarket     
08:54 < GitHub101> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/fpVj3
08:54 < GitHub101> joinmarket-clientserver/master 3027c03 AdamISZ: Merge #242: Fix bugs in bigstring and test_commands...
08:54 -!- GitHub101 [GitHub101@gateway/service/github.com/x-fsbrmbqsipzimmar] has left #joinmarket []     
08:55 -!- GitHub41 [GitHub41@gateway/service/github.com/x-irgzxrpsgxjczlgg] has joined #joinmarket     
08:55 < GitHub41> [joinmarket-clientserver] AdamISZ closed pull request #242: Fix bugs in bigstring and test_commands (master...fix-commands) https://git.io/fpV9q
08:55 -!- GitHub41 [GitHub41@gateway/service/github.com/x-irgzxrpsgxjczlgg] has left #joinmarket []     
09:02 < undeath> arubi: isn't coincurve_patch_ignore_sys_libsecp missing a return? or is that optional?
09:03 < arubi> it'll return the value of the last exited command
09:03 < undeath> ok, nice
09:10 -!- mr_paz [~mr_paz@84.39.112.89] has joined #joinmarket     
09:11 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
09:12 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
09:48 -!- achow101 [~achow101@unaffiliated/achow101] has quit [Ping timeout: 264 seconds]     
10:03 -!- achow101 [~achow101@unaffiliated/achow101] has joined #joinmarket     
10:48 -!- GitHub73 [GitHub73@gateway/service/github.com/x-ozbilmgvvmwqslzn] has joined #joinmarket     
10:48 < GitHub73> [joinmarket-clientserver] undeath opened pull request #243: remove slowaes (master...remove-slowaes) https://git.io/fpwmh
10:48 -!- GitHub73 [GitHub73@gateway/service/github.com/x-ozbilmgvvmwqslzn] has left #joinmarket []     
10:53  * arubi loves when tons of lines are removed
10:53 < undeath> :D     
11:09 -!- raedah [~x@184.75.221.179] has quit [Ping timeout: 252 seconds]     
11:11 -!- rdymac [uid31665@gateway/web/irccloud.com/x-areagrgrnglqqtoh] has quit [Quit: Connection closed for inactivity]     
12:02 < waxwing> https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/241/files#diff-4f3c2beacd536175f30616463ddaeb33R397 <-- probably just dumb, but i don't understand why 'wb' is appropriate here, if we're writing a text file?
12:02 < undeath> i was wondering, too
12:14 < belcher> the thing that 'w' as opposed to 'wb'" does is convert "\n" to "\r\n" on windows
12:14 < belcher> some people just have a muscle-memory habit of using 'wb' for everything? so idk
12:14 < belcher> if you only use linux then they're the same
12:31 < waxwing> that's what i thought i knew :) but with this whole future thing, i'm not sure, there was presumably a reason.
12:32 < undeath> yes, I'd actually expect it to error out if you pass a str to it
12:44 < waxwing> Lightsword, starting sendpayment gives this error in the option parsing: https://0bin.net/paste/igMyaREHivZszYy5#d+BYU9rfwc0AV2ddAC3lpES2ueIiZJBl4TqUtQG5hsO
12:46 -!- beIcher [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
12:50 -!- beIcher [~user@unaffiliated/belcher] has joined #joinmarket     
12:52 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 252 seconds]     
13:09 -!- belcher [~belcher@unaffiliated/belcher] has joined #joinmarket     
13:30 < waxwing> seems like replacing type=str with type="string" works
13:31 < undeath> that looks so wrong
13:32 < waxwing> here: https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/241/files#diff-cc68ef76b7c622c229f24ede1540d990R63
13:32 < waxwing> well but if you look in optparse.py
13:33 < undeath> i'm not saying it is wrong :)
13:34 < waxwing> line 552
13:34 < waxwing> oh, ok
13:35 < undeath> how did you figure out it should be "string"?
13:36 < waxwing> sorry line 522
13:36 < waxwing> there's a list self.TYPES there
13:36 < undeath> I'd more expect it to by type=newstr
13:36 < waxwing> the list of types is a list of strings. one of the strings is "string"
13:36 < waxwing> :)     
13:36 < undeath> heh     
13:37 -!- mr_paz [~mr_paz@84.39.112.89] has quit [Ping timeout: 246 seconds]     
13:39 < waxwing> undeath, is this one obvious? trying tumbler.py got:
13:39 < waxwing>   File "tumbler.py", line 45, in main
13:39 < waxwing>     maxcjfee = get_max_cj_fee_values(jm_single().config, options_org)
13:39 < waxwing>   File "/home/adam/code/Joinmarket/joinmarket-clientserver/scripts/cli_options.py", line 118, in get_max_cj_fee_values
13:39 < waxwing>     if len(filter(lambda x: x is None, fee_values)):
13:39 < waxwing> TypeError: object of type 'itertools.ifilter' has no len()
13:40 < undeath> missing list()
13:40 < waxwing> ok, wrap the whole thing in list()?
13:40 < waxwing> oh list before len
13:40 < undeath> len(list(…))
13:40 < waxwing> right
13:40 < undeath> in py3 most things are iterators, not lists anymore
13:40 < waxwing> yeah i remember this with set
13:41 < undeath> however, since it's if len()
13:41 < undeath> you could just do if any()
13:41 < waxwing> huh, never used that idiom. so like `if any(filter(...` ?
13:41 < waxwing> or does it still need list
13:41 < undeath> yep     
13:41 < undeath> no, that avoids the list
13:42 < undeath> because in fact you only want to know if there is some item left
13:43 < undeath> i think you could even remove the filter
13:43 < undeath> if any(lambda x: x is None, fee_values)
13:44 < undeath> oh no
13:44 < undeath> that's something different now
13:44 < undeath> maybe it's gettling a little to late for me right now
13:46 < undeath> any(x is None for x in fee_values)
13:46 < undeath> that's what we are looking for
13:51 < waxwing> sounds good. i'm not sure what the difference is between that version and `if any(filter(lambda x: x is None, fee_values)):`?
13:51 < undeath> any(filter()) will actually fail
13:51 < waxwing> this line also crashes: https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/440823b9670c0a232d42773c1c54c5972398ccfa/jmclient/jmclient/taker.py#L483
13:51 < waxwing> exceptions.TypeError: can only concatenate list (not "dict_keys") to list
13:51 < undeath> filter() will evaluate to a list of None or an empty list
13:52 < undeath> and any() would then determine that to be always False
13:52 < undeath> err, not a list but an iterator, anyway
13:52 < waxwing> oh. i hadn't ever used any, didn't know that.
13:52 < waxwing> "any"
13:52 < undeath> any looks for some item that does not evaluate to False
13:53 < waxwing> right. the any(filter( def doesn't work, will edit my comment, thanks.
13:55 < undeath> it yoes that for self.utxos.values() ?
13:56 < waxwing> i guess, do i have to do another list() cast or something?
13:57 < undeath> i'm unsure what the sum() does there
13:57 < waxwing> oh i just remembered that is some weird idiom for flattening lists
13:57 < waxwing> yeah that. the sum( , []) thing
13:57 < undeath> ah, I see
13:57 < undeath> so, same as reduce() ?
13:58 < waxwing> i don't recall. belcher might
13:58 < waxwing> i mean obv i can figure it out, it's just not in my head right now.
14:00 < belcher> its add them up iirc
14:00 < belcher> one_element + two_element + three_element ...
14:00 < belcher> i think
14:00 < belcher> list addition i mean
14:02 < undeath> reduce(lambda x, a: x+a, self.utxos.values(), []) may or may not work
14:02 < undeath> it seems to do the same thing
14:04 < undeath> err, nvm
14:04 < undeath> I really should stop trying to think for today
14:10 < Lightsword> what’s the reason for this? I’m trying to rebase my branch https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/242/files#diff-af6945b804e27cd4faaba66363230036R160
14:11 < undeath> it's mentioned in the pr comment
14:11 < undeath> […] due to an unupgraded command syntax in JM_SETUP […]
14:12 < Lightsword> ah     
14:14 < undeath> undeath | reduce(lambda x, a: x+a, self.utxos.values(), [])
14:14 < undeath> this should actually work I guess
14:15 < undeath> but then, I have no idea why sum() wouldn't :/
14:25 < waxwing> Lightsword, yes sorry it's confusing, the PR (242) did two separate but connected things: updated the test which simply didn't function right, and also fix the syntax in bigstring.py so it actually works with large strings, in the post-future-ized version of the code. but i guess you got it already :)
14:48 < undeath> waxwing | exceptions.TypeError: can only concatenate list (not "dict_keys") to list
14:48 < undeath> are you sure the error is actually with self.utxos.values() and not somewhere else?
14:48 < undeath> because that line has nothing to do with dict keys
14:49 < undeath> maybe some wrongly typed data ended up in self.utxos.values()
14:51 < undeath> https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/440823b9670c0a232d42773c1c54c5972398ccfa/jmclient/jmclient/taker.py#L329
14:51 < undeath> I think this is missing the list() call
15:26 -!- arubi_ [~ese168@gateway/tor-sasl/ese168] has joined #joinmarket     
15:27 -!- arubi [~ese168@gateway/tor-sasl/ese168] has quit [Ping timeout: 256 seconds]     
15:55 < Lightsword> waxwing, yeah going to look at that more
16:08 -!- undeath [~undeath@hashcat/team/undeath] has quit [Quit: WeeChat 2.3]     
17:28 -!- raedah [~x@184.75.221.3] has joined #joinmarket     
17:45 -!- arubi_ [~ese168@gateway/tor-sasl/ese168] has quit [Remote host closed the connection]     
17:46 -!- arubi [~ese168@gateway/tor-sasl/ese168] has joined #joinmarket     
17:49 -!- d3spwn_ [~Jimmy@a83-161-157-200.adsl.xs4all.nl] has joined #joinmarket     
17:52 -!- d3spwn [~Jimmy@83.161.157.200] has quit [Ping timeout: 244 seconds]     
17:59 -!- AgoraRelay [~jmrelayfn@p5DE4A628.dip0.t-ipconnect.de] has quit [Ping timeout: 244 seconds]     
18:11 -!- AgoraRelay [~jmrelayfn@p5DE4A82F.dip0.t-ipconnect.de] has joined #joinmarket