1 2015-06-07 00:19:58 <phantomcircuit> Luke-Jr, it's not and it disables like evrything including the wallet
 2 2015-06-07 00:20:04 <phantomcircuit> i dont expect many people to actually run it
 3 2015-06-07 00:20:27 <phantomcircuit> the only reason i can think of to run it is to get a verified list of block hashes
 4 2015-06-07 00:20:50 <phantomcircuit> and iirc there aren't any spv clients that can trust a node for headers but get transactions data from other nodes
 5 2015-06-07 01:53:20 <Luke-Jr> phantomcircuit: I meant that in future tense.
 6 2015-06-07 01:55:21 <gmaxwell> the current level of support is basically useless, but if we didn't get it in the feature was going to never get merged.
 7 2015-06-07 01:55:55 <gmaxwell> now the rest of the support for it, e.g. getting tip blocks, storing random ranges,  and working with the wallet can all make progress on their own time without the basic pruning patch continually bitrotting.
 8 2015-06-07 02:53:30 <gmaxwell> wumpus: is it known that the rpc interface has changed in git master so that / is escaped in strings?
 9 2015-06-07 02:53:44 <gmaxwell> e.g. in getpeerinfo
10 2015-06-07 02:53:45 <gmaxwell> "subver": "\/Satoshi:0.10.2\/",
11 2015-06-07 02:54:39 <gmaxwell> This seems wrong to me.
12 2015-06-07 02:59:54 <Luke-Jr> sounds like a Univalue bug
13 2015-06-07 03:00:07 <Luke-Jr> s/a/another/ <.<
14 2015-06-07 03:00:22 <gmaxwell> well duh, but I didn't know if it was a known one.
15 2015-06-07 03:03:11 <gmaxwell> keep in mind, on most of my systems I haven't been able to run master or 0.11 since the scheduler stuff was merged; without applying workarounds; so I'm not exactly current with whats running.
16 2015-06-07 03:10:25 <Arnavion> (It doesn't matter for JSON - "\/" is the same as "/" )
17 2015-06-07 03:11:51 <Luke-Jr> gmaxwell: :|
18 2015-06-07 04:44:05 <Rozal> Did u guys fix Bitcoin yet?
19 2015-06-07 05:07:20 <petertodd> phantomcircuit: and pruning isn't even very useful yet to get a verified list of block hashes: https://github.com/bitcoin/bitcoin/pull/6247
20 2015-06-07 05:08:20 <phantomcircuit> petertodd, eh a minimal bitcoin p2p node is pretty easy
21 2015-06-07 05:08:36 <petertodd> phantomcircuit: an RPC call is easier :)
22 2015-06-07 05:11:35 <phantomcircuit> petertodd, well.... if you're using a library they're basically the same
23 2015-06-07 05:11:44 <phantomcircuit> if you're not using libraries the rpc call is 10000x harder
24 2015-06-07 05:12:02 <sipa> ?
25 2015-06-07 05:12:23 <sipa> getblock will simply fail if the block data is not available
26 2015-06-07 05:12:27 <sipa> iirc
27 2015-06-07 05:14:34 <sipa> phantomcircuit: "if you're using a library they're basically the same" -> running an SPV node is the same as doing an RPC call? wut?
28 2015-06-07 05:15:48 <phantomcircuit> sipa, if you're using a library for both rpc and speaking bitcoin it's pretty similar really
29 2015-06-07 05:16:07 <phantomcircuit> especially since all the bitcoin speaking libraries will decode the result
30 2015-06-07 05:16:52 <sipa> i must be misunderstanding you
31 2015-06-07 05:17:37 <phantomcircuit> sipa, you have a local bitcoin node, you want to get a header and already have the hash
32 2015-06-07 05:17:50 <sipa> ah, you mean use p2p to fetch a header
33 2015-06-07 05:17:53 <phantomcircuit> it's not much more complicated to connect as an spv node and get the hash than it is to make an rpc call
34 2015-06-07 05:18:11 <phantomcircuit> yeah
35 2015-06-07 05:19:35 <sipa> yeah, i was confused... generally i don't consider using p2p for querying trusted data, so i saw the alternative to querying rpc to be "running an spv client in the caller, and validating the header chain yourself"
36 2015-06-07 05:19:46 <sipa> nvm
37 2015-06-07 05:22:18 <phantomcircuit> sipa, the header doesn't really need to be validated since the getheader call needs the hash anyways
38 2015-06-07 05:22:25 <phantomcircuit> validating the result is trivial
39 2015-06-07 05:27:11 <phantomcircuit> petertodd, needs "ismainchain"
40 2015-06-07 05:30:36 <gmaxwell> petertodd: should harmonize with the verifytxoutproof rpc.
41 2015-06-07 06:28:56 <wumpus> gmaxwell: I'm not sure if that's a bug or not, I don't know JSON/RFC4627's escaping rules, but it shouldn't hurt
42 2015-06-07 13:35:54 <mjerr> why are the fees on the testnet so much higher (factor 10) for a transaction that should get mined within 1 block
43 2015-06-07 13:36:14 <mjerr> I would have guessed since there is so little going on, that the fees are lower aswell
44 2015-06-07 13:53:46 <flyingkiwi> mjerr, I asked this myself. And I answered myself thats because there are 95k transactions in memory pool. But thats superficial knowledge
45 2015-06-07 14:02:12 <mjerr> flyingkiwi, that might be it, many devs send coins and dont care about them anymore i guess, especially in the beginning of developing something
46 2015-06-07 18:25:26 <michagogo> Who admins the bitcoin.org webserver?
47 2015-06-07 18:26:11 <michagogo> A path under /bin that doesn't exist should default to a 302 redirect and not a 301, and I want to know who to poke
48 2015-06-07 18:26:21 <michagogo> (that and/or a cache control header should be sent)
49 2015-06-07 18:36:36 <Luke-Jr> michagogo: why 302?
50 2015-06-07 18:37:28 <michagogo> Luke-Jr: unless otherwise specified, 301 is cached forever
51 2015-06-07 18:37:33 <jonasschnelli> hmm... anyone knows why CExtPubKey.pubkey.GetID() is *not* equal to CExtKey.key.GetPubKey().GetID()? IMO it should?!
52 2015-06-07 18:37:34 <michagogo> (at least in Chrome)
53 2015-06-07 18:38:40 <michagogo> Which is a problem for paths that are checked before they exist
54 2015-06-07 18:39:06 <michagogo> 301 means "Moved Permanently", and browsers treat it as such
55 2015-06-07 18:40:37 <michagogo> Oh, maybe 307?
56 2015-06-07 18:41:07 <sipa> jonasschnelli: it should, unless one is compressed and the other is not
57 2015-06-07 18:41:26 <michagogo> Anyway, one of the 3xx codes, but not 301
58 2015-06-07 18:41:44 <sipa> jonasschnelli: or if you're talking about derived keys, if the derivatrion was hardened, that won't work
59 2015-06-07 18:43:12 <jonasschnelli> sipa: i have a non hardened childkey CExtKey. Why is this: CExtKeyChild.Neuter().pubkey.GetID() =! CExtKeyChild.key.GetPubKey().GetID()?
60 2015-06-07 18:43:58 <jonasschnelli> my CExtKeyChild instance is m/44'/0'/0'/0 (non hardened)
61 2015-06-07 18:46:18 <jonasschnelli> sipa: generating the keys through the chainPath from master to m/44'/0'/0'/0/0 is deterministic and gives always the same bitcoin address. But when i try to get CKey directly over CExtKey i get a different address (over CExtKey.key.GetPubKey()).
62 2015-06-07 18:47:16 <jonasschnelli> when i do CExtKey.Neuter() then use pubkey.GetKeyID() i get the correct address/key.
63 2015-06-07 18:51:03 <jonasschnelli> Nevermind & damit. I should really stop now and go to bed. :-) . Mixed up some variables. All fine.
64 2015-06-07 18:51:20 <Luke-Jr> michagogo: why is it a redirect anyway?
65 2015-06-07 18:51:39 <Luke-Jr> should be 404
66 2015-06-07 18:53:45 <michagogo> You mean as opposed to 404? Idk, I guess I assumed there was a reason for it
67 2015-06-07 18:54:22 <michagogo> But the one thing it shouldn't be is 301
68 2015-06-07 18:56:26 <Luke-Jr> michagogo: I guess maybe it's somewhere on https://github.com/bitcoin/bitcoin.org/
69 2015-06-07 19:03:09 <harding> michagogo: Saivann Carignan admins the server, and you can find his contact info (and mine) through here: https://bitcoin.org/en/about-us#maintenance --- but for non-urgent stuff, you're best off opening an issue on the repository Luke-Jr linked to immediately above.
70 2015-06-07 19:07:18 <harding> michagogo: also, I'm not sure, but I think the reason we use 301 is because we want Google to update its links to old files/directories and 301 is the only status code they'll do it for.  (But I've bumped into 301 permanant caching enough times in development of the site to dislike it, so feel free to open an issue.)
71 2015-06-07 19:12:31 <michagogo> Update its links to old files? What does that mean exactly?
72 2015-06-07 19:15:21 <BlueMatt> cfields: I know you wanted to fix it some other way, but I got tired of build not working, so https://github.com/bitcoin/bitcoin/pull/6248
73 2015-06-07 19:16:45 <harding> michagogo: If example.com/foo is a popular page with lots of incoming links but you move its content to example.com/bar and 301 foo->bar, then Google will link to example.com/bar anytime it would've previously linked to example.com/foo.
74 2015-06-07 19:17:31 <michagogo> Er, where would that be relevant to bitcoin.org/bin?
75 2015-06-07 19:17:41 <michagogo> Or rather, when
76 2015-06-07 19:18:32 <michagogo> And yeah, it makes sense to 301 that, because the content is, in fact, "moved permanently"
77 2015-06-07 19:18:42 <michagogo> It's making it a default that's weird
78 2015-06-07 19:19:20 <harding> michagogo: presumably the test binaries which go into /bin/bitcoin-core-VERSION/test/ during the RC cycle but with ../test being removed once the finals are released.
79 2015-06-07 19:19:47 <michagogo> Um, does Google really link to those all that much?
80 2015-06-07 19:19:49 <harding> michagogo: anyway, Saivann set the policy, so I recommend opening an issue and talking to him.
81 2015-06-07 19:21:00 <harding> michagogo: heh, I'll open an issue.  What's your GitHub username so I can @ mention you?
82 2015-06-07 19:21:10 <michagogo> <-
83 2015-06-07 19:21:16 <harding> Ok, simple enough :-)
84 2015-06-07 19:21:31 <michagogo> Thanks
85 2015-06-07 19:41:00 <harding> michagogo: https://github.com/bitcoin/bitcoin.org/issues/880
86 2015-06-07 22:18:19 <Stormdev> If I want to test some transaction using python-bitcoinlib or libbitcoin in C++
87 2015-06-07 22:19:13 <Stormdev> I need to run a node on my computer? or just create a bitcoin.conf correctly?
88 2015-06-07 22:24:14 <sipa> Stormdev: bitcoin.conf is part of bitcoind's configuration. If you don't run bitcoind, you don't need bitcoin.conf...
89 2015-06-07 22:26:41 <Stormdev> ok
90 2015-06-07 22:27:00 <Stormdev> so when I want to run bitcoin.rpc.proxy I have to specify an url
91 2015-06-07 22:27:21 <sipa> I do not know what bitcoin.rpc.proxy is.
92 2015-06-07 22:27:40 <Stormdev> well in bitcoinlib
93 2015-06-07 22:28:07 <Stormdev> it's a class used to create a proxy to a bitcoin RPC service
94 2015-06-07 22:29:06 <sipa> well you need to configure the address of the bitcoind you're connecting to
95 2015-06-07 22:29:20 <sipa> whether that's one you run yourself, or another one you can access...
96 2015-06-07 22:30:18 <Stormdev> hum I just want to connect to testnet
97 2015-06-07 22:32:37 <sipa> then you need to run bitcoind to connect to testnet
98 2015-06-07 22:32:57 <sipa> the rpc protocol is not a means to connect to testnet, it's a means to connect to a bitcoind instance
99 2015-06-07 22:33:41 <Stormdev> ok thanks, it's clear for me now