1 2014-08-31 02:40:51 <dcousens> does a Tx sequence number have any effect on a transactions behaviour in the mempool?
2 2014-08-31 02:40:54 <dcousens> (nowadays)
3 2014-08-31 03:38:54 <bassguitarman> I have a question for the devs
4 2014-08-31 03:39:17 <bassguitarman> I am about to start working on watch only addresses in the bitcoin client
5 2014-08-31 03:39:24 <bassguitarman> are you against this at all?
6 2014-08-31 03:47:58 <Luke-Jr> bassguitarman: working on in what sense?
7 2014-08-31 03:48:09 <Luke-Jr> bassguitarman: it's a bad idea, and already supported in Bitcoin Core.
8 2014-08-31 03:48:27 <bassguitarman> It is?
9 2014-08-31 03:48:30 <Luke-Jr> yes
10 2014-08-31 03:49:14 <bassguitarman> is it natively supported?
11 2014-08-31 03:49:18 <Luke-Jr> ⦠yes
12 2014-08-31 03:49:23 <bassguitarman> I cant find any reference, sorry :)
13 2014-08-31 03:49:30 <Luke-Jr> bassguitarman: if you want something to work on toward making it better (ie, no longer "watch only addresses"), I suggest focussing on HD wallet and multiwallet support
14 2014-08-31 03:49:38 <Luke-Jr> that's because it's not released until 0.10
15 2014-08-31 03:49:54 <bassguitarman> Luke-Jr, that makes sense
16 2014-08-31 03:50:15 <bassguitarman> what is multiwallet?
17 2014-08-31 03:51:43 <Luke-Jr> bassguitarman: the ability to load multiple wallets in the same instance of Bitcoin Core
18 2014-08-31 03:51:51 <stapler117> ACTION can't wait untill HD wallet in bitcoin-core
19 2014-08-31 05:49:53 <BlueMatt> valid point: https://bitcoin.org/en/download's source link points to master, but all other links on that page are to 0.9.2.1
20 2014-08-31 08:10:44 <jacob___> i downloaded bitcoin client on my linux, but frack!! is there an easier way to download the blockchain? maybe torrent?
21 2014-08-31 08:13:16 <fanquake> jacob__ https://bitcointalk.org/index.php?topic=145386.0
22 2014-08-31 08:13:20 <kuzetsa> jacob___: yes, there is
23 2014-08-31 08:13:23 <kuzetsa> thanks fanquake
24 2014-08-31 08:13:48 <kuzetsa> jacob___ that link is for the torrent (the one fanquake linked)
25 2014-08-31 08:14:47 <jacob___> ok cool
26 2014-08-31 08:14:53 <kuzetsa> :)
27 2014-08-31 08:14:58 <jacob___> thanks kuzetsa
28 2014-08-31 08:15:05 <kuzetsa> heh, I didn't do anything
29 2014-08-31 08:15:17 <kuzetsa> just confirmed it was the same link I was trying to dig up and mentioned as much lol
30 2014-08-31 08:15:20 <jacob___> you potentially saved me A LOT of time
31 2014-08-31 08:38:26 <jacob___> whats the status on ethereum,.., did they produce a workable client yet?
32 2014-08-31 08:38:50 <BlueMatt> jacob___: not here
33 2014-08-31 12:41:29 <wumpus> gmaxwell: I'd prefer to have only one regtest mode as well, on the other hand if some tests really need specific sets of parameters to work with I'm not going to block that
34 2014-08-31 12:42:34 <wumpus> gmaxwell: there should anyhow be no tests for specific chains in the production code, the idea is to have flags on the Params object and test those
35 2014-08-31 12:44:44 <wumpus> anyhow if it's possible to have the tests use just the current mainnet, regtest or testnet that'd be preferable
36 2014-08-31 14:30:44 <jtimon> weird SSL error with pull tester but not with travis https://github.com/bitcoin/bitcoin/pull/4796
37 2014-08-31 14:32:14 <jtimon> wumpus is it related to testing standardness stuff without mining?
38 2014-08-31 14:33:18 <jtimon> maybe we should just regtest's behavior to be more like mainnet, should be easy to do in chainparams
39 2014-08-31 14:34:08 <jtimon> You're talking about #4732, right?
40 2014-08-31 15:12:38 <dgenr8> dcousens: nSequence < (max uint value) enables tx to be non-final based on nLockTime. nothing else.
41 2014-08-31 16:25:33 <dcousens> dgenr8: what do you mean by non-final?
42 2014-08-31 17:11:04 <jtimon> dcousens probably main.cpp::IsFinalTx()
43 2014-08-31 17:12:58 <azariah4> when creating the tx hash to sign with SIGHASH_ALL and SIGHASH_ANYONECANPAY , should the input count always be set to 01?
44 2014-08-31 17:14:46 <azariah4> noticed bitcoinj seems to clear all inputs and just add the one input that is signed: https://github.com/bitcoinj/bitcoinj/blob/master/core/src/main/java/com/google/bitcoin/core/Transaction.java#L992
45 2014-08-31 17:17:36 <azariah4> ah, I should RTFM. https://en.bitcoin.it/wiki/OP_CHECKSIG#Procedure_for_Hashtype_SIGHASH_ANYONECANPAY
46 2014-08-31 17:17:47 <jtimon> azariah I'm not sure I understand the question
47 2014-08-31 17:18:17 <jtimon> but you may want to read class script.cpp::CTransactionSignatureSerializer
48 2014-08-31 17:19:32 <jtimon> if (fAnyoneCanPay)
49 2014-08-31 17:19:32 <jtimon> // In case of SIGHASH_ANYONECANPAY, only the input being signed is serialized
50 2014-08-31 17:19:32 <jtimon> nInput = nIn;
51 2014-08-31 17:19:32 <jtimon> // Serialize the prevout
52 2014-08-31 17:19:32 <jtimon> void SerializeInput(S &s, unsigned int nInput, int nType, int nVersion) const {
53 2014-08-31 17:19:33 <jtimon> ::Serialize(s, txTo.vin[nInput].prevout, nType, nVersion);
54 2014-08-31 17:20:08 <azariah4> yepp that's exactly what I was looking for
55 2014-08-31 17:20:13 <jgarzik> "If programming languages were weapons" http://bjorn.tipling.com/if-programming-languages-were-weapons </off-topic>
56 2014-08-31 17:22:08 <Ademan> haha js is a sword without a hilt
57 2014-08-31 17:23:45 <Ademan> That seems unfair about PHP though, I heard you can create a really great Bitcoin node from scratch with PHP
58 2014-08-31 17:25:23 <sipa> lol
59 2014-08-31 17:27:12 <jtimon> Ademan, I don't get it
60 2014-08-31 17:27:52 <jtimon> did anyone actually tried that?
61 2014-08-31 17:28:14 <jtimon> a php bitcoin node from scratch?
62 2014-08-31 17:34:20 <Ademan> jtimon: yeah, karpeles did
63 2014-08-31 17:35:15 <jtimon> oh, I see
64 2014-08-31 17:35:36 <jtimon> well it was mccaleb who started that, no?
65 2014-08-31 17:41:00 <Ademan> It's possible that the bitcoin-node-in-php was an addition by karpeles, but I don't know
66 2014-08-31 17:43:48 <sipa> Ademan: mtgox's own bitcoin client was in c++ (allegedly)
67 2014-08-31 17:44:24 <Ademan> sipa: hrm, I thought I had heard otherwise, but they may have been referring to the website and I misunderstood
68 2014-08-31 17:44:54 <sipa> the website was php yes
69 2014-08-31 17:45:41 <Adlai> jgarzik: thanks for that :D
70 2014-08-31 17:48:59 <Ademan> I was just trying to figure out why Luke was receiving emails for errors at Gox, but then I realized it was regarding transaction submission
71 2014-08-31 17:52:58 <anddam> can bitcoin-qt be run without the server? the idea would be to use it as a GUI for an already running bitcoind instance
72 2014-08-31 17:53:13 <sipa> ni
73 2014-08-31 17:53:15 <sipa> no
74 2014-08-31 17:53:30 <sipa> you can run it with or without the rpc server
75 2014-08-31 17:53:35 <sipa> but not without the core
76 2014-08-31 17:53:49 <anddam> sipa: I actually appreciated the "ni"
77 2014-08-31 17:54:01 <sipa> i want you to find a shrobbery
78 2014-08-31 17:54:16 <anddam> sure, let me start the dictionary app first
79 2014-08-31 17:55:21 <anddam> sipa: that's a curious request
80 2014-08-31 17:56:38 <anddam> different topic: do I need to explicitly do something after importing a priv key in order to generate the public counterpart?
81 2014-08-31 17:56:51 <anddam> importprivkey at https://en.bitcoin.it/wiki/Original_Bitcoin_client/API_calls_list says "There's no need to import public key, as in ECDSA (unlike RSA) this can be computed from private key"
82 2014-08-31 17:57:20 <sipa> what do you want to do?
83 2014-08-31 17:58:41 <anddam> import a private key
84 2014-08-31 17:59:06 <anddam> oh, I just figured it was shown in https://en.bitcoin.it/wiki/How_to_import_private_keys_v7%2B
85 2014-08-31 17:59:20 <anddam> the public key is automatically calculated, is that correct?
86 2014-08-31 17:59:26 <sipa> yes
87 2014-08-31 17:59:39 <anddam> ok, I just never did that and was curious
88 2014-08-31 17:59:52 <sipa> if you imported the key, bitcoin-qt knows the key
89 2014-08-31 17:59:59 <anddam> about the GUI/core separation is there a plan to move into that direction?
90 2014-08-31 18:00:13 <sipa> yes, but hardly more than a plan
91 2014-08-31 18:00:25 <anddam> sipa: my question was focused on the public key, since I was thinking of importing the private key
92 2014-08-31 18:01:00 <anddam> I mean I see that since public key can be calculated talking about a single key or a pair makes no difference
93 2014-08-31 18:01:26 <anddam> thanks for the info
94 2014-08-31 18:33:38 <jtimon> sipa I made #4754 easier to review
95 2014-08-31 19:32:24 <sipa> jtimon: thabks; will review soon
96 2014-08-31 19:32:39 <jtimon> cool, thanks
97 2014-08-31 19:38:15 <anddam> my client is rescanning the blockchain, is the process cpu or disk bound?
98 2014-08-31 19:38:22 <anddam> I'm on i5 with an SSD
99 2014-08-31 19:38:49 <anddam> what kind of time could I expect with current blockchain height?
100 2014-08-31 19:41:10 <Luke-Jr> anddam: #bitcoin
101 2014-08-31 19:41:20 <anddam> k
102 2014-08-31 20:50:39 <jtimon> in util::GetDataDir(bool fNetSpecific = true) we can cache the path for specific modes/networks
103 2014-08-31 20:51:29 <jtimon> is this really necessary? Can you start testnet and then move to mainnet without restarting bitcoind?
104 2014-08-31 20:51:56 <sipa> no
105 2014-08-31 20:52:21 <sipa> in particular, that would require resetting the blockchain data
106 2014-08-31 20:52:26 <sipa> which isn't really supported
107 2014-08-31 20:52:39 <jtimon> so I can remove that parameter and only cache one path, right?
108 2014-08-31 20:52:39 <sipa> (and is a pain in the ass for tests, so it'd be nice to change that)
109 2014-08-31 20:52:47 <sipa> ah
110 2014-08-31 20:53:00 <sipa> no, the initial config file can be read before the network is known at all
111 2014-08-31 20:53:12 <sipa> that's the reason for the netspecific
112 2014-08-31 20:53:40 <jtimon> mhmm, what about not caching in that first read? I could replace it with a bool cache param
113 2014-08-31 20:54:39 <sipa> wumpus probably knows better why and what those things are for
114 2014-08-31 20:55:17 <jtimon> it would be nice to be able to remove that BaseParams().NetworkID()...
115 2014-08-31 21:33:50 <aljosa> is it possible to use bitcoin libraries to create decentralized network and use it to distribute content and verify content source?
116 2014-08-31 21:35:18 <jtimon> #bittorrent ?
117 2014-08-31 21:37:01 <jtimon> #tahoe-lafs ?
118 2014-08-31 21:42:26 <Luke-Jr> aljosa: of course it's possible, just incredibly stupid and inefficient; that's what BitTorrent, git-annex, and similar technologies are designed for
119 2014-08-31 21:42:42 <Luke-Jr> Bitcoin isn't designed for distributing content or anything like that
120 2014-08-31 21:50:18 <joedoe-> hehe
121 2014-08-31 21:50:33 <joedoe-> dont be like smtp creators ;>
122 2014-08-31 21:51:20 <aljosa> i don't need it to distribute content, i need to guarantee that network agrees which node in network created content and that it can't be altered after ownership is set
123 2014-08-31 21:53:19 <joedoe-> I am not sure although I think similiar project already exists
124 2014-08-31 21:53:19 <joedoe-> w8
125 2014-08-31 21:54:15 <joedoe-> nah, wrong, nevermind
126 2014-08-31 22:07:58 <jtimon> sipa I got it without changing the parameter #4803
127 2014-08-31 22:17:30 <Luke-Jr> [21:51:20] <aljosa> i don't need it to distribute content, i need to guarantee that network agrees which node in network created content and that it can't be altered after ownership is set <-- sounds like timestamping, which is certainly possible with merged-mining, but you have no need for a blockchain