1 2011-07-14 00:07:30 <Kiba> hello folks
  2 2011-07-14 00:52:16 <Zagitta> Why is it that the H value of the blockheader hash must be 0?
  3 2011-07-14 00:54:43 <luke-jr> Zagitta: because all valid difficulties start with the first 32 bits of the hash being 0
  4 2011-07-14 00:55:59 <Zagitta> luke-jr: thanks
  5 2011-07-14 01:17:44 <Zagitta> grr, i still don't get the custom packing used to store the difficulty
  6 2011-07-14 01:33:25 <forrestv> Zagitta, what about it? i can help
  7 2011-07-14 01:34:43 <Zagitta> forresttv: wait hang on a sec, i got an idea
  8 2011-07-14 01:43:43 <Zagitta> forrestv: sanity check... We agree that this hashed header: 8F5325EA73E0D5EB0648593E2615A6F96AC1C224079465A7ECD15935C46A6485 is above difficulty of 1? aka: ffffffffffffffffffffffffffffffffffffffffffffffffffffffff00000000
  9 2011-07-14 01:46:01 <forrestv> Zagitta, difficulty 1 is 0xffff0000000000000000000000000000000000000000000000000000
 10 2011-07-14 01:46:36 <forrestv> but it's higher anyway
 11 2011-07-14 01:47:05 <Zagitta> okay well i'm just trying to see if i can avoid touching nBits
 12 2011-07-14 01:47:33 <forrestv> wait
 13 2011-07-14 01:47:36 <forrestv> what do you mean 'higher'
 14 2011-07-14 01:47:51 <forrestv> do you mean that it qualifies as a block?
 15 2011-07-14 01:47:56 <Zagitta> yes
 16 2011-07-14 01:48:05 <forrestv> because it doesn't. its hash is higher than the target
 17 2011-07-14 01:48:49 <Zagitta> okay good then it's starting to make more sense
 18 2011-07-14 01:50:28 <Zagitta> i should really learn to convert hex to decimal mentally some day
 19 2011-07-14 01:52:24 <Zagitta> forrestv: thanks, now i should be able to avoid submitting any shares to bitcoind unless they actually are blocks
 20 2011-07-14 01:55:11 <lfm> difficulty 1.0 is 00000000ffff0000000000000000000000000000000000000000000000000000
 21 2011-07-14 01:55:23 <lfm> nBits = 0x1d00ffff
 22 2011-07-14 01:56:35 <nanotube> https://en.bitcoin.it/wiki/Difficulty#How_is_difficulty_stored_in_blocks?
 23 2011-07-14 01:58:06 <lfm> I think that compact form was part of the bignum lib, right?
 24 2011-07-14 02:00:56 <Zagitta> @nanotube: that example is absolutly useless
 25 2011-07-14 02:02:56 <Zagitta> https://github.com/bitcoin/bitcoin/blob/master/src/rpc.cpp#L214 is better
 26 2011-07-14 02:10:28 <marvin__> ;;bc,stats
 27 2011-07-14 02:10:30 <gribble> Current Blocks: 136187 | Current Difficulty: 1563027.9961162 | Next Difficulty At Block: 137087 | Next Difficulty In: 900 blocks | Next Difficulty In About: 5 days, 19 hours, 30 minutes, and 0 seconds | Next Difficulty Estimate: 1661834.82925870
 28 2011-07-14 02:10:46 <Joric> reference secp256k1 implementation, removed ctypes dependency https://github.com/joric/pywallet
 29 2011-07-14 02:24:33 <luke-jr> note that difficulty has 2 definitions now :p
 30 2011-07-14 02:32:19 <lfm> you mean the main net difficulty and the share difficulty?
 31 2011-07-14 02:48:12 <denisx> the new poclbm stats do not reflect a different target ;(
 32 2011-07-14 02:49:14 <denisx> is the programmer sometimes in this channel?
 33 2011-07-14 02:50:06 <lfm> ya I think so
 34 2011-07-14 02:50:27 <luke-jr> denisx: they should
 35 2011-07-14 02:50:44 <luke-jr> I personally saw the code to handle it
 36 2011-07-14 02:51:24 <denisx> rr.btcmp.com:7332 [411.277 MH/s (~222 MH/s)] [Rej: 0/42 (0%)]
 37 2011-07-14 02:51:31 <denisx> then I must have a really bad day
 38 2011-07-14 02:52:08 <lfm> rejected 0 of 42, that seems ok
 39 2011-07-14 02:52:33 <luke-jr> denisx: it might not work if the pool changes target
 40 2011-07-14 02:52:42 <luke-jr> but if it's consistent, it should
 41 2011-07-14 02:52:50 <denisx> luke-jr: no, I did not change it on the fly
 42 2011-07-14 02:53:17 <denisx> but that would maybe the next step ;)
 43 2011-07-14 02:55:29 <lfm> why is it two different mh/s?
 44 2011-07-14 02:55:58 <denisx> lfm: because the second value is computed by the number of accepted shares
 45 2011-07-14 02:56:34 <lfm> ok that could be random variation
 46 2011-07-14 02:56:38 <denisx> and it does not reflect that with diff-2 it is worth two shares or 2**33 hashes
 47 2011-07-14 02:57:05 <lfm> oh diff 2.0? that looks pretty normal then
 48 2011-07-14 02:57:58 <lfm> what were you expecting to say it was a bad day?
 49 2011-07-14 02:58:19 <denisx> lfm: if that would be normal I would have a bad day
 50 2011-07-14 02:58:34 <denisx> but I think poclbm can't handle it
 51 2011-07-14 02:59:30 <lfm> it seems you got more than expected
 52 2011-07-14 03:00:19 <denisx> lfm: no, the first value is what my GPU makes, the second is computed from the accepted shares
 53 2011-07-14 03:00:49 <lfm> so with diff  2.0 you earned as if you had 444 mh/s
 54 2011-07-14 03:02:01 <gribble> The expected generation output, at 411.277 Khps, given the supplied difficulty of 2, is 206.837039441 BTC per day and 8.6182099767 BTC per hour.
 55 2011-07-14 03:02:01 <lfm> ;;bc,gend 411.277 2
 56 2011-07-14 03:04:03 <lfm> ;;bc,gen 411.277
 57 2011-07-14 03:04:04 <gribble> The expected generation output, at 411.277 Khps, given current difficulty of 1563027.9961162 , is 0.000264661976567 BTC per day and 1.1027582357e-05 BTC per hour.
 58 2011-07-14 03:04:23 <lfm> ;;bc,gen 411277
 59 2011-07-14 03:04:25 <gribble> The expected generation output, at 411277 Khps, given current difficulty of 1563027.9961162 , is 0.264661976567 BTC per day and 0.011027582357 BTC per hour.
 60 2011-07-14 03:04:32 <gribble> The expected generation output, at 411277 Khps, given the supplied difficulty of 2, is 206837.039441 BTC per day and 8618.2099767 BTC per hour.
 61 2011-07-14 03:04:32 <lfm> ;;bc,gend 411277 2
 62 2011-07-14 03:05:07 <gribble> The average time to generate a block at 411277 Khps, given the supplied difficulty of 2, is 20 seconds
 63 2011-07-14 03:05:07 <lfm> ;;bc,calcd 411277 2
 64 2011-07-14 03:05:54 <lfm> ok by that you should have a lot more then 42 shares in a day, yes
 65 2011-07-14 03:07:16 <denisx> lfm: I think you are totally misunderstanding me
 66 2011-07-14 03:07:28 <lfm> ya probably
 67 2011-07-14 03:12:09 <Zagitta> nothing but coding for 3 days straight is starting to make me sick of it
 68 2011-07-14 03:12:34 <lfm> can't help you with that.
 69 2011-07-14 03:13:28 <Zagitta> lfm: useless as allways then... jkjk :p
 70 2011-07-14 03:16:37 <lfm> Alice Cooper said on the radio tonight Beunos Aries has the most psychiatrists per capita, you may want to move there. Hows that for helpful? (jk)
 71 2011-07-14 03:23:21 <IO-> one of my clients is a real Alice Coopers Town Bar in phx az
 72 2011-07-14 03:23:29 <IO-> the owner said cooper has never once come in
 73 2011-07-14 03:23:37 <IO-> but a lot of other stars have
 74 2011-07-14 03:25:18 <sacarlson> any experts on namecoin code in here or where would they hang out?
 75 2011-07-14 03:25:37 <IO-> sorry can't help
 76 2011-07-14 03:25:53 <IO-> i'm looking up tickets to beunos aries
 77 2011-07-14 03:36:05 <AndyBr> going there now? it's freezing
 78 2011-07-14 03:50:50 <lfm> He claims to be sending his radio show from Arizona
 79 2011-07-14 03:52:33 <IO-> ya he lives here
 80 2011-07-14 03:53:15 <erle-> my gnome doesnt eat the new ico file
 81 2011-07-14 03:53:18 <erle-> in git
 82 2011-07-14 03:53:27 <erle-> no preview, no use as icon
 83 2011-07-14 03:53:37 <erle-> firefox and gimp preview it well
 84 2011-07-14 03:57:34 <AndyBr> oh, about BA having psychiatrists, that's true. they are obsessed about that shit
 85 2011-07-14 03:57:40 <AndyBr> and they have a shit ton of models :D
 86 2011-07-14 04:02:15 <lfm> AndyBr: you live there?
 87 2011-07-14 04:02:38 <AndyBr> lfm: couple of months of the year, yea
 88 2011-07-14 04:02:44 <AndyBr> well, like 3-4 months
 89 2011-07-14 04:03:04 <lfm> when its warm? grin
 90 2011-07-14 04:03:47 <erle-> AndyBr, BA? models?
 91 2011-07-14 04:04:01 <lfm> Beunos Aries
 92 2011-07-14 04:04:50 <AndyBr> buenos aires, capital of argentina
 93 2011-07-14 04:04:54 <erle-> from where have you gone there?
 94 2011-07-14 04:05:11 <AndyBr> yes, when it's warm. it has reversed temperature from where i normally live, which is norway
 95 2011-07-14 04:06:23 <erle-> haha, yea
 96 2011-07-14 04:06:27 <erle-> winter now :)
 97 2011-07-14 04:06:59 <erle-> i have never been more south than the canarian isles
 98 2011-07-14 04:07:08 <Zagitta> If i have the blockheader hash stored in a byte array and want to check for H != 0, what index would i need to check at?
 99 2011-07-14 04:07:13 <AndyBr> yeah. you can't really buy warm clothes there (they dont care), nobody has proper heating, no insulation. -pain-
100 2011-07-14 04:08:41 <lfm> Zagitta: it would be 4 bytes at the most significant end depending how you look at it
101 2011-07-14 04:09:24 <Zagitta> oh yeah the edianess
102 2011-07-14 04:10:11 <lfm> Zagitta: ya the endianness of mining gets kinda complicated and bitcoin doesnt really help by switching it around for some things
103 2011-07-14 04:10:50 <lfm> partly cuz sha256 is kinda defined as bigendian and bitcoin is mostly little endian
104 2011-07-14 04:10:52 <Zagitta> lfm: yeah.... :S
105 2011-07-14 04:11:39 <Zagitta> lfm: okay WTH... now i've tried both ends of the array but i still get nothing but H != 0 hashes
106 2011-07-14 04:12:19 <lfm> ya well only 1 in 4 billion hashes has H=0
107 2011-07-14 04:12:24 <denisx> the something is wrong ;)
108 2011-07-14 04:12:32 <denisx> +n
109 2011-07-14 04:14:04 <Zagitta> lfm: if that's the case then is diablominer/phoenix still reporting them to me?
110 2011-07-14 04:14:35 <lfm> I dont know, I think they only send when they think the hash should be h==0
111 2011-07-14 04:15:19 <lfm> like they check but who knows they might have bugs
112 2011-07-14 04:15:39 <exgran> LOLZ @ #bitcoin-begging
113 2011-07-14 04:15:49 <exgran> cant make an honest buck...
114 2011-07-14 04:15:53 <JFK911> yes please help me bury my grandmother
115 2011-07-14 04:16:02 <JFK911> the undertaker takes btc
116 2011-07-14 04:16:04 <exgran> yikes.
117 2011-07-14 04:16:06 <Zagitta> lfm: well right now i'm suspecting my code more than anything
118 2011-07-14 04:16:07 <lfm> shes gonna die soon?
119 2011-07-14 04:16:29 <lfm> Zagitta: yup, that would seem to be the best bet
120 2011-07-14 04:17:27 <Zagitta> lfm: question just is WHAT i'm doing wrong, there's so many things it could be ><
121 2011-07-14 04:27:33 <Zagitta> good night
122 2011-07-14 05:13:43 <CIA-103> bitcoinj: hearn@google.com * r144 /trunk/pom.xml: Add distribution management section to Maven pom.xml. Patch from Gary Rowe. Update issue 13. http://bitcoinj.googlecode.com/svn-history/r144/
123 2011-07-14 05:33:16 <AndyBr> i'm creating a bitcoin service and need some redundancy in case one box running bitcoind goes down. anyone have experience with this? can i leave wallet.dat on a share/SAN/other shared storage?
124 2011-07-14 05:57:50 <lfm> AndyBr: the wallet has keys for the next 100 transactions if you use new keys for each.
125 2011-07-14 05:58:22 <lfm> after that it could get out of sync with a backup
126 2011-07-14 05:59:10 <AndyBr> lfm: i see. what i meant was running two instances of bitcoind, on separate servers, both accessing the same data dir
127 2011-07-14 06:00:20 <lfm> I am quite sure you could only run one at a time, if they never run at the same time it should be ok
128 2011-07-14 06:00:58 <lfm> the socket for a single bitcoin should be accessible from two clients tho
129 2011-07-14 06:02:17 <AndyBr> right, talking about the server falling down here
130 2011-07-14 06:03:06 <lfm> if the bitcoin server dies then you need the wallet backed up again and it needs to be within the last 100 txn
131 2011-07-14 06:03:12 <AndyBr> the next best solution would be that the servers are swapped in case of failure, starting bitcoind on Server2 if Server1 goes down
132 2011-07-14 06:03:31 <AndyBr> well, the wallet would be stored in a SAN, and be unaffected
133 2011-07-14 06:03:41 <AndyBr> just the operating system on the server would crash
134 2011-07-14 06:04:12 <lfm> ok so long as they dont try to run both servers at once I think it would be ok
135 2011-07-14 06:05:57 <lfm> bitcoin uses a lock file in the data dir to prevent two instances trying to use the same dir
136 2011-07-14 06:07:11 <cuddlefish> ooooh, yes.
137 2011-07-14 06:07:22 <cuddlefish> deal with a supplier went through.
138 2011-07-14 06:07:35 <lfm> grats
139 2011-07-14 06:07:36 <cuddlefish> you can now buy almost ANYTHING with Bitcoins.
140 2011-07-14 06:07:45 <lfm> url?
141 2011-07-14 06:07:51 <cuddlefish> i mean almost ANYTHING you could find in walmart
142 2011-07-14 06:07:58 <cuddlefish> lfm: No website yet.
143 2011-07-14 06:08:15 <lfm> oh so I cant buy almost anything yet
144 2011-07-14 06:08:18 <cuddlefish> you can
145 2011-07-14 06:08:24 <cuddlefish> i'll do it through IRC/forum
146 2011-07-14 06:08:30 <lfm> I want a bycycle
147 2011-07-14 06:08:36 <lfm> bicycle
148 2011-07-14 06:08:42 <cuddlefish> once sec
149 2011-07-14 06:09:57 <lfm> delivered to Calgary (in Canada)
150 2011-07-14 06:12:03 <lfm> We do have Walmarts here tho
151 2011-07-14 06:16:09 <cuddlefish> lfm: Hmm, no bicycles, just parts. probably too big, sorry
152 2011-07-14 06:19:58 <lfm> ok, ill look forward to a web site
153 2011-07-14 06:22:12 <yossarian_> Hi @all.
154 2011-07-14 06:22:32 <yossarian_> I'm trying to install bitcoin version .24 on ubuntu 11.04
155 2011-07-14 06:22:47 <yossarian_> Does anybody know how?
156 2011-07-14 06:22:50 <yossarian_> :)
157 2011-07-14 06:29:53 <yossarian_> Oh. That was easy.
158 2011-07-14 06:29:57 <yossarian_> Thanks anyways. :D
159 2011-07-14 06:33:44 <genewitch> just out of curiosity, why does bitcoin.exe stay running after i tell it to quit in the systray?
160 2011-07-14 06:34:05 <genewitch> like the icon disappears, but it's still in task manager
161 2011-07-14 06:48:08 <lfm> genewitch: sometimes bitcoin takes longer then you might expect to shutdown. just give it a little more time. If it hasnt shut down after a minute or two, let us know.
162 2011-07-14 06:51:50 <genewitch> lfm: it took me longer than a minute or two to log in to screen again
163 2011-07-14 06:51:57 <genewitch> i'll reboot in a second and let you know if it happens again
164 2011-07-14 08:23:30 <prof7bit> "
165 2011-07-14 08:25:16 <justmoon> prof7bit, nobody is disputing that, however nobody has pinpointed the bug yet - and to make the parsing robust against it can't hurt imho
166 2011-07-14 08:27:16 <prof7bit> the magic bytes boundary sequence mechanism isn't implemented correctly anyways, there is nothing preventing the appearance of these bytes randomly anywhere else in the protocol.
167 2011-07-14 08:28:08 <justmoon> prof7bit, normally the magic bytes won't be needed, they only act as a last resort to re-sync if the parser ever goes out of step
168 2011-07-14 08:28:54 <justmoon> theoretically, even if the parser desyncs and then mistakes some data for the magic bytes, it'll reject the message as invalid and then resync with the next occurance of the magic bytes
169 2011-07-14 08:29:27 <justmoon> and again, all that should never happen/be necessary
170 2011-07-14 08:30:31 <prof7bit> yesterday i had my own partial implemntation running for the first time (only version and verack parsing for now, the rest only read and ignored the exact byte numbers according to the header) and it ran for some time and showed me inv and ping and suddenly came out of sync.
171 2011-07-14 08:31:18 <cuddlefish> prof7bit: real men use Construct.
172 2011-07-14 08:31:19 <justmoon> prof7bit, should not happen, there is a bug in your code
173 2011-07-14 08:31:25 <cuddlefish> it is the best parser in the whole wide world
174 2011-07-14 08:31:30 <cuddlefish> it will parse anything
175 2011-07-14 08:31:31 <cuddlefish> ANYTHING
176 2011-07-14 08:31:35 <prof7bit> I haven't looked into the sources yet, I fear they are even more ugly than my own code
177 2011-07-14 08:31:39 <cuddlefish> and look beautiful doing it
178 2011-07-14 08:32:01 <cuddlefish> ;;google construct python parser
179 2011-07-14 08:32:02 <gribble> Construct (python library) - Wikipedia, the free encyclopedia: <http://en.wikipedia.org/wiki/Construct_(python_library)>; 31.1. parser  Access Python parse trees  Python v2.7.2 documentation: <http://docs.python.org/library/parser.html>; Ned Batchelder: Python parsing tools: <http://nedbatchelder.com/text/python-parsers.html>
180 2011-07-14 08:32:10 <justmoon> prof7bit, bitcoinjs logs if it encounters inter-message garbage and I haven't seen that message in months of all kinds of testing on testnet and mainnet
181 2011-07-14 08:32:14 <cuddlefish> first link. the second library's borying
182 2011-07-14 08:35:19 <prof7bit> i looked into the bitcoinj souces today (also for the first time, only to see what they have done and their architecture is strikingly similar to what I have done in Pascal) and I found this source comment. I will do more testing and logging today. maybe it was a bug in my implementation. it SHOULD not (never) happen if a client implements the protocol correctly. there is *NO* way for bytes to randomly appear or disappear on
183 2011-07-14 08:35:20 <prof7bit> a TCP connection.
184 2011-07-14 08:36:40 <justmoon> prof7bit, yep, the version, verack stuff is tricky. you need to keep two protocol version numbers (sendVer and recvVer) and make the checksum handling dependent on those to correctly initiate the connection
185 2011-07-14 08:37:42 <justmoon> cuddlefish, have you implemented bitcoin's var_int with construct? curious to see how that would look
186 2011-07-14 08:37:59 <prof7bit> yes. this alone makes the protocol more ugly by an estimated factor of roughly 2.5
187 2011-07-14 08:38:06 <cuddlefish> justmoon: what does var_int do
188 2011-07-14 08:38:12 <justmoon> prof7bit, agreed :|
189 2011-07-14 08:38:29 <justmoon> cuddlefish, https://en.bitcoin.it/wiki/Protocol_specification#Variable_length_integer
190 2011-07-14 08:38:45 <justmoon> it's an integer type whose length depends on its first byte
191 2011-07-14 08:38:46 <prof7bit> my ReadMessage method could be so simple... but it is not allowed to be.
192 2011-07-14 08:38:59 <cuddlefish> justmoon: Ah, that's trivial
193 2011-07-14 08:39:19 <cuddlefish> construct allows switches
194 2011-07-14 08:39:33 <justmoon> prof7bit, the only way to get it exactly right is just by imitating the official client line by line imho
195 2011-07-14 08:39:57 <justmoon> if you want to have compatibility with ancient versions, it gets even crazier
196 2011-07-14 08:40:17 <cuddlefish> justmoon: "wrong version? drop connection" is the smart way
197 2011-07-14 08:40:25 <cuddlefish> for a non-stupid client
198 2011-07-14 08:40:35 <cuddlefish> satoshi's a brilliant systems engineer but a shitty programmer
199 2011-07-14 08:40:49 <prof7bit> my implementation allows for the exceptions that are mentioned in the wiki page, this seems rather trivial
200 2011-07-14 08:41:29 <prof7bit> it only concerns individual messages, but this thing with two different headers is outright crazy
201 2011-07-14 08:42:58 <justmoon> prof7bit, did you see this yet: if (GetTime() > 1329696000) ?
202 2011-07-14 08:43:02 <justmoon> in net.h
203 2011-07-14 08:43:59 <justmoon> the client will actually change behavior in feb 2012 (specifically it'll stop supporting older clients)
204 2011-07-14 08:47:30 <prof7bit> oh.my.god.
205 2011-07-14 08:48:21 <prof7bit> why did i heva the crazy idea to try to implement this. i ca already see where this will end.
206 2011-07-14 08:48:46 <justmoon> prof7bit, ???_(??)_/???
207 2011-07-14 08:51:38 <lfm> date -d @1329696000
208 2011-07-14 08:51:56 <moa7> so just built 0.3.25 on ubuntu 11.04 no problems .... except the gui doesn't come up when launching ...debug log looks fine ... any ideas
209 2011-07-14 08:51:59 <moa7> ?
210 2011-07-14 08:52:25 <cuddlefish> moa7: oh boy, regression
211 2011-07-14 08:52:25 <erus`> have you tried turning it on and off again?
212 2011-07-14 08:52:55 <moa7> cuddlefish: how's ya mean?
213 2011-07-14 08:52:57 <prof7bit> i spent 3 days already only to try to find the most simple and *elegant* way to detect and instantiate incoming message classes and the code just wont start to lose its ugliness.
214 2011-07-14 08:53:03 <moa7> yep launched several times now
215 2011-07-14 08:53:14 <cuddlefish> moa7: there was the same bug in .21 or .22
216 2011-07-14 08:53:57 <sipa> back then it was a problem with an incompatibility between wx2.9 and some window manager
217 2011-07-14 08:54:05 <justmoon> wait a second, there is a 0.3.25?!
218 2011-07-14 08:54:14 <sipa> justmoon: current git head is 0.3.25
219 2011-07-14 08:54:19 <justmoon> k
220 2011-07-14 08:54:22 <sipa> it will probably be changed to 0.4.0 before release
221 2011-07-14 08:54:29 <justmoon> yeah, that's why I was confused
222 2011-07-14 08:54:42 <sipa> moa7: i believe the bitcoin-release repo has patches for wx to make it compatible
223 2011-07-14 08:54:43 <moa7> just wanted to check out wallet newness stuff.
224 2011-07-14 08:54:48 <moa7> with gui.
225 2011-07-14 08:55:18 <prof7bit> wx is not crap. you just have to know how to use it.
226 2011-07-14 08:55:28 <justmoon> sipa, aren't all gui libraries pretty crazy dependencies?
227 2011-07-14 08:55:29 <prof7bit> wx is a brilliant piece of code
228 2011-07-14 08:55:30 <sipa> wx2.8 would have been fine
229 2011-07-14 08:55:57 <sipa> but depending on a dev release is just asking for trouble with packaging
230 2011-07-14 08:56:11 <moa7> bitcoin-release repo?
231 2011-07-14 08:56:18 <prof7bit> yes. 2.9 was a bad decision. and there was no reason to chose 2.9
232 2011-07-14 08:56:34 <justmoon> sipa, kind of funny, node-bitcoin-p2p used to depend on node's dev release, I eventually had to backport it for current stable :D
233 2011-07-14 08:56:36 <sipa> translation stuff is much nicer in 2.9
234 2011-07-14 08:56:41 <lfm> Satoshi said he used 2.9 due to some unicode support.
235 2011-07-14 08:56:54 <sipa> BlueMatt made a backport to 2.8
236 2011-07-14 08:57:14 <erus`> i dont like the macros in wx
237 2011-07-14 08:57:58 <cuddlefish> erus`: i don't like wx
238 2011-07-14 08:58:05 <sipa> moa7: it's in mainline it seems, see contrib/wx-patches
239 2011-07-14 08:58:18 <sipa> https://github.com/bitcoin/bitcoin/blob/master/contrib/wx-patches/README
240 2011-07-14 08:59:41 <moa7> okay so gotta patch wxwidgets ... may as well go back to my beloved fedora
241 2011-07-14 08:59:48 <prof7bit> and i have never had any unicode problems with wx 2.8, at least not when using it from python.
242 2011-07-14 09:00:00 <sipa> wx2.8 supports unicode just fine
243 2011-07-14 09:00:16 <lfm> moa7: why patch wx? just use 2.9?
244 2011-07-14 09:00:33 <moa7> just running with what is on system ...
245 2011-07-14 09:00:38 <prof7bit> 2.9 has some improvements for Mac. but 2.8 runs on Mac also.
246 2011-07-14 09:00:42 <lfm> not if you patch it
247 2011-07-14 09:00:46 <moa7> to see what real sysadmins have to deal with
248 2011-07-14 09:01:15 <sipa> note that fedora has a patches openssl which removes EC stuff
249 2011-07-14 09:01:18 <sipa> *patched
250 2011-07-14 09:01:40 <moa7> I know so if I'm patching stuff I may as well go and do it all there
251 2011-07-14 09:02:14 <moa7> patching/compiling libs, deps etc
252 2011-07-14 09:02:21 <lfm> moa7: seems to me if you're patching stuff you might as well use the patch called 2.9
253 2011-07-14 09:02:27 <cuddlefish> moa7: go all in and use Gentoo
254 2011-07-14 09:02:40 <sipa> cuddlefish: he was talking about sysadmins, not people with too much time :)
255 2011-07-14 09:04:01 <prof7bit> there is no need to patch wx. the proper way is to work around an undesired behaviour of a lib with code in your own application. and this is true for all libs.
256 2011-07-14 09:04:09 <moa7> lfm: is 11.04 sys wxwidget is 2.8
257 2011-07-14 09:04:22 <cuddlefish> if you're patching a lib you're not using the right lib.
258 2011-07-14 09:05:07 <moa7> devs need to start compiling on clean systems
259 2011-07-14 09:05:13 <moa7> asking for trouble.
260 2011-07-14 09:05:14 <prof7bit> +1
261 2011-07-14 09:05:25 <cuddlefish> sipa: have you put your importing patch into .4?
262 2011-07-14 09:05:38 <sipa> moa7: there is a deterministic virtual machine, reconfigured from scratch for each build
263 2011-07-14 09:05:57 <cuddlefish> sipa: configured? No. List of packages and apt-get
264 2011-07-14 09:06:08 <prof7bit> a typical dev system for testing and release building should have a pretty standard mainstream OS version installed.
265 2011-07-14 09:06:38 <sipa> cuddlefish: yes, that's what i mean
266 2011-07-14 09:06:46 <moa7> okay maybe 11.04 is a little too new.
267 2011-07-14 09:06:50 <prof7bit> bleeding edge experiments are for stuff that will be released in 3 years.
268 2011-07-14 09:06:57 <sipa> cuddlefish: as much as possible is not custom
269 2011-07-14 09:07:13 <sipa> cuddlefish: it will go in 0.4 most likely
270 2011-07-14 09:07:20 <moa7> sipa: thnx for your help.
271 2011-07-14 09:07:21 <lfm> prof7bit: bitcoin will get outa beta in 3 years
272 2011-07-14 09:07:31 <cuddlefish> sipa: yaay.
273 2011-07-14 09:07:34 <prof7bit> it might be dead in 3 years.
274 2011-07-14 09:07:41 <cuddlefish> sipa: i'll give you a 1 BTC private key if it is :P
275 2011-07-14 09:07:48 <sipa> cuddlefish: haha :D
276 2011-07-14 09:07:49 <lfm> ya, itll be dead but it will be outa beta
277 2011-07-14 09:07:50 <prof7bit> i see this danger
278 2011-07-14 09:08:07 <mtrlt> anything might be dead in 3 years
279 2011-07-14 09:08:09 <mtrlt> or tomorrow
280 2011-07-14 09:08:22 <sipa> BlueMatt: did you see JS's suggestion about boost::signal?
281 2011-07-14 09:08:26 <sipa> on the mailinglist
282 2011-07-14 09:09:12 <BlueMatt> yea responding now, though its an entirely separate suggestion than CHub
283 2011-07-14 09:09:17 <prof7bit> maybe i have my (much faster, much more compatible, much more feature-rich) client ready until then and can take over the leadership.
284 2011-07-14 09:09:24 <BlueMatt> his is wrt uis which get stuff from wallet, not from CHub (IMHO)
285 2011-07-14 09:09:41 <wafflefish> BlueMatt: .. what
286 2011-07-14 09:09:42 <justmoon> prof7bit, get in line :P
287 2011-07-14 09:09:42 <lfm> prof7bit: go for it! sounds good
288 2011-07-14 09:10:07 <sipa> BlueMatt: yes yes, it is entirely separate
289 2011-07-14 09:10:34 <sipa> BlueMatt: one is communication between nodedb/net/wallet/mempool, the other is communication between wallet/rpc/ui
290 2011-07-14 09:10:44 <BlueMatt> yep
291 2011-07-14 09:10:48 <sipa> BlueMatt: nonetheless, boost::signal could be used for CHub as well
292 2011-07-14 09:11:27 <sipa> it has priorities, tracking of listeners that have ceased to exist, combination of results, ...
293 2011-07-14 09:20:31 <BlueMatt> hmmm...It doesnt appear to support calling methods of objects instead of just calling a random function
294 2011-07-14 09:20:47 <BlueMatt> which makes the whole thing so much simpler
295 2011-07-14 09:21:16 <sipa> it's one line of code to do the dispatch
296 2011-07-14 09:22:07 <BlueMatt> yea, but becomes 20 lines of code to add each new listener...
297 2011-07-14 09:22:19 <sipa> true
298 2011-07-14 09:22:45 <BlueMatt> could be useful to do the commit stuff that way, but I really prefer to be able to call functions from objects instead of just functions
299 2011-07-14 09:22:57 <sipa> that may be a reason to not use it for CHub
300 2011-07-14 09:23:36 <sipa> or even for wallet listeners
301 2011-07-14 09:23:56 <sipa> but it would rid you of quite some implementation work to write the dispatching
302 2011-07-14 09:24:18 <BlueMatt> the dispatching is a simple foreach...
303 2011-07-14 09:24:31 <sipa> ... for each signal
304 2011-07-14 09:24:36 <BlueMatt> true
305 2011-07-14 09:25:17 <BlueMatt> I like them, but I dont think it quite fits into CHub, but Ill probably use it for doing the dispatching of commits
306 2011-07-14 09:25:21 <sipa> anyway, i don't really care whether boost::signal is used or not, but at least for the wallet listening it would make it very easy to implement it right now
307 2011-07-14 09:25:47 <sipa> meh, either use it for everything or do something yourself for everything within CHub
308 2011-07-14 09:26:11 <BlueMatt> no, Im saying to dispatch the functions internal to CHub
309 2011-07-14 09:26:16 <BlueMatt> not external ones
310 2011-07-14 09:26:59 <sipa> not sure about that, but that's something best left to decide for the one who implements it
311 2011-07-14 09:27:06 <sipa> i.e. you :)
312 2011-07-14 09:27:22 <BlueMatt> lol, ok well I dont plan on writing that part for quite some time so...
313 2011-07-14 09:27:33 <sipa> haha
314 2011-07-14 09:27:34 <BlueMatt> maybe commit it without commit handler threading...
315 2011-07-14 09:27:44 <BlueMatt> then write the threaded parts to that
316 2011-07-14 09:27:48 <sipa> i wouldn't do thread stuff now
317 2011-07-14 09:27:57 <sipa> currently it isn't threaded either, and it works
318 2011-07-14 09:28:12 <BlueMatt> yep
319 2011-07-14 09:29:09 <sipa> anyway, for the wallet listener i would use boost::signal, and implement it very soon
320 2011-07-14 09:29:21 <BlueMatt> you can do that entirely separate from me
321 2011-07-14 09:29:29 <sipa> sure
322 2011-07-14 09:29:38 <sipa> i wasn't implying that you should do that implementation
323 2011-07-14 09:29:46 <sipa> (nor myself)
324 2011-07-14 09:29:50 <BlueMatt> I was implying you should ;)
325 2011-07-14 09:29:55 <sipa> meh
326 2011-07-14 09:30:10 <BlueMatt> wait and see if JS does it...
327 2011-07-14 09:30:13 <WakiMiko_> BlueMatt: trying your latest nightly, options->encrypt wallet->cancel shouldnt produce an error message about a too short passphrase, it should just return the gui
328 2011-07-14 09:30:21 <sipa> haha
329 2011-07-14 09:30:35 <sipa> at least 3rd time you get that bug reported, BlueMatt
330 2011-07-14 09:30:36 <WakiMiko_> also, same for "reenter your passphrase" dialog and the change passphrase dialog
331 2011-07-14 09:30:39 <WakiMiko_> oh
332 2011-07-14 09:30:40 <BlueMatt> yep
333 2011-07-14 09:30:41 <WakiMiko_> im sorry
334 2011-07-14 09:30:42 <WakiMiko_> lol
335 2011-07-14 09:30:43 <sipa> don't be
336 2011-07-14 09:30:52 <BlueMatt> no, it needs fixed
337 2011-07-14 09:30:57 <BlueMatt> just havent gotten around to it yet
338 2011-07-14 09:31:15 <WakiMiko_> alright, also devs: thanks for the huge amount of unpaid work you put in all of this!!
339 2011-07-14 09:31:29 <BlueMatt> sipa: can you add a 0.4 tag to https://github.com/bitcoin/bitcoin/issues/407
340 2011-07-14 09:31:55 <sipa> WakiMiko_: how do you mean unpaid? oh, you didn't know there secretly goes 0.1% of all transactions to us?
341 2011-07-14 09:31:58 <sipa> *oops*
342 2011-07-14 09:32:03 <WakiMiko_> haha
343 2011-07-14 09:32:13 <BlueMatt> all those fees, yea miners dont get those
344 2011-07-14 09:34:22 <WakiMiko_> i just quit bitcoin and restarted it, it didnt ask for a passphrase. does it only ask when sending a tx?
345 2011-07-14 09:34:29 <BlueMatt> yes
346 2011-07-14 09:34:35 <WakiMiko_> makes sense
347 2011-07-14 09:35:46 <BlueMatt> read the README and the commitmsg
348 2011-07-14 09:36:32 <BlueMatt> the commitmsg of https://github.com/TheBlueMatt/bitcoin/commit/4e87d341f75f13bbd7d108c31c03886fbc4df56f and README from https://github.com/TheBlueMatt/bitcoin/commit/b6b039d84ed3d1616cb97ee45ff24ec343efbed0
349 2011-07-14 09:37:22 <WakiMiko_> reading, thanks!
350 2011-07-14 09:38:59 <prof7bit> out of curiosity: is there anywhere a nice graph depicting the block size vs. time?
351 2011-07-14 09:40:45 <BlueMatt> lol that trademark troll who tried to trademark bitcoin already abandoned it...
352 2011-07-14 09:41:06 <prof7bit> or transaction count vs. time?
353 2011-07-14 09:46:41 <prof7bit> s/nice graph/any graph someone made anywhere on the web so i don't have to make on myself/
354 2011-07-14 09:47:10 <sipa> my god #412 speeds startup up
355 2011-07-14 09:47:18 <prof7bit> because i am lazy and it is really not important.
356 2011-07-14 09:48:05 <BlueMatt> sipa: yes very much, go ahead and pull it
357 2011-07-14 09:48:24 <sipa> i was about to
358 2011-07-14 09:55:40 <BlueMatt> god I hate github's bugtracking system...
359 2011-07-14 10:02:53 <phantomcircuit> blueadept, lol
360 2011-07-14 10:02:56 <phantomcircuit> BlueMatt,
361 2011-07-14 10:03:04 <phantomcircuit> god xchats autocomplete is stupid
362 2011-07-14 10:03:18 <phantomcircuit> so i realized something today
363 2011-07-14 10:03:23 <wafflefish> phantomcircuit: because you always want to talk to the person who is first alphabetically.
364 2011-07-14 10:03:36 <phantomcircuit> if i do inverted pair matching the rates wont match up
365 2011-07-14 10:04:21 <phantomcircuit> SELL GBP/BTC SELL BTC/GBP can match perfectly fine (with potentially 1 pip of rounding in the rate)
366 2011-07-14 10:04:38 <wafflefish> phantomcircuit: rounding?
367 2011-07-14 10:04:43 <wafflefish> are you using flo-oats?
368 2011-07-14 10:04:44 <phantomcircuit> but recording the trade is messed up because 1 of them uses rate and the other 1/rate
369 2011-07-14 10:04:53 <phantomcircuit> wafflefish, no fuck floats
370 2011-07-14 10:05:10 <phantomcircuit> if im doing inverted pair matching i have to calculate one of the rates as 1/rate
371 2011-07-14 10:05:26 <phantomcircuit> which almost always will cause a very small rounding error
372 2011-07-14 10:05:30 <phantomcircuit> a maximum of 1 pip
373 2011-07-14 10:05:36 <wafflefish> phantomcircuit: I see
374 2011-07-14 10:05:43 <phantomcircuit> solution? make 1 pip be retarded small
375 2011-07-14 10:06:26 <phantomcircuit> either way the rounding occurs in calculating the want rate and not in calculating the actual used rate
376 2011-07-14 10:06:35 <wafflefish> right
377 2011-07-14 10:06:36 <phantomcircuit> so most of the time it wont actually result in any rounding at all
378 2011-07-14 10:07:23 <phantomcircuit> basically im doing this so we dont need a market maker...
379 2011-07-14 10:09:16 <wafflefish> phantomcircuit: how does that help
380 2011-07-14 10:09:17 <phantomcircuit> wafflefish, like right now if you wanted to sell exactly $20 USD for BTC
381 2011-07-14 10:09:20 <phantomcircuit> you cant do it
382 2011-07-14 10:09:26 <wafflefish> phantomcircuit: sure you can...
383 2011-07-14 10:09:39 <phantomcircuit> lol no you cant
384 2011-07-14 10:09:53 <phantomcircuit> you can approximate the rate and hope that the order goes through immediately
385 2011-07-14 10:10:10 <wafflefish> phantomcircuit: Ah, I see what youuuu mean
386 2011-07-14 10:10:14 <phantomcircuit> but because bitcoins are bought/sold as BTC/USD you can only buy/sell specific amounts of bitcoins
387 2011-07-14 10:10:49 <phantomcircuit> oh snap i know what i can do
388 2011-07-14 10:11:05 <phantomcircuit> (sold|bought)_amount (bought|sold)_amount
389 2011-07-14 10:11:11 <phantomcircuit> base_amount quote_amount
390 2011-07-14 10:11:22 <phantomcircuit> shazam dont store the rate calculate it on the fly
391 2011-07-14 10:11:47 <phantomcircuit> guess he didn't agree
392 2011-07-14 10:13:11 <lfm> I think there is accounting standards for those sorts of calculations
393 2011-07-14 10:13:46 <moa7> denominaire extraordinaire
394 2011-07-14 10:14:18 <lfm> denominaire extraordinaire?
395 2011-07-14 10:14:42 <phantomcircuit> lfm, i dont believe there is actually
396 2011-07-14 10:15:31 <lfm> phantomcircuit: ok maybe there is a multiplicity of standards then! grin
397 2011-07-14 10:15:31 <phantomcircuit> if you're rounding in accounting you're doing something wrong usually
398 2011-07-14 10:15:45 <phantomcircuit> for purposes of calculating sales tax and such things there are rules though
399 2011-07-14 10:16:15 <lfm> phantomcircuit: ya, thats what I mean. like they must have standards for exchange rates and whatnot
400 2011-07-14 10:16:18 <phantomcircuit> there are rules for brokers accepting orders, if the order would result in rounding it should be rejected
401 2011-07-14 10:16:48 <phantomcircuit> like for USD you can legally only place trades with 4 places after the decimal for amounts < 1.00
402 2011-07-14 10:17:00 <phantomcircuit> so 1.0001 is an illegal amount for a broker to accept
403 2011-07-14 10:17:04 <phantomcircuit> but 0.9999 isn't
404 2011-07-14 10:30:20 <sipa> for both
405 2011-07-14 10:30:34 <BlueMatt> sipa: odd, CIA typically posts them all I thought
406 2011-07-14 10:30:39 <BlueMatt> the merge and all the commits
407 2011-07-14 10:31:10 <sipa> i've found it to be quite inconsistent :)
408 2011-07-14 10:31:19 <BlueMatt> clearly
409 2011-07-14 10:40:40 <moa7> what's up with 3-letter codes ... comes from days of teletypes and 2K mem limits
410 2011-07-14 10:40:45 <prof7bit> exactly.
411 2011-07-14 10:40:55 <prof7bit> xau <-- spot gold
412 2011-07-14 10:41:08 <lfm> moa7: ya, I thot XZBTC would be fine for the standards wonks
413 2011-07-14 10:41:48 <prof7bit> they have only 6 bytes reserved in their monstrous java servers
414 2011-07-14 10:41:50 <moa7> BTCH
415 2011-07-14 10:41:55 <lfm> like that have to fit all their stuff onto 80 col punch card or something
416 2011-07-14 10:42:15 <moa7> lol
417 2011-07-14 10:42:28 <phantomcircuit> they probably do
418 2011-07-14 10:42:46 <phantomcircuit> old systems like that will incorporate every technology used in any way over the past 30 years
419 2011-07-14 10:42:50 <UukGoblin> we need to fit it on 80 col amphipod message line
420 2011-07-14 10:43:03 <lfm> they lost to cobol code for international exchange back in 1984
421 2011-07-14 10:43:09 <lfm> source
422 2011-07-14 10:43:29 <phantomcircuit> lol
423 2011-07-14 10:44:02 <phantomcircuit> citi groups intercountry internal clearing house uses shit from the early 80s
424 2011-07-14 10:44:05 <phantomcircuit> it's a bad joke
425 2011-07-14 10:48:50 <prof7bit> as long as it works its still ok and cheaper than reimplementing it by young programmers who cannot program anymore nowadays. cobol compilers still exist.
426 2011-07-14 10:48:55 <Guest72992> Project Bitcoin-Test build #7: STILL FAILING in 7.4 sec: http://www.bluematt.me/jenkins/job/Bitcoin-Test/7/
427 2011-07-14 10:48:56 <Guest72992> * dev: Single DB transaction for all addresses in a message
428 2011-07-14 10:49:45 <phantomcircuit> prof7bit, i realized halfway through building this exchange that doing it in c++ would have been easier than python
429 2011-07-14 10:49:48 <phantomcircuit> and i was shocked
430 2011-07-14 10:50:22 <prof7bit> C++ is not easier than *anything*. you must be making a mistake.
431 2011-07-14 10:50:34 <phantomcircuit> there isn't a good heap implementation in python
432 2011-07-14 10:50:57 <prof7bit> maybe itrs easier than brainfuck or perl but not much more
433 2011-07-14 10:51:00 <phantomcircuit> and when you look at the language dealing with cheap pointers isn't possible
434 2011-07-14 10:51:55 <phantomcircuit> the heap implementation is based on sorting a list into a heap
435 2011-07-14 10:51:58 <phantomcircuit> it's pretty weird
436 2011-07-14 10:52:02 <lfm> APL?
437 2011-07-14 10:52:26 <phantomcircuit> python
438 2011-07-14 10:52:32 <lfm> you never heard of heapsort?
439 2011-07-14 10:52:33 <prof7bit> Ada or Pascal maybe?
440 2011-07-14 10:52:49 <phantomcircuit> lfm, heapsort is nice
441 2011-07-14 10:52:53 <phantomcircuit> but i need an actual heap
442 2011-07-14 10:53:06 <phantomcircuit> https://secure.wikimedia.org/wikipedia/en/wiki/Heap_(data_structure)
443 2011-07-14 10:53:09 <UukGoblin> malbolge
444 2011-07-14 10:53:12 <lfm> like a priority queue?
445 2011-07-14 10:53:21 <prof7bit> Pascal for example has the same features as C++ but is orders of magnitudes less error prone and easier to read
446 2011-07-14 10:53:23 <b4epoche> phantomcircuit:  you building an exchange?
447 2011-07-14 10:53:23 <phantomcircuit> lfm, yes... but not a queue
448 2011-07-14 10:53:34 <phantomcircuit> yes
449 2011-07-14 10:53:37 <b4epoche> is it at all 'open' yet?
450 2011-07-14 10:53:50 <phantomcircuit> not yet
451 2011-07-14 10:54:06 <phantomcircuit> there isn't anything of interest really yet
452 2011-07-14 10:54:21 <UukGoblin> prof7bit, pascal has a precompiler? or a templating engine?
453 2011-07-14 10:54:31 <prof7bit> pascal has generics
454 2011-07-14 10:54:32 <phantomcircuit> users/accounts/limit & market orders
455 2011-07-14 10:54:42 <phantomcircuit> but i haven't finished the actual order matching yet
456 2011-07-14 10:54:48 <b4epoche> I'm making a "Trader" edition of the OSX client that connects to exchanges for 'in-app' trading
457 2011-07-14 10:54:50 <phantomcircuit> because i insisted on allowing inverted pairings
458 2011-07-14 10:54:52 <phantomcircuit> which is making this like
459 2011-07-14 10:54:57 <phantomcircuit> a million times harder
460 2011-07-14 10:55:07 <UukGoblin> and it definitely didn't have all the features c++ had then
461 2011-07-14 10:55:09 <UukGoblin> but srsly
462 2011-07-14 10:55:11 <UukGoblin> malbolge
463 2011-07-14 10:55:13 <UukGoblin> !
464 2011-07-14 10:55:22 <UukGoblin> best. language. evarr
465 2011-07-14 10:55:26 <prof7bit> 10 years is a long time. we have 2011 now
466 2011-07-14 10:56:04 <lfm> prof7bit: what version of pascal are you talking about?
467 2011-07-14 10:56:06 <prof7bit> the free pascal compiler is developed rapidly and actively, about a dozen commits a day, each day
468 2011-07-14 10:56:58 <prof7bit> 13 years development, 5 million lines of code, all major OS and architectures, huge standard library for all purposes
469 2011-07-14 10:57:29 <lfm> prof7bit: ok so you're not talking about Wirth's original Pascal any more.
470 2011-07-14 10:57:36 <prof7bit> no. its 2011
471 2011-07-14 10:57:49 <UukGoblin> really?
472 2011-07-14 10:57:51 <lfm> because I knew it
473 2011-07-14 10:57:54 <prof7bit> noboy is using 1980 C also
474 2011-07-14 10:57:56 <b4epoche> Fortran77 ftw!
475 2011-07-14 10:58:30 <UukGoblin> malbolge
476 2011-07-14 10:58:36 <UukGoblin> the one and only
477 2011-07-14 10:58:53 <prof7bit> all are referring to what gcc does nowadays when talking about C++, so I am refering to FPC when referring to Pascal. Older versions are history.
478 2011-07-14 11:00:13 <lfm> I guess I shouldnt try to claim I can program pascal any more.
479 2011-07-14 11:00:20 <prof7bit> we have generics, we have interfaces, we have reference counted strings or arbitrary size, etc.
480 2011-07-14 11:00:37 <phantomcircuit> SELL GBP/BTC 10 @ 0.1 SELL BTC/GBP 10 @ 10.0
481 2011-07-14 11:00:43 <phantomcircuit> like those orders perfectly match
482 2011-07-14 11:01:38 <lfm> is that 10 btc or 10 gbp?
483 2011-07-14 11:02:07 <phantomcircuit> 2 orders
484 2011-07-14 11:02:21 <phantomcircuit> selling 10 GBP for BTC 0.1 BTC per GBP
485 2011-07-14 11:02:34 <phantomcircuit> selling 10 BTC for GBP 10 GBP per BTC
486 2011-07-14 11:02:34 <prof7bit> lfm: your old pascal knowledge is not lost. you can still use it. the language was extended in such a way that it did not break anything and not destroy the simplicity and elegance.
487 2011-07-14 11:03:01 <prof7bit> it all fits seamlessly into the language
488 2011-07-14 11:03:19 <prof7bit> it feels completely natural
489 2011-07-14 11:03:38 <phantomcircuit> lol
490 2011-07-14 11:03:46 <phantomcircuit> lfm, see what im saying though about the rates?
491 2011-07-14 11:04:07 <phantomcircuit> you convert the earlier one to a buy and fulfill them
492 2011-07-14 11:04:29 <lfm> phantom ya the rates match but the amounts dont seem to match, wouldnt on of the amounts be 1 and the other 10?
493 2011-07-14 11:04:59 <phantomcircuit> yeah
494 2011-07-14 11:05:05 <phantomcircuit> whatever partial matches
495 2011-07-14 11:05:10 <lfm> ok
496 2011-07-14 11:05:15 <phantomcircuit> you cant have a non partial order in the orderbook anyways,
497 2011-07-14 11:05:28 <phantomcircuit> FOK/IOC/GTC
498 2011-07-14 11:06:16 <phantomcircuit> oh did i mention suports FOK/IOC?
499 2011-07-14 11:06:45 <lfm> international olympic commitee?
500 2011-07-14 11:07:18 <phantomcircuit> immediate or cancel
501 2011-07-14 11:07:49 <lfm> oh, nice, i guess.
502 2011-07-14 11:08:18 <phantomcircuit> so you can place a "market" order with a limit
503 2011-07-14 11:08:44 <UukGoblin> hm, btw, can anyone sensibly read the whole output of "free"?
504 2011-07-14 11:08:51 <b4epoche> phantomcircuit:  when you get up and running let me know so I can integrate your api into my app
505 2011-07-14 11:08:57 <prof7bit> and if you are looking for a nice IDE / RAD-tool for pascal be sure to check out Lazarus (the windows download of Lazarus already comes bundled completely with compiler, debugger and everything in a simple setup.exe->next->next->finish, on the other platforms you install the compiler separately)
506 2011-07-14 11:09:03 <UukGoblin> it reports buffers and cache on the first line, then on the second line, all the figures are different, and none of them add up
507 2011-07-14 11:09:52 <prof7bit> Note to C++ users: don't do that or you might lose interest in C++ within 2 weeks exposure to Lazarus/FPC
508 2011-07-14 11:10:15 <phantomcircuit> windows
509 2011-07-14 11:10:20 <phantomcircuit> is this some kind of joke?
510 2011-07-14 11:10:28 <phantomcircuit> gtfo
511 2011-07-14 11:10:30 <prof7bit> same warning for Java / Netbeans fans
512 2011-07-14 11:10:44 <phantomcircuit> <-- notepad
513 2011-07-14 11:11:11 <lfm> A fine editor so long as you dont want to edit more than 64 k
514 2011-07-14 11:11:14 <phantomcircuit> piratpad
515 2011-07-14 11:11:24 <phantomcircuit> ok i lied
516 2011-07-14 11:11:25 <phantomcircuit> gedit
517 2011-07-14 11:11:33 <prof7bit> phantomcircuit: please elaborate on the "joke" thing, i don't get your point.
518 2011-07-14 11:11:38 <lfm> sed?
519 2011-07-14 11:11:53 <sipa> sed is a stream editor
520 2011-07-14 11:11:54 <phantomcircuit> BUTTERFLY WINGS
521 2011-07-14 11:12:01 <sipa> if you want to edit files, use ed
522 2011-07-14 11:12:11 <justmoon> phantomcircuit, default color scheme? (gedit "classic")
523 2011-07-14 11:12:16 <sipa> or a steady hand and a magnetic needle
524 2011-07-14 11:12:22 <phantomcircuit> sipa, ++
525 2011-07-14 11:12:38 <lfm> edlin?
526 2011-07-14 11:13:30 <phantomcircuit> lol i really suck at poker
527 2011-07-14 11:13:38 <phantomcircuit> i cant do even basic arithmetic in my head
528 2011-07-14 11:16:14 <phantomcircuit> so anyways
529 2011-07-14 11:16:16 <phantomcircuit> like i was saying
530 2011-07-14 11:16:36 <phantomcircuit> SELL GBP/BTC 10 @ 0.1
531 2011-07-14 11:16:43 <phantomcircuit> SELL BTC/GBP 0.1 @ 10.0
532 2011-07-14 11:16:48 <sipa> ;;bc,blocks
533 2011-07-14 11:16:49 <gribble> 136241
534 2011-07-14 11:17:17 <phantomcircuit> the rate is inverted in 1 and the same in the other
535 2011-07-14 11:17:40 <phantomcircuit> so i cant record the actual rate in the "native" pairing for both orders
536 2011-07-14 11:17:47 <phantomcircuit> which is more than a little annoying
537 2011-07-14 11:17:49 <prof7bit> you are introducing confusion into the game
538 2011-07-14 11:18:54 <phantomcircuit> prof7bit, yes but the normal use case for "normal" users is to either sell all their fiat or sell all their btc
539 2011-07-14 11:19:00 <prof7bit> a trader only wants to have to do with either XXXYYY or YYYXXX but not be confused by mixing these
540 2011-07-14 11:19:07 <phantomcircuit> with a single pairing we couldn't do both
541 2011-07-14 11:19:49 <phantomcircuit> prof7bit, see the problem?
542 2011-07-14 11:20:11 <lfm> also a trader will accept an exchange which is BETTER than what he asked for
543 2011-07-14 11:20:15 <prof7bit> I am trading a lot of EURUSD (for fun and profit) and i never even *think* about a hypothetical USDEUR
544 2011-07-14 11:20:49 <phantomcircuit> yeah but you're trading with quotes right?
545 2011-07-14 11:21:12 <phantomcircuit> without quotes you're stuck just guessing
546 2011-07-14 11:21:23 <prof7bit> yes. The EURUSD quotes as quoted by my broker. I never needed an USDEUR quote for anything.
547 2011-07-14 11:21:48 <phantomcircuit> yeah but i mean you have a quote, so you can calculate the exact amount of USD you get by selling an amount of EUR
548 2011-07-14 11:21:52 <phantomcircuit> and vice versa
549 2011-07-14 11:21:57 <phantomcircuit> without quotes you cant do that
550 2011-07-14 11:23:15 <prof7bit> i don't sell USD for Euros ;-) I only have dollars
551 2011-07-14 11:23:31 <prof7bit> but i might buy euros and sell them again
552 2011-07-14 11:23:42 <prof7bit> or borrow them and sell even more
553 2011-07-14 11:24:06 <lfm> so usd is your base currency
554 2011-07-14 11:24:38 <lfm> and you might be trading with someone whose base currency is eur but it doesnt really matter
555 2011-07-14 11:24:39 <prof7bit> its my account currency and i only think in terms of dollars whenever I log into my platform
556 2011-07-14 11:25:44 <phantomcircuit> yes but you can only do that because you're dealing with quotes
557 2011-07-14 11:25:58 <phantomcircuit> if you where placing orders directly on an exchange you would have to worry about it
558 2011-07-14 11:26:36 <phantomcircuit> or am i missing something here?
559 2011-07-14 11:26:38 <phantomcircuit> i dont think i am
560 2011-07-14 11:27:47 <coderrr> has anyone compiled a 32bit client on 64bit linux ?
561 2011-07-14 11:28:02 <prof7bit> the allowable volumes are quoted only in terms of one of the currencies, in EUR in this case. if i place an order i either buy or sell X EUR (in exchange for dollars). I dont place orders to buy or sell y dollars for euros.
562 2011-07-14 11:28:45 <phantomcircuit> yes but you know the rate
563 2011-07-14 11:29:00 <phantomcircuit> so you *can* calculate it
564 2011-07-14 11:29:04 <prof7bit> i know how many UR it is by simply counting the euros i hold and i know how many dollars it would be by looking at the current quote.
565 2011-07-14 11:29:09 <forrestv> how does the testnet-in-a-box work?
566 2011-07-14 11:29:09 <prof7bit> EUR
567 2011-07-14 11:29:11 <phantomcircuit> x EUR * y rate = z USD
568 2011-07-14 11:29:19 <forrestv> how does it keep the difficulty low with a stock client?
569 2011-07-14 11:29:23 <phantomcircuit> since y is quoted it's fixed and you can solve for x
570 2011-07-14 11:29:45 <phantomcircuit> but on a fluid exchange you couldn't do that
571 2011-07-14 11:29:55 <phantomcircuit> you'd be guessing, hoping to hit a moving target
572 2011-07-14 11:30:15 <prof7bit> if i have one lot then this is 100000 eur. no need for calculting here, i see it directly. if I want to know the dollars then I simply multiply with the quote.
573 2011-07-14 11:30:29 <phantomcircuit> exactly
574 2011-07-14 11:30:43 <phantomcircuit> if you where placing that order directly with an exchange you would not have a quote
575 2011-07-14 11:30:49 <phantomcircuit> just a limit
576 2011-07-14 11:30:53 <phantomcircuit> make sense?
577 2011-07-14 11:31:43 <prof7bit> i would for example say buy 1 Lot @ 1.3800 or i would say sell 1 Lot @ 1.4300
578 2011-07-14 11:32:04 <phantomcircuit> yes
579 2011-07-14 11:32:04 <prof7bit> Lot is always EUR, the quote is always USD
580 2011-07-14 11:32:15 <phantomcircuit> but you're buying at *exactly* that amount, because you have a quote
581 2011-07-14 11:32:30 <prof7bit> you can't mix this in the same orderbook
582 2011-07-14 11:32:38 <phantomcircuit> but if instead you placed an order on the actual exchange you would get your rate or better
583 2011-07-14 11:33:52 <phantomcircuit> prof7bit, lol see what im saying though?
584 2011-07-14 11:34:12 <phantomcircuit> the features that traders want are directly counter to the basic operations that normal users want
585 2011-07-14 11:34:20 <prof7bit> Not really sure. You still have the resulting average quote in USD always and never in EUR
586 2011-07-14 11:34:53 <phantomcircuit> prof7bit, ok lets try this another way
587 2011-07-14 11:35:06 <phantomcircuit> prof7bit, you have some EUR and need USD to fill a short
588 2011-07-14 11:35:16 <prof7bit> you must decide for one of the two options. either run the orderbook and the quotes and the charts in XXX/YYY or do it in YYY/XXX
589 2011-07-14 11:35:31 <phantomcircuit> you go to your broker and get a quote, then you solve for how much EUR you have to sell to cover the short, right?
590 2011-07-14 11:36:20 <phantomcircuit> but what if you *cant* get a quote?
591 2011-07-14 11:36:27 <phantomcircuit> you can *only* place an order
592 2011-07-14 11:36:53 <phantomcircuit> get what im saying?
593 2011-07-14 11:37:05 <prof7bit> i can say here I have 1 Lot of EUR, sell them for USD at this dollar price. or I say i need 1 Lot of EUR buy than at that dollar price. i'm always using the dollar price.
594 2011-07-14 11:37:23 <phantomcircuit> yeah i get that
595 2011-07-14 11:37:57 <phantomcircuit> what im saying is that doing it that way limits the accuracy of trying to exchange EUR for USD
596 2011-07-14 11:38:32 <prof7bit> or I would say here i have 1 BTC, sell them at x$ or buy 1BTC at x$ but not buy x$ at yBTC
597 2011-07-14 11:39:20 <phantomcircuit> yes
598 2011-07-14 11:39:35 <phantomcircuit> which means if you have to cover a short, you're forced to over shoot it right
599 2011-07-14 11:39:58 <prof7bit> it would create confusion and rounding errors and odd lot sizes in the order book. thats why it is never done.
600 2011-07-14 11:42:23 <phantomcircuit> prof7bit, rounding errors would be limited to 1 pip on the rate
601 2011-07-14 11:42:31 <phantomcircuit> but yeah i see your point
602 2011-07-14 11:42:40 <prof7bit> the problem with being able to cover shorts (or related things) before bad things happen in a thin market is only relevant if the market is thin and bad things might happen
603 2011-07-14 11:43:26 <prof7bit> and if you are borrowing money (leverage)
604 2011-07-14 11:44:48 <prof7bit> this might end in a disaster, either for the trader or for the dealer, depending on what is written in your contract
605 2011-07-14 11:45:52 <prof7bit> but it might still need a while until we see leveraged bitcoin trading.
606 2011-07-14 11:46:45 <prof7bit> the first few experiments in this direction might  surely be funny to watch (from a distance)
607 2011-07-14 11:51:24 <phantomcircuit> prof7bit, heh
608 2011-07-14 11:52:06 <prof7bit> if a dealer tries to hedge his exposure to millions of leveraged bitcoins in the spot market
609 2011-07-14 11:52:57 <phantomcircuit> anyways the first thing most people do is try and sell all of their bitcoins
610 2011-07-14 11:53:05 <phantomcircuit> or use all their fiat to buy bitcoins
611 2011-07-14 11:53:22 <phantomcircuit> there's no way to cover both angles with a normal system without lots of guessing
612 2011-07-14 11:56:44 <prof7bit> you generally cannot predict exactly what happens in a market if you start buying or selling non-trivial amounts. other participants will see this and react to it by placing/canceling their own orders. you can only guess when looking at the orderbook before you place an order.
613 2011-07-14 11:57:53 <moa7> wisdom of the crowd
614 2011-07-14 11:58:24 <moa7> or mob mentality?
615 2011-07-14 11:59:07 <Eliel_> both? :D
616 2011-07-14 12:00:22 <moa7> a capricious beast
617 2011-07-14 12:12:27 <phantomcircuit> prof7bit, hmm i guess
618 2011-07-14 12:39:25 <phantomcircuit> prof7bit, balls i dont know what to do now
619 2011-07-14 12:39:33 <phantomcircuit> i could modify this to be a normal forex exchange easily
620 2011-07-14 12:44:53 <phantomcircuit> prof7bit, you still there?
621 2011-07-14 12:45:29 <prof7bit> sorry, was inside my IDE...
622 2011-07-14 12:46:07 <phantomcircuit> prof7bit, so which would you vote for GBP/BTC or BTC/GBP
623 2011-07-14 12:46:23 <phantomcircuit> im just going to bite the bullet and move this all to a normal exchange format
624 2011-07-14 12:46:42 <phantomcircuit> and implement the behavior in the ui to approximate buy/sell of the quote
625 2011-07-14 12:47:58 <phantomcircuit> i guess BTC/GBP
626 2011-07-14 12:48:01 <phantomcircuit> just to be cheaky
627 2011-07-14 12:48:04 <phantomcircuit> cheeky
628 2011-07-14 12:48:41 <prof7bit> i would vote for BTCGBP (price of BTC quoted in GBP), just to not confuse the users. Maybe you find a way to implement the other side (or some easy to use calculator) in some easy to understand way later
629 2011-07-14 12:48:43 <Joric> dear mother of god, bitcoinj has 6 megabytes of ec2 related sources
630 2011-07-14 12:48:58 <Joric> i only wrote like 50 lines
631 2011-07-14 12:49:02 <TD> ec2?
632 2011-07-14 12:49:24 <TD> you mean ecdsa right
633 2011-07-14 12:49:26 <Joric> i used reference i2d_ECPrivateKey/i2o_ECPublicKey implementation
634 2011-07-14 12:49:50 <TD> bitcoinj includes a complete copy of the bouncy castle library
635 2011-07-14 12:49:57 <Joric> there's nothing special really, python supports big numbers out of the box
636 2011-07-14 12:50:01 <TD> it might be taken out in favor of shading the library at compile time
637 2011-07-14 12:50:10 <Joric> yeah, probably
638 2011-07-14 12:50:27 <TD> i don't see your point though. including the whole crypto library is easier and better than reimplementing the ECDSA algorithms
639 2011-07-14 12:55:58 <mmartian_> ;;bc,stats
640 2011-07-14 12:56:00 <gribble> Current Blocks: 136255 | Current Difficulty: 1563027.9961162 | Next Difficulty At Block: 137087 | Next Difficulty In: 832 blocks | Next Difficulty In About: 5 days, 8 hours, 43 minutes, and 44 seconds | Next Difficulty Estimate: 1655426.95092866
641 2011-07-14 13:27:42 <gmaxwell> BlueMatt: if you're around, can you try to probe me again? I yelled at my ISP...
642 2011-07-14 13:28:24 <nanotube> hehe
643 2011-07-14 13:28:41 <BlueMatt> probing...
644 2011-07-14 13:28:52 <BlueMatt> Connection timed out
645 2011-07-14 13:29:11 <gmaxwell> alas. Okay.
646 2011-07-14 13:29:26 <BlueMatt> nope, bumped connection time out, and its back to the same as before
647 2011-07-14 13:29:29 <nanotube> i can try telneting to you to port 8333 gmaxwell if you like
648 2011-07-14 13:29:31 <BlueMatt> no packets after syn ack
649 2011-07-14 13:29:40 <phantomcircuit> IMA PROBE THE SHIT OUT OF... er nvm
650 2011-07-14 13:30:49 <BlueMatt> oh now its working...
651 2011-07-14 13:30:59 <BlueMatt> gmaxwell: ???
652 2011-07-14 13:32:16 <gmaxwell> Yea, its still falling over when there are high numbers of connections. I bumped everyone off, and capped it down. oh well. more investigation later.
653 2011-07-14 13:32:39 <BlueMatt> m have fun
654 2011-07-14 13:32:42 <BlueMatt> maybe its your patches?
655 2011-07-14 13:33:52 <gmaxwell> Anything is possible.
656 2011-07-14 13:35:38 <prof7bit> the client should allow incoming local p2p connections even if it has already max connections to other peers (to make testing easier)
657 2011-07-14 13:35:43 <gmaxwell> er, well, I know its not bitcoin itself because I can succesfully connect to it all day long locally.
658 2011-07-14 13:35:55 <BlueMatt> mmm
659 2011-07-14 13:36:05 <gmaxwell> prof7bit: to make DOS attack easier, you mean? :)
660 2011-07-14 13:36:29 <prof7bit> dos attack on my own pc, me attacking myself?
661 2011-07-14 13:37:14 <prof7bit> i want my client to connect the official client but it wont let me in because it is full already.
662 2011-07-14 13:38:10 <prof7bit> i am not at the point where i can test it against other people's clients in the network, i just want it to connect my own client
663 2011-07-14 13:39:14 <gmaxwell> prof7bit: then run a node thats firwalled off from inbound connections (and perhaps isolated) for that purpose. If you're testing something new you should really test against testnet instead.
664 2011-07-14 13:39:55 <prof7bit> i need something that will bombard me with inv messages now, testnet is too quiet.
665 2011-07-14 13:43:06 <dikidera> i've been thinking of how to do it and it almost seems impossible
666 2011-07-14 13:43:18 <dikidera> each thread on a cpu miner to work on the same getwork
667 2011-07-14 13:44:16 <dikidera> every possible piece of code i think about, in the end doesnt seem like it will work like how i want it
668 2011-07-14 13:44:38 <b4epoche> why is anyone thinking about cpu  mining?
669 2011-07-14 13:44:46 <dikidera> anything better to do?
670 2011-07-14 13:45:15 <b4epoche> I've got plenty better to do&  but maybe you don't ;-)
671 2011-07-14 13:47:10 <dikidera> deja vu...
672 2011-07-14 13:47:36 <dikidera> i've probably done this in a parallel universe already...
673 2011-07-14 13:47:59 <Joric> i accidentaly the whole wallet
674 2011-07-14 13:48:12 <denisx> so, poclbm will be fixed in the next release to show stats correctly with diff-2
675 2011-07-14 13:48:41 <Joric> looks like 0.3.25 goes to infinite loop if both 'skey' and 'key' are presented
676 2011-07-14 13:48:48 <dikidera> heehaw
677 2011-07-14 13:48:56 <BlueMatt> Joric: there is no such thing as a skey
678 2011-07-14 13:48:56 <dikidera> Son, i am disappoint
679 2011-07-14 13:49:02 <BlueMatt> dikidera: stfu
680 2011-07-14 13:49:03 <Joric> ckey, whatever
681 2011-07-14 13:49:14 <BlueMatt> Joric: no, it returns null, or should...
682 2011-07-14 13:49:23 <BlueMatt> uh, false sorry
683 2011-07-14 13:49:59 <Joric> had to add version checking to avoid accidentaly the whole wallet in the future
684 2011-07-14 13:50:20 <BlueMatt> "to avoid accidentaly the whole wallet" makes no sense
685 2011-07-14 13:50:30 <BlueMatt> and there already is version checking
686 2011-07-14 13:50:36 <dikidera> denisx:i found that doing diff 2 shares results in a few percent more hashes, which are sent to bitcoinD via pushpool
687 2011-07-14 13:51:03 <BlueMatt> lol
688 2011-07-14 13:51:10 <dikidera> *few percent more hashes, which meet the conditions to be sent to bitcoin
689 2011-07-14 13:51:17 <BlueMatt> lol
690 2011-07-14 13:51:19 <dikidera> compared to doing diff-1
691 2011-07-14 13:51:24 <BlueMatt> lol
692 2011-07-14 13:51:27 <denisx> dikidera: no, impossible
693 2011-07-14 13:51:31 <dikidera> i guarantee it
694 2011-07-14 13:51:34 <BlueMatt> lol
695 2011-07-14 13:51:53 <denisx> dikidera: ok, then join my pool and hash there! ;)
696 2011-07-14 13:53:48 <BlueMatt> Joric: can you provide more info, there is no infinite loop possibility (there is no loop) in loading ckeys or keys
697 2011-07-14 13:53:56 <denisx> dikidera: and if you bring in 5GH/s I will start a diff-4 port for you! ;)
698 2011-07-14 13:55:59 <BlueMatt> at that rate, you should be mining at real diff, get a couple blocks/sec
699 2011-07-14 14:03:00 <Joric> BlueMatt, http://pastebin.com/eT4mrkbv don't know what i've done but it refuses to load with that wallet, eats 0% cpu and just hangs there
700 2011-07-14 14:03:31 <BlueMatt> can you give a bitcointools dump?
701 2011-07-14 14:03:55 <Joric> hang on
702 2011-07-14 14:04:48 <Joric> http://pastebin.com/7uMLi6Ps
703 2011-07-14 14:05:44 <Joric> note the unencrypted key in the beginning
704 2011-07-14 14:06:17 <BlueMatt> is that the latest bitcoin?
705 2011-07-14 14:06:23 <Joric> nightly build
706 2011-07-14 14:06:24 <BlueMatt> bitcointools
707 2011-07-14 14:06:27 <BlueMatt> sorry
708 2011-07-14 14:06:41 <Joric> bitcointools are fresh
709 2011-07-14 14:06:56 <BlueMatt> oh gavin did a different addrIncoming handling...
710 2011-07-14 14:07:24 <BlueMatt> yea that will fail to load but it will return...
711 2011-07-14 14:07:31 <BlueMatt> (unless something broke during merge)
712 2011-07-14 14:08:04 <Joric> do you mind if i'll send this wallet.dat to you, it's unused
713 2011-07-14 14:08:17 <moa7> sipa: fwiw, 0.3.25 on ubuntu 11.04 wxWidgets 2.9.2 doesn't need patch that 2.9.1 did
714 2011-07-14 14:10:41 <b4epoche> BlueMatt:  I'm feeling like doing some rebasing&  can you help?  and I promise I'll take notes this time
715 2011-07-14 14:10:57 <BlueMatt> you mean onto master?
716 2011-07-14 14:10:59 <Joric> BlueMatt, the wallet http://goo.gl/GS6ZZ
717 2011-07-14 14:11:12 <BlueMatt> Joric: testnet?
718 2011-07-14 14:11:17 <b4epoche> well, whatever the latest (semi-)stable version is
719 2011-07-14 14:11:36 <Joric> no it's not testnet
720 2011-07-14 14:11:43 <Joric> just newly created
721 2011-07-14 14:11:51 <BlueMatt> m whatever
722 2011-07-14 14:11:57 <BlueMatt> how did you make it?
723 2011-07-14 14:12:21 <Joric> imported an unencrypted key entry in bitcointools
724 2011-07-14 14:14:14 <BlueMatt> Joric: nope master for me just closes with a "Load wallet failed..."
725 2011-07-14 14:14:24 <Joric> okay then
726 2011-07-14 14:14:35 <BlueMatt> (as it should)
727 2011-07-14 14:15:20 <BlueMatt> ah, actually, I see...
728 2011-07-14 14:15:30 <BlueMatt> yep wires got crossed and stuff that wasnt rebased was merged
729 2011-07-14 14:15:30 <Joric> oh btw did you cut out that double instance checking on win32? it's pretty much dated
730 2011-07-14 14:16:46 <Joric> BlueMatt, https://github.com/bitcoin/bitcoin/blob/master/src/init.cpp line 308 until #endif, should be removed imo
731 2011-07-14 14:16:49 <BlueMatt> Joric: sipa https://github.com/bitcoin/bitcoin/pull/413
732 2011-07-14 14:17:24 <BlueMatt> Joric: ???
733 2011-07-14 14:18:10 <BlueMatt> Joric: dont think anyone has bothered to look to deep, submit a pull redhatzero q
734 2011-07-14 14:18:12 <BlueMatt> req*
735 2011-07-14 14:18:23 <Joric> BlueMatt, that single instance check is older than the lock file code, the whole block is obsolete
736 2011-07-14 14:18:41 <BlueMatt> does the lock file stuff work on win32?
737 2011-07-14 14:19:14 <Joric> yes i've tried it on the same datadir, it says locked or something like that
738 2011-07-14 14:19:23 <BlueMatt> on winxp?
739 2011-07-14 14:19:41 <Joric> well... on win7
740 2011-07-14 14:21:59 <Joric> it's currently an issue 75 https://github.com/bitcoin/bitcoin/issues/75
741 2011-07-14 14:22:18 <BlueMatt> so submit a pull reqeust
742 2011-07-14 14:22:31 <BlueMatt> and check that it works on xp
743 2011-07-14 14:22:40 <BlueMatt> (obviously googling the relevant locking code works too)
744 2011-07-14 14:23:02 <BlueMatt> afternoon jgarzik
745 2011-07-14 14:26:09 <FellowTraveler> I have realized some of the custom things I want to do will require doing Bitcoin/blockchain calls from within my own C++ code
746 2011-07-14 14:26:14 <FellowTraveler> can the BTC client be used as a lib?
747 2011-07-14 14:26:19 <BlueMatt> no
748 2011-07-14 14:26:20 <FellowTraveler> or is there some preferable lib?
749 2011-07-14 14:26:27 <FellowTraveler> or just copy some code if I need it?
750 2011-07-14 14:27:04 <jgarzik> FellowTraveler: you can always do that :)
751 2011-07-14 14:27:19 <BlueMatt> FellowTraveler: well there is an effort to clean up the code to make it useable as a lib, but not really good to do it yet
752 2011-07-14 14:27:19 <Joric> just modify bitcoind or something
753 2011-07-14 14:27:33 <FellowTraveler> BlueMatt who is in charge of that effort?
754 2011-07-14 14:27:42 <BlueMatt> FellowTraveler: anyone coding on it...
755 2011-07-14 14:27:45 <BlueMatt> which atm I think is just me
756 2011-07-14 14:28:08 <BlueMatt> But I think JS might be doing other stuff soon too
757 2011-07-14 14:28:15 <BlueMatt> (judging by his post on the mailing list)
758 2011-07-14 14:28:54 <FellowTraveler> Basically I need to be able to send and receive bitcoin payments INSIDE the C++.  And the important part is a customized piece of bitcoind that does payments to multiple parties (like escrow, where the parties vote to release funds)
759 2011-07-14 14:29:03 <FellowTraveler> that's why I need it inside my server code
760 2011-07-14 14:29:06 <Joric> though i afraid bitcoind is linked with wxwidgets aswell
761 2011-07-14 14:29:17 <BlueMatt> Joric: no its not
762 2011-07-14 14:29:25 <FellowTraveler> so people can bail their bitcoins in/out of an OT server, but it's really bailed into escrow voting pool, so you don't have to trust the server.
763 2011-07-14 14:29:57 <FellowTraveler> for me to accomplish that, I need to be able to send/receive inside my own code.  Can you give me a stripped down piece of code that does that.
764 2011-07-14 14:30:02 <FellowTraveler> like, the minimum I would need to include?
765 2011-07-14 14:30:08 <BlueMatt> you could do that
766 2011-07-14 14:30:17 <BlueMatt> if you want to copy/paste code which is probably the easiest
767 2011-07-14 14:30:27 <FellowTraveler> well I can paste a function in, if that's all it will take
768 2011-07-14 14:30:42 <b4epoche> why not RPC?
769 2011-07-14 14:30:45 <BlueMatt> no, bitcoin doesnt generate escrow txes
770 2011-07-14 14:30:48 <BlueMatt> so youll have to code it
771 2011-07-14 14:30:57 <FellowTraveler> but if stuff needs to be sectioned off into lib form, or headers need to be fixed up or whatever, that's the part I don't know that much
772 2011-07-14 14:30:59 <BlueMatt> or do escrow anything
773 2011-07-14 14:31:12 <FellowTraveler> don't worry about the escrow part, I just need basic send/receive inside my C++ code
774 2011-07-14 14:31:45 <FellowTraveler> I will do the escrow no prob after that is working.
775 2011-07-14 14:32:04 <BlueMatt> shouldnt be harder than calling a committransaction function or two
776 2011-07-14 14:32:06 <b4epoche> I'm not sure why RPC wouldn't be a good solution for FellowTraveler
777 2011-07-14 14:32:22 <b4epoche> although I have to admit that I really know little about the details of RPC
778 2011-07-14 14:32:48 <FellowTraveler> b4epoche the reason is because once I have the basic send/receive inside my code, I'm going to modify it to do my special escrow / low trust servers / voting pool thing.
779 2011-07-14 14:32:53 <BlueMatt> b4epoche: you cant use rpc to do with w/o modifying rpc
780 2011-07-14 14:33:21 <FellowTraveler> I have a whole protocol figured out and it involves a modified version of BTC.
781 2011-07-14 14:34:19 <b4epoche> oh, I just thought you needed it bitcoind as send/receive 'engine' with you doing the splitting and stuff elsewhere
782 2011-07-14 14:34:42 <denisx> is there anywhere explaind how this feestuff works. it seems non-deterministic
783 2011-07-14 14:35:44 <b4epoche> there's a random number generator in there to determine fees
784 2011-07-14 14:37:29 <BlueMatt> lol
785 2011-07-14 14:37:52 <CIA-103> bitcoin: Matt Corallo master * r3a10d1c / src/db.cpp : Fix bad return values in LoadWallet. - http://bit.ly/qYKNpC https://github.com/bitcoin/bitcoin/commit/3a10d1c49390f26daed14d6362f72c154ac191e7
786 2011-07-14 14:37:53 <CIA-103> bitcoin: Jeff Garzik master * r065d743 / src/db.cpp : Merge pull request #413 from TheBlueMatt/loadwalletret ... https://github.com/bitcoin/bitcoin/commit/065d74301990d6cfce35ca3a6be264d0b975619b
787 2011-07-14 14:39:31 <BlueMatt> denisx: its very deterministic, its on the wiki