|
Ah mon péché mignon, oublier de repasser d'une chaîne de caractères hexa à la valeur hexa...
Du coup j'ai le bon hash
Mais la vérification ne marche pas, bizarre
De toute façon si chez toi ça marche c'est que c'est bon, c'est juste ma fonction de vérification qui déconne
Dernière question, pourquoi tu as enlevé l'octet final de la signature? Le 01?
|
|
|
|
|
C'est bizarre
La fonction Hash c'est bien sha256^2 non? Je viens d'essayer sha256 quand même mais ça ne marche pas non plus
|
|
|
|
Je m'étais trompé dans le montant, après correction j'ai le même message que toi Mais la vérification fonctionne chez toi? Et tu as le même hash que moi? Mon vérifieur me dit que 3044022044d64dd439555d659ec6d3a530a2eb2a5747e8f29522aa87636acc33bc1a54d40220720f8f8d776b4aba6916892bd5994d0a71436d738cd106907ecec57182ee8eb101 n'est pas une signature de Hash("0100000001177a1f8fdfe69a2ad022f056bf45ef8162c7c8a1672919b885308b3b16162307000000001976a9145f10ada9157add69864482ef8c99ef8787943f0488acffffffff0130c11d00000000001976a9145c3f294acb5059a968c4a35700b5432a7397002b88ac0000000001000000")
=
14e230ca4b3d2963630bb186dee8146fcf401d0de03eccc9926cf33944ce216cpar l'adresse 19ff... |
|
|
|
|
Je viens de comparer les deux scripts, je ne vois pas la différence en dehors du fait que le wiki utilise comme scriptPubkey "Pubkey OP_CHECKSIG" et la transaction que j'utilise "OP_DUP OP_HASH160 <hash> OP_VERIFY OP_CHECKSIG"
|
|
|
|
|
Ah je croyais que c'était une transaction ce code-là...
Je vais pouvoir creuser de ce côté-là
Merci pour tes infos et ta propostion, effectivement je veux bien ton code ça m'aiderait je pense
|
|
|
|
I'm happy to see some of you guys agree with destroying Facebook! I mean after all who controls the internet ? how are we sure its not the governments ? And after watching the social network why would anyone trust zuckerburg ( bloodsucker) with ur personal info? Oh the tragedy come November 5!
So, is it just going to be some bad-ass DDoS attack? I remember when The Social Network (2010) came out; there was a small DDoS for a few days. I'm thinking some anons will take advantage of the scurry of admins trying to get the DDoS to stop and while they are doing that they will probubly find a hole durning this panic  That would be nice |
|
|
|
Lol, I hope you know that Bonker is putting these ridiculous threads out as jokes. You shoulda seen his idea to forever leave a fridge with the door open next to your GPU in another thread.
Epic! |
|
|
|
Bonjour Je pense être à deux doigts de résoudre mon problème mais j'y suis pas encore et comme il est assez pointu (en tout cas pour moi) je comprendrais peut-être mieux une explication venant d'un francophone Ce que je veux c'est signer une transaction Je sais signer un message, je connais la syntaxe d'une transaction, ce qu'il me manque c'est le message à signer Je sais que c'est un Hash, donc un sha256^2 Je sais que l'argument de ce Hash est la transaction modifiée, en fonction de SIGHASH Voici la partie non sûre : Si j'utilise SIGHASH_ALL, le message est la transaction où : tous les scriptSig(non encore connu encore puisque justement je les veux) sont supprimés à l'emplacement du scriptSig à la position n, je mets le scriptPubKey de l'input correspondant (une fois signé, j'obtiendrais alors le scriptSig pour l'input n) j'ajoute à la fin 01000000 car j'utilise SIGHASH_ALL De plus, suite à une réponse de Mike sur le forum dev, j'ai lu la fonction SignatureHash qui initialise le message comme ceci : "CDataStream ss(SER_GETHASH);" Sachant que SER_GETHASH = 4, il faudrait donc que je rajoute au début de ma transaction modifiée 0400000000000000, soit 4 en int64 Donc finalement, j'arrive à ça (j'ai pris une vraie tx pour pouvoir comparer : http://blockexplorer.com/rawtx/a467d767729385c191d3e463ceb8a436a1c0dda9d362955909c64196db86e573 ) : 04000000 or 0400000000000000 or nothing
01000000 #version
01 #1 input
177a1f8fdfe69a2ad022f056bf45ef8162c7c8a1672919b885308b3b16162307 #inverted hash of the tx
00000000 #index
#old scriptpubkey instead of scriptsig
__________________________________________________________
19 #length of scriptPubKey
76 #OP_DUP
a9 #OP_HASH160
14 #length of hash
5f10ada9157add69864482ef8c99ef8787943f04 # NOT inverted hash160 of my pubkey
88 #OP_VERIFY
ac #OP_CHECKSIG
__________________________________________________________
ffffffff #sequence
01 #1 output
80841e0000000000 #value
19 #length of scriptPubKey
76 #OP_DUP
a9 #OP_HASH160
14 #length of hash
5c3f294acb5059a968c4a35700b5432a7397002b # NOT inverted hash160 of pubkey to which I send money
88 #OP_VERIFY
ac #OP_CHECKSIG
00000000 #locktime
01000000 #hashtype = SIGHASH_ALL
Je sais qu'il y a des développeurs dans le coin, si vous êtes occupés postez même un mini message si ça prendrait trop de temps, j'approfondirai de mon côté |
|
|
|
Ah merci  J'avais vu 2-3 modifications à faire, je les ferai dès que j'ai un peu de temps J'aurai bien aidé plus mais je pense pas que mon niveau en anglais me le permette... |
|
|
|
|
I don't think I can be much help for writing in the wiki because my english is too bad for that, but I can help fighting spam if there are not enough volunteers
|
|
|
|
Hi, If you want to have a deterministic wallet but are afraid of forgetting your passphrase, you can now use one of your file to create your wallet Moreover, a thief will have to check every file on your computer to find your generated keys Download Fillet and run "./fillet.py --file ~/me.png --keynumber 1 --size 10 --pwpath ~/bitcoin/pywallet/" All flags are optional except --file Pywallet is used to convert your key to a Bitcoin address When you will want to import a key, pywallet will do that easily |
|
|
|
Thanks, it still doesn't work, but that helped me I chose SIGHASH_ALL and nvin=0, so SignatureHash becomes: CTransaction txTmp(txTo);
scriptCode.FindAndDelete(CScript(OP_CODESEPARATOR)); //no codeseparator, scriptPubkey was just OPDUP UPHASH HASH VERIFY CHECKSIG
// Blank out other inputs' signatures
for (int i = 0; i < txTmp.vin.size(); i++)
txTmp.vin.scriptSig = CScript();
txTmp.vin[0].scriptSig = scriptCode;
// Serialize and hash
CDataStream ss(SER_GETHASH);
ss.reserve(10000);
ss << txTmp << nHashType;
return Hash(ss.begin(), ss.end()); The only thing I'm not sure is the blue line I think it initializes ss with "04 00 00 00 00 00 00 00" ( 1 << 2 as an int64) I tested "04 00 00 00" and "" also But none works Also, is that ss? (I took a transaction with only 1 input and 1 output this time: http://blockexplorer.com/rawtx/a467d767729385c191d3e463ceb8a436a1c0dda9d362955909c64196db86e573 ) 04000000 or 0400000000000000 or nothing
01000000 #version
01 #1 input
177a1f8fdfe69a2ad022f056bf45ef8162c7c8a1672919b885308b3b16162307 #inverted hash of the tx
00000000 #index
#old scriptpubkey instead of scriptsig
__________________________________________________________
19 #length of scriptPubKey
76 #OP_DUP
a9 #OP_HASH160
14 #length of hash
5f10ada9157add69864482ef8c99ef8787943f04 # NOT inverted hash160 of my pubkey
88 #OP_VERIFY
ac #OP_CHECKSIG
__________________________________________________________
ffffffff #sequence
01 #1 output
80841e0000000000 #value
19 #length of scriptPubKey
76 #OP_DUP
a9 #OP_HASH160
14 #length of hash
5c3f294acb5059a968c4a35700b5432a7397002b # NOT inverted hash160 of pubkey to which I send money
88 #OP_VERIFY
ac #OP_CHECKSIG
00000000 #locktime
01000000 #hashtype = SIGHASH_ALL
|
|
|
|
|
Elle n'est pas publiquement modifiable?
|
|
|
|
Last question: I know how to sign something with a privkey, but what is the message signed in <sig>? I found that on the wiki: The other component is an ECDSA signature over a hash of a simplified version of the transaction. It, combined with the public key, proves the transaction was created by the real owner of the address in question. Various flags define how the transaction is simplified and can be used to create different types of payment. which doesn't really help me, I still don't know what "simplified version" it is I found the table here too: https://en.bitcoin.it/wiki/Script#ScriptsI understand it but OP_CHECKSIG verifies <sig> is from <pubKey>, I don't see the message signed I couldn't find help in the OP_CHECKSIG page neither: https://en.bitcoin.it/wiki/OP_CHECKSIGI also tried with this etotheipi's image: http://dl.dropbox.com/u/1139081/BitcoinImg/OpCheckSigDiagram.pngStill doesn't work, I tried with Hash256 as sha256 and sha256^2 but none worked |
|
|
|
Thanks for the graph, etotheipi I love the bitcoin address graph too |
|
|
|
iv got the new version running 1.1-23
what does each field mean?
key=bitcoin private key?
message=duh
signature=?
public=bitcoin public key?
version=do i need to change that, or is that there for future reasons
Yes sorry I should have said that before Three cases: 1. You just want some info, then just fill Key as usual 2. You want to sign, then fill Key and Msg to sign Msg with Key 3. You want to verify, then fill Msg, Sig and Pubkey to verify that Sig is a signature of Msg by Pubkey |
|
|
|
|
What a dumb!
I was signing my test string instead of the msg form
Try your signatures as if the message were "coucou", it will work...
New release pushed, it will surely work better
|
|
|
|
Still nothing. I tried it with a different key. By the way, if you use --web, it ignores --wallet= on the command line. It'd be nice if that carried over. I went used Dump Your Wallet and copied out one of the "sec" values, scroll down to the info/sign/verify section, paste in the key, Get Info. Results: Address (Bitcoin): 17LeVE2fjEK7ncqxCToU4McytDHLUhKpyx
Privkey (Bitcoin): [secret]
Hexprivkey: [secret]
Pubkey: 04321a3c84e2e380215518fc952fbd782f20ca8940366f7575953c136dfb585b33a93e9969f546d13fd2afbf407dfd44a745b1421d027ee4f16de31a0d7a3b7edb
Type "maybe different words?" into the message box, Get Info. Result: Signature of 'maybe different words?' by 17LeVE2fjEK7ncqxCToU4McytDHLUhKpyx: 304402206e1704baf1c1238077b420b74c07dcdf6964155614005fe20089ce83445e2895022009ebcd27124a52ecdc90a4398438c5c400cc9dcd259560ec90d752a876946178
Pubkey: 04321a3c84e2e380215518fc952fbd782f20ca8940366f7575953c136dfb585b33a93e9969f546d13fd2afbf407dfd44a745b1421d027ee4f16de31a0d7a3b7edb
Go back up, clear out the Key: field. Paste the sig and public key into the correct boxes, Get Info. Result: Signature of 'maybe different words?' by 17LeVE2fjEK7ncqxCToU4McytDHLUhKpyx is NOT 304402206e1704baf1c1238077b420b74c07dcdf6964155614005fe20089ce83445e2895022009ebcd27124a52ecdc90a4398438c5c400cc9dcd259560ec90d752a876946178
I changed the message to "test" just to make sure the question mark wasn messing up the URL encoding, and it still failed verification. Have you tested this one? msg: "Hi github users :p" "pubkey" : "04ef6e366cd6b0b8fbf02c0c25ad39fe892b90c597875899fdc9db16941cf43fb8c429e0534cb8b 972f5cc9f1a50f36dc3352ffad427f073e1c64a145828a3be6e", "sign" : "3046022100a80b6e0c7c54c54ba943f4e3cde12f5a7dc5313e3f0a15ce868f01683ced64fa02210 08b4ad7d3800a11c241dcef7aaf44c8224a7d9f1e54d3e669bf4887036b6d10af" I hope it's ok because here it works Good idea for the wallet flag, will do it as soon as I resolve that |
|
|
|
so if i want to use the new version of pywallet, do i need to run it without ecdsa, or is it just to test to make sure it works without crashing?
I just pushed the fix Now if you run it, it should write that ecdsa is not present, and run normally except that you won't be able to sign/verify |
|
|
|
|
Show Posts
