Bitcoin-Qt next-test 2013-07-21

[Luke-Jr]

next-test is a branch of the master bitcoind & Bitcoin-Qt code with as many pull requests merged as possible.
These pull requests are often not well-tested, and may contain serious bugs.

Please note these might possibly corrupt your wallet, destroy your coins (on the network), or worse. No warranty of any kind of provided. BACKUP YOUR WALLET

• Windows: installer ‖ zip (sigs)
• Source code: git ‖ tar.gz ‖ zip

Today's next-test includes the following pull requests (green are merged now; red are disputed):

• Parse URIs with non-BTC amounts
• -acceptnonstdtxn option to skip "non-standard transaction" checks
• Support for Tonal Bitcoin units (ᵇTBC, ˢTBC, and TBC)
• JSON-RPC method: prioritisetransaction <txid> <priority delta>
• A set of build scripts on top of gitian and vagrant for automating gitian builds...
• Change window titles to "Bitcoin-Qt - <purpose>" / misc related renames
• Specify fees explicitly via JSON-RPC
• CreateNewBlock: Child-pays-for-parent / Add transaction fee later
• Add small-data OP_DROP transactions as standard transactions.
• FreeBSD and OpenBSD build fixes
• Support for specific BIP 22 rejection reasons and BIP 23 block proposal
• Base58 en-/de-coding
• Fixes for when listtransactions encounters non-standard outputs
• let user select wallet file with -wallet=foo.dat
• Add wget(1)-friendly crypted key dump via GET /wallet-ckeys.json
• RPC: Avoid cleartext passwords by default
• Restore FIRST_CLASS_MESSAGING
• sipa's optimized secp256k1 library
• P2P: do not return addresses older than 7 days in "addr" messages
• break after ProcessMessage() to give other peers a chance.
• Update for the Windows Installer
• wallet encapsulation
• Only create signatures with even S, and verification mode to check.
• Support for building against system LevelDB
• Filter listaddressgroupings
• Discourage fee sniping with nLockTime
• Fix off-by-one errors in the interpretation of IsFinal()
• Yet another Coin Control Release
• Added -txnotify to call a script when a transaction is received
• Allow createrawtransaction to accept duplicate addresses
• Add user interface to set dust limit and filtered addresses
• Fix LevelDB removing /dev/null at build time
• Added CWalletManager class to support loading of multiple wallets.
• Network activity toggle
• 0mq - Zero MQ Support.
• Adding blockhash parameter to getrawtransaction - Issue #2077
• Enable tx replacement on testnet.
• Payment Protocol Work
• Bugfix: Store last/next wallet resend times unique per CWallet object
• extend core proxy options and handling
• ListTransactionsOfBlock RPC method
• Resolve path to allow for Bitcoin directory to be a symlink
• Watch for double spends of wallet transactions
• Bitcoin-Qt: updates to options handling code
• Partial solaris support
• always show wallet encryption commands in `help`
• Inline signature serializer
• display txfee in first sendCoinsDialog message box (issue #1715)
• Detect any sufficiently long fork and alert the user just like any other alert
• Leveldb subtree at 1 12 0
• Relay OP_RETURN data TxOut as standard transaction type.
• UTF-8 support for JSON-RPC
• Free tx area should be configurable for fee calculation, also
• Cleanup code by using forward declarations and other methods...
• RPC: keypoolrefill now permits optional size parameter, to bump keypool
• Reduce keepalive ping/timeout
• Prune provably-unspendable outputs
• Log aborted block database rebuilds
• Add an address-based transaction index
• Minor cleanup of connection and confirmation icons. Additionally removes unused v0.3.0. image resources.
• Child-pays-for-parent mempool
• Add michagogo to gitian download scripts
• Add forcerescan
• fix invalid conversion error with MinGW 4.8.1 in net.cpp
• Bitcoin-Qt: save and restore position of debug window
• Use shared lock for wallet registration functions.
• BIP 32 derivation + test vectors
• P2SH related RPC improvements
• Bugfix: Move boost/version include from db.cpp to walletdb.cpp where BOOST_VERSION is used
• getblocktemplate: longpolling support
• makefile.mingw: Use normal boost libraries instead of debug for Windows
• update SelectParamsFromCommandLine() handling/order
• Add block blacklisting RPC
• Allow SendMessages to run partially without cs_main
• Bugfix: Include block creation and SSL options headings in usage

[hazek]

I set the topic as sticky for a while because I think that if anything needs extensive testing it's the software that handles our money.

[BCB]

Testing.

[BCB]

in bitcoin-qt in console

typed "walletpassphrase xxxxxx 1"

error:



selected "retry"



Window 7 Professional
64 Bit

[Gavin Andresen]

I set the topic as sticky for a while because I think that if anything needs extensive testing it's the software that handles our money.

I would much rather people help test individual pull requests or 'git HEAD' rather than "random collection of changes". Putting eleven unrelated changes together makes figuring out what caused a bug/regression much harder.

[hazek]

I will set other threads asking people to test as sticky as well. If there is already such a thread please pm it to me.

[Luke-Jr]

I don't think this needs to be stickied; thanks though!
As far as I am aware, no useful test results have ever come from next-test itself, just from my process of putting it together.
If anyone has found bugs for specific pullrequests, or even reported success for any pullrequests using next-test, please feel free to correct me on this.

It's handy as a risky client (I use it myself), and if people want to take advantage of it for that or for testing, they're welcome to.
However, since no other clients have stickies, I don't see any reason next-test should be special in this regard.

I agree it would be nice to have a clear forum thread describing the basic process of testing pullrequests individually.

in bitcoin-qt in console

typed "walletpassphrase xxxxxx 1"

error:

This is an example of an (unfortunately) not useful bug report
There are probably a dozen pullrequests this could have been triggered by, and I cannot reproduce it myself (it works fine for me).

BCB, if you want to help debug this further, I suggest you go through the list of individual pull requests in the opening post one at a time (they're links)
For each one, do:
1. Look for the most recent post by BitcoinPullTester saying "Automatic sanity-testing: PASSED, see <link>"; if there is one, click on that first link.
2. Browse to the bitcoin/ subdirectory, and download bitcoin-qt.exe
3. Try to reproduce the problem using that bitcoin-qt.exe
4. If it has the bug, go back to the main pullrequest page and post a comment there.

[Luke-Jr]

Note that this will cause older versions (including 0.8.3 and git master!) to think they need to reindex the block database!
This is due to unspent (and unspendable) outputs being pruned (which older code thinks is a problem because they don't understand it's unspendable).
p2pool uses these every block it finds.

[inceptor]

Note that this will cause older versions (including 0.8.3 and git master!) to think they need to reindex the block database!
This is due to unspent (and unspendable) outputs being pruned (which older code thinks is a problem because they don't understand it's unspendable).
p2pool uses these every block it finds.

yes agreed
used before older version same happened

better use online wallet like block chain etc?

[phelix]

+1 for CoinControl

[Blockhead]

This is causing me problems, since I am trying to get my coins out of my wallet, but it doesn't work right now.  It started to re-index, but hangs at about 12 days ago.  I am using v0.8.1

[kuzetsa]

+1 for CoinControl

Indeed.

Thanks cozz 

[Machinery]

I'm doing some testing. Thanks to the programmers who are doing the hard work.

[faiza1990]

+1 for CoinControl

Indeed.

Thanks cozz 

agree

[pythonista]

Hi Luke,

Thanks for your hard work.

Would it be possible for the importaddress function for watch-only wallets to be merged in? This is something that merchants & services desperately need - to be free from thirdparty sites like blockchain.info

Here it is:
https://github.com/bitcoin/bitcoin/pull/2861

[Luke-Jr]

Would it be possible for the importaddress function for watch-only wallets to be merged in? This is something that merchants & services desperately need - to be free from thirdparty sites like blockchain.info

Here it is:
https://github.com/bitcoin/bitcoin/pull/2861

That's just watch-only addresses, and not necessarily useful when you need watch-only wallets (ie, every real-world use case). It missed the merge window for this next-test (it was outdated code and unmergable when I made this), but perhaps it will make it into the next one. Please also remember next-test is not necessarily a good idea to run in production.

You can easily get a watch-only wallet today by using keypoolrefill with an absurdly high number (5 times your expected transaction volume is good) on your spending client, making a backup, and using wallet encryption on that backup with a crazy-long password (you can/should throw away) to destroy the private keys. Note that this will break after <keypoolrefill number> transactions, and is probably pretty dangerous to try for non-experts, but if done right can get you a watch-only address that works until Bitcoin-Qt has real support for them.

sipa is working on HD wallet support for Bitcoin-Qt (a prerequisite to real watch-only wallets), so I expect we should see that pop up sometime before 0.10 is released.

[phelix]

+1 for CoinControl

Indeed.

Thanks cozz 

agree

My karma was meant for Luke because he keeps including CC. 

[bernard75]

sipa is working on HD wallet support for Bitcoin-Qt (a prerequisite to real watch-only wallets), so I expect we should see that pop up sometime before 0.10 is released.

could you link to it or him?

[WSDN]

Thanks Luke jr for this exellent job! we are testing right now!