3
1
Here is an example of multiple transactions in the same block, all spending from the same output. I created these using the Electrum wallet, telling Electrum to spend a little bit of the BTC from the same 1AD9xV6aeQMPMzfv5Nzk8yNDgtyV3LaUkB address each time
See: https://blockchain.info/address/1AD9xV6aeQMPMzfv5Nzk8yNDgtyV3LaUkB

When I try doing a similar thing myself with Python and the Pycoin library, the second transaction gets flagged as a "double spend" when I submit it to blockchain.info/pushtx. The response I get is:
An outpoint is already spent in [DBBitcoinTx [txIndex=101360418, getEstimatedBTCSent()=0, isGeneratedBlockReward()=false, getTotalBTCSent()=0, getHash()=1766bb1b732db7b3eef1bea27a572651ee343d79ca6c5b9e866e2f97e8df85aa, getVersion()=1, getSize()=258, toString()=BitcoinTx{hash=1766bb1b732db7b3eef1bea27a572651ee343d79ca6c5b9e866e2f97e8df85aa, version=1, lockTime=0, in=[], out=[]}, getClass()=class piuk.blockchain_db.DBBitcoinTx]] [(101355273, 1)]
Like in the first example, the total spent by all the transactions never exceeds the total available at that address, so I'm curious why this is considered a "double spend".
How should you structure a series of transactions spending from the same address so that a double spend is not triggered?


The three transactions you show spend different outputs. – Nick ODell – 2015-09-09T09:20:10.953
Blockchain.info is an authority on nothing, their rules are not Bitcoins rules. – Anonymous – 2015-09-09T09:21:54.097
3same address doesn't mean same output. It's only a doublespend if it spends the same outputs. Addresses can have more than one output. – Murch – 2015-09-09T09:43:34.083