From niftynei at gmail.com  Thu Nov 29 00:21:50 2018
From: niftynei at gmail.com (lisa neigut)
Date: Wed, 28 Nov 2018 18:21:50 -0600
Subject: [Lightning-dev] Dual Funding Proposal
In-Reply-To: <Pv0ITmVb0lqZq-Nt1dVuc7RqhxjL3FGnEI0hYU_Rob4uI00xz1z6p3ed_kRPSWj1EieBhEnhDhLUddWrxrRSzqQjUxlwTEW4w-zZXpGt--k=@protonmail.com>
References: <CAM1a7P1CSFzNVjv2mREOAPE6SUvsdcC_t7n0YWX5fxQ=GGQp5g@mail.gmail.com>
	<Pv0ITmVb0lqZq-Nt1dVuc7RqhxjL3FGnEI0hYU_Rob4uI00xz1z6p3ed_kRPSWj1EieBhEnhDhLUddWrxrRSzqQjUxlwTEW4w-zZXpGt--k=@protonmail.com>
Message-ID: <CAM1a7P0d+dxLaPGPxN4w5be-hpBWFfS1m+TUh-R2VU2nF5GdFw@mail.gmail.com>

On Tue, Nov 27, 2018 at 11:26 PM ZmnSCPxj <ZmnSCPxj at protonmail.com> wrote:

> Good morning Lisa,
>
> Minor comments only, have not studied in detail:
>
>
>
> ____ `accept_channel2`:
>
> [32:temporary_channel_id]
>
> ?  // unchanged
>
> [33:first_per_commitment_point]
>
> [?: options_tlv]
>
> options_tlv:
>
>    1.
>
>    Type: 1 `option_upfront_shutdown_script`
>
>            [2:len]
>
>           Value: `shutdown_scriptpubkey`
>
>
> I believe an even type is more appropriate, since the other side MUST
> enforce that sthudown only goes to the specified script?
>
>
Good point. This definitely should be an even type number as you suggest.
My initial thought was that being inside an `options_tlv` would preclude
the need to maintain the even/oddness designation, but I think this is a
mistake.

____`funding_puts2`
>
> This message exchanges the input and output information necessary to
> compose the funding transaction.
>
> [32:temporary_channel_id]
>
> [`2`:`num_inputs`]
>
> [`num_inputs*input_info`]
>
> [`2`:`num_outputs`]
>
> [`num_outputs`*ouput_info`]
>
> 1. subtype: `input_info`
>
> 2. data:
>
>  * [`8`:`satoshis`]
>
>  * [`32`:`prevtxid`]
>
>  * [`4`:`prevtxoutnum`]
>
>  * [`2`:`scriptlen`]
>
>  * [`scriptlen`:`script`]
>
>  * [`2`:`max_extra_witness_len`]
>
>  * [`2`:`wscriptlen`]
>
>  * [`wscriptlen`:`wscript`]
>
>
> `script` here is the `scriptPubKey`?  This is needed for `hashPrevouts` in
> BIP143 I believe.
>
> What is the `wscript`?  Is this the `scriptCode` in BIP143?
>
> Are non-SegWit inputs disallowed?
>

This borrows heavily from Rusty's splicing proposal; whatever was specified
there should be assumed to also be specified here for inputs.  Rationale
being there should be no difference between the input requirements for a
splice versus a dual fund.


>
>
> 1. subtype: `output_info`
>
> 2. data:
>
>  * [`8`:`satoshis`]
>
>  * [`2`:`scriptlen`]
>
>  * [`scriptlen`:`script`]
>
> Requirements:
>
> The sending node:
>
>    -
>
>    MUST ensure each `input_info` refers to an existing UTXO
>    -
>
>    MUST ensure the `output_info`.`script` is a standard script
>    -
>
>    MUST NOT spend any UTXOs specified in funding_puts2 until/unless the
>    channel establishment has failed
>
>
> If a violation of this is detected, what MUST we do?
>

I'm not sure what you mean here, it's not exactly enforceable other than
from an implementation perspective. If the funding transaction is invalid
because of an input (ie a double spend), the funding transaction won't be
confirmed. This should be discovered when the node attempts to broadcast
the funding transaction, and can be handled (ie treated as a channel
failure) there.


> Regards,
> ZmnSCPxj
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/lightning-dev/attachments/20181128/600e60f7/attachment.html>