Another missing part:

--- a/src/qt/dash.cpp
+++ b/src/qt/dash.cpp
@@ -451,10 +451,6 @@ void BitcoinApplication::addWallet(WalletModel* walletModel)
     if (m_wallet_models.empty()) {
         window->setCurrentWallet(walletModel->getWalletName());
     }
-
-    connect(walletModel, SIGNAL(coinsSent(WalletModel*, SendCoinsRecipient, QByteArray)),
-        paymentServer, SLOT(fetchPaymentACK(WalletModel*, const SendCoinsRecipient&, QByteArray)));
-
     m_wallet_models.push_back(walletModel);
 #endif
 }

same as https://github.com/bitcoin/bitcoin/pull/17165/files#diff-2e949323a5be6fd47f7af05ba778bbd8f0c11d9530b95de9e21690004d9ee40cL342

Nack. Dash uses BIP70 quite a lot in mobile, especially with Anypay. We need to carefully think if we really want to remove this. BIP70 will also be expanded to support DashPay features.

Nack. Dash uses BIP70 quite a lot in mobile, especially with Anypay. We need to carefully think if we really want to remove this. BIP70 will also be expanded to support DashPay features.

Fair enough, but that also means that a) we're completely on our own because Bitcoin will no longer be supporting it, b) Core would remain to some extent permanently tethered to OpenSSL and Protobuf and c) The argument in favor of removing BIP70 is that existing solutions already exist that do not justify carrying the debt associated with maintaining BIP70.

That being said, there is speculation that BIP70's removal was not due to strictly technical reasons alone and I can see why maybe having it disabled but still available in one release before axing it may be a better solution.

@QuantumExplorer

Dash uses BIP70 quite a lot in mobile, especially with Anypay.

This PR doesn't affect mobile/SPV wallets in any way.

BIP70 will also be expanded to support DashPay features.

Core (L1) is not supposed to know anything about DashPay (L2) or its features. I don't think such expansions would be merged here in the future.

This PR doesn't affect mobile/SPV wallets in any way

I meant that Dash users in mobile are known to use BIP70, for payments with Anypay. I am not sure if people might also want to use the Core client for such payments.

Core (L1) is not supposed to know anything about DashPay (L2) or its features.

This one is a little tricky. I am not sold that Layers should be defined as products. In my mind Layer 1 is the blockchain layer. Layer 2 is the state layer, and Layer 3 is the application data layer (state can be encrypted for instance). While L1 should not know anything about L2, Core can certainly have updates that would bring it Layer 3 capabilities and maybe even some state knowledge. This would not be for Masternodes, but for wallets. This would be especially useful for exchanges that use our core wallet, but want an Identity they can use for their customers to send funds to over DashPay.

In the end, the Bitcoin Core wallet is moving away from all user-facing features and is becoming a node implementation. We need to decide if we want to follow that path.

... Core can certainly have updates that would bring it Layer 3 capabilities and maybe even some state knowledge. This would be especially useful for exchanges that use our core wallet, but want an Identity they can use for their customers to send funds to over DashPay.

Hmm.. my understanding is that to talk to/over DashPay you need a Platform-based app (built with a Platform SDK) and that's exactly what various services (exchanges etc.) would have to build/use.

Anyway, these changes are GUI only and exchanges (and other services) do not use GUI (at least I reeeally hope the don't lol), so this PR should not affect them in any way. Moreover, you can't simply run Core to provide BIP70-compatible payments as a service (no such functionality ever existed in Core), you need some another piece of software to do this, so again, no harm is done to BIP70-compatible services which use Core as a backend.

In the end, the Bitcoin Core wallet is moving away from all user-facing features and is becoming a node implementation. We need to decide if we want to follow that path.

In the long run, yes, imo. I don't think there are any other options for mass adoption scenario besides masternodes and other full nodes run by various services being a backbone of the network and the vast majority of regular users transacting via light wallets. Core should become just an interface to the network to which you could connect any app (which talks the same language ofc) with whatever capabilities you need, be it a wallet, an explorer or any custom logic software of your choice.

I'm not sure I'm a fan of removing BIP70 support from Dash Core GUI. Lots of users currently use Dash Core as their primary user application. Once we have split qt from dash core fully like bitcoin has done, then I would fully support removing BIP70 and from that OpenSSL from core itself as long as qt can retain BIP70 functionality.

Won't be merged in the immediate future, let's close this for now

This pull request has conflicts, please rebase.