Mark RBFed txs as removed from mempool

[ben-kaufman]

This PR marks RBFed transactions as removed from mempool.

Testing: Make an onchain transaction, then go to its activity details and boost it.

[ben-kaufman]

Need to update the e2e test to match the behaviour of not deleting the boosted activities.

[ovitrif]

ACK ✅

Test: Send onchain → Boost 🟢

For e2e test to pass pass, please check with @piotr-iohk tomorrow 🙏🏻

[piotr-iohk]

ACK ✅

Test: Send onchain → Boost 🟢

For e2e test to pass pass, please check with @piotr-iohk tomorrow 🙏🏻

Yes, it's the @boost_2 - Can do RBF test failing as it's not expecting the additional tx which is "Removed from mempool".

Question: Where is this change coming from? I haven't found it reflected in the Figma design, and it seems inconsistent with the current behavior in the Bitkit RN app, where the original transaction is not shown once it is RBF-ed - the wallet only displays the replacement transaction.

I’m also a bit concerned that "Removed from mempool" may feel alarming or unclear for non-technical users. If we want to expose this state, we might want to consider alternative wording or a different UX approach.

[ovitrif]

Yes, it's the @boost_2 - Can do RBF test failing as it's not expecting the additional tx which is "Removed from mempool".

Question: Where is this change coming from? I haven't found it reflected in the Figma design, and it seems inconsistent with the current behavior in the Bitkit RN app, where the original transaction is not shown once it is RBF-ed - the wallet only displays the replacement transaction.

I’m also a bit concerned that "Removed from mempool" may feel alarming or unclear for non-technical users. If we want to expose this state, we might want to consider alternative wording or a different UX approach.

@piotr-iohk We already have this in Bitkit RN, it was never in Figma.

It's one of those things which are so technical that devs must force it on UX even if product people would not like it.

While your concerns are valid, if we don't do this, then with the current capabilities we have in the apps, we end up with much worse UX, where the users see flashes and duplicates of a RBFed TX.

My suggestion unless you have a very strong opinion against it, in which case I am open to re-evaluate:
Please adapt the test so it accounts for this 🙏🏻 , Android implementation is also in pending PR:

• Save boosted txs bitkit-android#470

[piotr-iohk]

@piotr-iohk We already have this in Bitkit RN, it was never in Figma.

@ovitrif I don't think it is displayed like that in Bitkit RN (i.e. straight on the activity list as a separate tx). From what I see only the RBF-ed tx id is displayed in the "explore" view of boosting tx.

Anyway, to rephrase what I wanted to say before, I think that is fine to show the tx that was RBF-ed in activity list if we want to expose it. I'm just not fond of "Removed from mempool" label. I think that something else would work better, like: "Superseded"/"Replaced by higher-fee transaction"/etc. 🤷‍♂️

[ben-kaufman]

@piotr-iohk the reason for the label is that eventually we'd want also the other cases of dropped transactions to have that same label.

But we could rename it for this case if that's what we decide.

[ovitrif]

@piotr-iohk

Indeed, for the case when a tx is removed due to RBF (non-technically said: replaced), the current copy (wording) is not the most optimal, I agree.

Though as @ben-kaufman pointed out, the reason we went with this is because "Transaction Removed from Mempool" is also for another case, when a tx might get dropped because it never gets confirmed, or due to a reorg, or if the tx is not propagated enough. (sharing this from DM chats with Ben).

We can use a copy along the lines that you're suggesting:

"Replaced by higher fee transaction"

I would also like to invite @aldertnl for review & input of this text 🙏🏻 .

[ovitrif]

@aldertnl

The discussion is about this activity item variant in the list(s):

1. This PR adds it for transactions that are removed when doing RBF.
2. We (Ben and I) thought to use the standard text that we also have in RN (React Native) Bitkit.
3. Piotr pointed out that this variant is not in Figma, and that it might not be in RN Bitkit.
   • I know I saw it randomly, but I will have to do some tests.
4. Piotr suggested to use more specific copy, so we don't overload users with tech jargon.
5. Ben points out that there are other cases when this copy is more proper, and that we just reused it for this case.
6. We all agree we could find better words, and handle the other "removed from mempool" cases with this copy.
7. We've thought of "Replaced by higher fee transaction", but want your review/input on how to frame this better with words, in the activity list item subtitle.

[ovitrif]

@piotr-iohk OK to proceed with the current text and update to use improved copy in apps + e2e when received?!

[piotr-iohk]

@ben-kaufman @ovitrif tests updated to reflect current state: synonymdev/bitkit-e2e-tests#50
If a new transaction status (including StatusRemoved) is introduced later, the tests will be updated accordingly.

Additional notes for Android implementation: synonymdev/bitkit-android#470 (comment)

One request: if possible, please merge the Android and iOS changes around the same time. It makes it much easier to manage and update the e2e tests.

Looking ahead, it would be very helpful if future changes adding the same feature could use the same branch name in both Android and iOS repos. Our e2e CI first runs tests against a branch with a matching name in the bitkit-e2e-tests, so aligned naming improves the workflow and reduces additional work.

[ovitrif]

utAck

[ovitrif]

One request: if possible, please merge the Android and iOS changes around the same time. It makes it much easier to manage and update the e2e tests.

ack. Will do 👍🏻