refactor(send): execute prepared transaction in saga #5001

satish-ravi · 2024-02-28T23:35:09Z

Description

Execute prepared tx directly in the send saga instead of using the viem/saga's send payment method, which was simulating a contract call and then executing it.

Test plan

Manual, CI

Related issues

Fixes ACT-1049

Backwards compatibility

Yes

Network scalability

If a new NetworkId and/or Network are added in the future, the changes in this PR will:

Continue to work without code changes, OR trigger a compilation error (guaranteeing we find it when a new network is added)

satish-ravi · 2024-02-28T23:43:47Z

src/viem/saga.ts

This was using the wrapSendTransactionWithRetry method, which attempts to retry after a timeout of 90s, but the retry count was set to only 1, all of this is removed in the send saga, which now matches with what the swap saga is doing currently.
Also, this was firing events at each stage of the transaction, but the send saga already covers most of it, we only lose out on some intermediate events,transaction_hash_received and transaction_receipt_received, which seem more like debugging events.
I don't think either of these are required anymore, but can add these back if there's a strong reason

Is there harm in including these in the updated saga? Seems like there are clear places where they would fit?

If these are truly never used by anyone though, I think it's fine to get rid of them. Are there any other actual uses of these events in code though? If we remove them here, can we clean them up from the app entirely?

It would be weird just to include those tx events and don't fire other tx events. These aren't fired for swaps, so don't think we need to fire them for sends.
These are still used in the old contract-kit sends, we can clean them up once we remove it.

codecov · 2024-02-28T23:58:58Z

Codecov Report

Attention: Patch coverage is 93.22034% with 4 lines in your changes are missing coverage. Please review.

Project coverage is 85.49%. Comparing base (722017d) to head (b225571).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #5001      +/-   ##
==========================================
- Coverage   85.54%   85.49%   -0.05%     
==========================================
  Files         724      723       -1     
  Lines       29612    29462     -150     
  Branches     5127     5099      -28     
==========================================
- Hits        25331    25188     -143     
+ Misses       4046     4039       -7     
  Partials      235      235

Files	Coverage Δ
src/send/SendConfirmation.tsx	`91.33% <ø> (ø)`
src/send/actions.ts	`100.00% <100.00%> (ø)`
src/send/saga.ts	`68.05% <93.10%> (+9.85%)`	⬆️

... and 2 files with indirect coverage changes

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 722017d...b225571. Read the comment docs.

emerge-tools · 2024-02-29T03:09:44Z

1 build increased size

Name	Version	Download	Change	Install	Change	Approval
⚠️ Celo (test) org.celo.mobile.test	1.79.0 (144)	26.5 MB	⬆️ 2.4 MB (9.81%)	63.0 MB	⬆️ 2.7 MB (4.44%)	N/A

Celo (test) 1.79.0 (144)
org.celo.mobile.test

⚖️ Compare build
📦 Install build
⏱️ Analyze build performance

Total install size change: ⬆️ 2.7 MB (4.44%)
Total download size change: ⬆️ 2.4 MB (9.81%)

Largest size changes

Item	Install Size Change
📝 splashBackground@3x.jpg	⬆️ 600.2 kB
📝 background@3x.jpg	⬆️ 368.6 kB
📝 boost-rewards@3x.png	⬆️ 188.4 kB
📝 background@2x.jpg	⬆️ 176.1 kB
📝 boost-rewards@2x.png	⬆️ 90.1 kB

View Treemap

🛸 Powered by Emerge Tools

jophish · 2024-02-29T17:45:22Z

src/send/saga.ts

-      }
-    } else {
-      throw new Error('No address found on recipient')
+    if (tokenInfo?.symbol === 'CELO') {


nit: Maybe worth having a test for this event?

jophish · 2024-02-29T17:52:26Z

src/viem/saga.ts

Is there harm in including these in the updated saga? Seems like there are clear places where they would fit?

If these are truly never used by anyone though, I think it's fine to get rid of them. Are there any other actual uses of these events in code though? If we remove them here, can we clean them up from the app entirely?

jeanregisser

Great!!

Looking forward to when we can combine the logic for send/swap/wc too, but that will be more work :D

jeanregisser · 2024-02-29T18:19:17Z

src/send/saga.ts

+    )
+
+    if (receipt.status === 'reverted') {
+      throw new Error('transaction reverted')


nit: would be nice to include the txHash in the error message for debugging purposes.

jeanregisser · 2024-02-29T18:24:04Z

src/send/saga.ts

+    yield* put(showErrorOrFallback(err, ErrorMessages.SEND_PAYMENT_FAILED))
    yield* put(sendPaymentFailure())


Looks like we wouldn't want the error message if the pin was cancelled.

And maybe not the sendPaymentFailure()? Unless it's used for switching the button state back to enabled.

The showErrorOrFallback will show a pin required error message, and sendPaymentFailure just resets the isSending state so should be fine to include.
I was confused by this as well originally 😄. left a couple of comments

ok, I guess what I'm suggesting is we change the UX to not display an error message if the pin was cancelled, and not proceed. It's slightly annoying to see the error in that case, since the user decided to cancel.
We did that for swap.

wallet/src/swap/saga.ts

Lines 378 to 383 in fb39d99

} catch (err) {

if (err === CANCELLED_PIN_INPUT) {

Logger.info(TAG, 'Swap cancelled by user')

yield* put(swapCancel(swapId))

return

}

makes sense, I think there are a few differences with how send and swap UX works, like for swap we navigate immediately after submitting the tx to the chain but we block on the confirmation screen for send. Would be good to address all that in a separate PR with all the UX changes.

jeanregisser · 2024-02-29T18:24:38Z

src/send/saga.ts

+    }
+    Logger.error(`${TAG}/sendPaymentSaga`, 'Send payment failed', error)
+    ValoraAnalytics.track(SendEvents.send_tx_error, { error: error.message })
+    SentryTransactionHub.finishTransaction(SentryTransaction.send_payment)


Should the sentry transaction be in a finally block?

This was my original plan, but if the pin is cancelled, we don't start the transaction, so I included it here. But looking at finishTransaction, it looks like it does nothing if a tx with given name isn't found, so should be safe and clean to do it in finally.

### Description Execute prepared tx directly in the send saga instead of using the viem/saga's send payment method, which was simulating a contract call and then executing it. ### Test plan Manual, CI ### Related issues - Fixes ACT-1049 ### Backwards compatibility Yes ### Network scalability If a new NetworkId and/or Network are added in the future, the changes in this PR will: - [X] Continue to work without code changes, OR trigger a compilation error (guaranteeing we find it when a new network is added)

satish-ravi commented Feb 28, 2024

View reviewed changes

Base automatically changed from satish/act-1049-send-confirmation to main February 29, 2024 02:57

satish-ravi added 2 commits February 28, 2024 19:02

refactor(send): execute prepared transaction in saga

7a3bed2

fix build

f45d812

satish-ravi force-pushed the satish/act-1049-saga branch from 025f46c to f45d812 Compare February 29, 2024 03:02

jophish reviewed Feb 29, 2024

View reviewed changes

jeanregisser reviewed Feb 29, 2024

View reviewed changes

satish-ravi added 3 commits February 29, 2024 13:25

Merge branch 'main' into satish/act-1049-saga

8e34092

feedback

f9741a8

move finishTx to finally

1e35805

jeanregisser approved these changes Mar 4, 2024

View reviewed changes

satish-ravi added 2 commits March 4, 2024 12:24

Merge branch 'main' into satish/act-1049-saga

8da5fea

fix test

b225571

satish-ravi enabled auto-merge March 4, 2024 20:57

jophish approved these changes Mar 4, 2024

View reviewed changes

satish-ravi added this pull request to the merge queue Mar 4, 2024

Merged via the queue into main with commit 5b3b1ea Mar 4, 2024
16 checks passed

satish-ravi deleted the satish/act-1049-saga branch March 4, 2024 22:07

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

refactor(send): execute prepared transaction in saga #5001

refactor(send): execute prepared transaction in saga #5001

satish-ravi commented Feb 28, 2024 •

edited

satish-ravi Feb 28, 2024

jophish Feb 29, 2024

satish-ravi Feb 29, 2024

codecov bot commented Feb 28, 2024 •

edited

emerge-tools bot commented Feb 29, 2024 •

edited

jophish Feb 29, 2024

jophish Feb 29, 2024

jeanregisser left a comment

jeanregisser Feb 29, 2024

jeanregisser Feb 29, 2024

satish-ravi Feb 29, 2024

jeanregisser Mar 4, 2024

satish-ravi Mar 4, 2024

jeanregisser Feb 29, 2024

satish-ravi Feb 29, 2024

		yield* put(showErrorOrFallback(err, ErrorMessages.SEND_PAYMENT_FAILED))
		yield* put(sendPaymentFailure())

	} catch (err) {
	if (err === CANCELLED_PIN_INPUT) {
	Logger.info(TAG, 'Swap cancelled by user')
	yield* put(swapCancel(swapId))
	return
	}

refactor(send): execute prepared transaction in saga #5001

refactor(send): execute prepared transaction in saga #5001

Conversation

satish-ravi commented Feb 28, 2024 • edited

Description

Test plan

Related issues

Backwards compatibility

Network scalability

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codecov bot commented Feb 28, 2024 • edited

Codecov Report

emerge-tools bot commented Feb 29, 2024 • edited

1 build increased size

Celo (test) 1.79.0 (144)org.celo.mobile.test

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jeanregisser left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

satish-ravi commented Feb 28, 2024 •

edited

codecov bot commented Feb 28, 2024 •

edited

emerge-tools bot commented Feb 29, 2024 •

edited

Celo (test) 1.79.0 (144)
org.celo.mobile.test