Fix: Improve transaction matching algorithm to prevent suboptimal pairs#3328
Draft
TaprootFreak wants to merge 4 commits intodevelopfrom
Draft
Fix: Improve transaction matching algorithm to prevent suboptimal pairs#3328TaprootFreak wants to merge 4 commits intodevelopfrom
TaprootFreak wants to merge 4 commits intodevelopfrom
Conversation
Problem: The previous greedy matching algorithm could create suboptimal pairs, leading to unmatched transactions that should have been matched based on exact references or temporal proximity. Example case: - bank_tx 191175 (E2E-79792) created 07:31 - exchange_tx 125970 (DEPOSIT-79792) created 07:33 - These belong together (same reference, 2 min apart) - But exchange_tx 125970 was matched with bank_tx 190594 (5 days apart) - Result: 40k CHF stuck as "pending to Scrypt" for 34+ hours Solution: Implement two-phase optimal matching: Phase 1: Exact Reference Matching - Match bank_tx.endToEndId with exchange_tx.txId - Ensures transactions with matching references are paired first - Prevents false matches based purely on timing Phase 2: Optimal Timing-based Matching - Build cost matrix based on temporal proximity - Use greedy algorithm on sorted pairs (best matches first) - Ensures closest temporal matches are selected Changes: - Add matchByExactReference() method for Phase 1 - Add buildCostMatrix() and findOptimalMatches() for Phase 2 - Replace greedy iteration in filterSenderPendingList() with two-phase approach - Add comprehensive test cases for both phases - Add verbose logging for match tracking and debugging Benefits: - Prevents suboptimal matches that cause false "pending" balances - Prioritizes exact reference matches over approximations - Reduces manual intervention needed for stuck transactions - Improves overall matching accuracy and reliability
The new optimal matching algorithm (Phase 1: Exact Reference + Phase 2: Optimal Timing) matches more transactions successfully than the old greedy approach. Tests now expect: - Fewer unmatched transactions - Better pairing based on timing and exact references - All 13 tests passing
Add missing @dfinity dependencies that were previously only transitive dependencies: - @dfinity/agent@^3.4.3 - @dfinity/principal@^3.4.3 - @dfinity/candid@^3.4.3 These are required by ICP blockchain integration files and should be explicitly listed since they are directly imported.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
Das aktuelle greedy Matching führt zu suboptimalen Transaktions-Paarungen, wodurch Transaktionen fälschlicherweise als "pending" markiert werden.
Konkretes Beispiel:
Lösung
Implementierung eines zwei-Phasen optimalen Matching-Algorithmus:
Phase 1: Exakte Referenz-Matches
bank_tx.endToEndIdmitexchange_tx.txIdPhase 2: Optimales Timing-basiertes Matching
Änderungen
Code
matchByExactReference()für Phase 1buildCostMatrix()undfindOptimalMatches()für Phase 2filterSenderPendingList()nutzt jetzt zwei-Phasen AnsatzTests
Auswirkungen
Positiv
Risiken
Testing
Lokal
Empfohlen vor Merge
Verwandte Issues
Löst das Problem der 40k CHF "toScrypt pending" seit 01.03.2026
Checkliste