Currently, a large amount of heterogeneous event data with opaque event names (e.g., obscured IDs) are increasingly generated, e.g., the blockchain technology requires the transaction information to be recorded by different stakeholders in various ways from different blocks (i.e., distributed ledgers). Thus, a large amount of heterogeneous event data with opaque event names (e.g., obscured IDs) is generated on blockchains. To identify similar workflows on blockchains for collaborator selection, matching heterogeneous events is an important task. Existing techniques are inappropriate because they do not make full use of workflow features in event logs. We observe that several types of event constraints in event logs may serve as more discriminative features in event matching. Specifically, a general matching approach is proposed for identifying a right mapping that maximizes the similarity of event constraints for a pair of heterogeneous event logs. An advanced A* algorithm with a tight upper bound function is devised to improve the matching efficiency of the proposed approach. We conducted exr5tensive experiments to evaluate the effectiveness of our approach, and the experimental results demonstrate that our approach outperforms state-of-the-art matching approaches.