Qeasy Cloud
Get Started

Supplier Master Data Sync in Practice: Code Mapping and Incremental Scheduling from Kingdee Cloud Cosmic to Jushuitan

· 何金辉· Integration Solutions· 18 views· 5 min read
JushuitanKingdee Cloud供应商主数据基础资料同步增量调度轻易云

What This Strategy Solves

Synchronizing supplier master data from an ERP to a front-end business system looks like a simple "data push," but in reality it is the foundation for subsequent purchase orders, receiving, and reconciliation. If code mapping, status determination, and the incremental starting point are not designed correctly from the start, archive inconsistencies will appear 3-6 months later: buyers cannot find the right supplier in the front-end system, and payment reconciliation becomes difficult. At one retail customer's site, we saw that in the fourth month after go-live, finance found over 200 supplier records in the front-end system that had already been disabled in the ERP, blocking reconciliation. The core value of this strategy is to treat supplier master data as a prerequisite dependency for later transactional synchronization—get "archive consistency" right first, then talk about orders and inventory.

Data Flow and Field Mapping

The overall flow is unidirectional: Kingdee Cloud Cosmic (source) → Qeasy data integration platform (middleware, responsible for cleaning, mapping, and transformation) → Jushuitan (target). The middleware does not persist data; it only performs stream processing.

Key field mapping (typical mapping; actual project results depend on business confirmation):

Business meaningKingdee Cloud Cosmic source fieldMiddleware handlingJushuitan target field
Supplier codeFSupplierId / NumberPass-through as unique keysupplier_code
Supplier nameFNameTrim spaces, full/half-width conversionsupplier_name
Mnemonic / short nameFShortNameCross-check with nameshort_name
ContactFContactPass-throughcontact
PhoneFPhoneNormalize separatorsphone
AddressFAddressMerge province/city/districtaddress
Enabled statusFUseOrgId/ForbidStatusMap to enabled/disabled enumstatus
Category / groupFGroupTransform via code mapping tablesupplier_category

We strongly recommend centralizing code mapping in Qeasy rather than scattering it across strategies. A common pattern at customer sites is to maintain all basic data mappings (item, supplier, customer, warehouse) in a single maintenance view, so that adding a new code only requires one edit.

How to Configure on Qeasy

In the Qeasy data integration platform, this strategy is typically configured with the following key points:

  1. Source connection: Select Kingdee Cloud Cosmic, fill in the public cloud tenant access endpoint and credentials, and use a query-type API (such as supplier list query). Enable pagination to avoid oversized single calls.
  2. Target connection: Select Jushuitan and use its supplier master data write API. Jushuitan is sensitive to code uniqueness; ensure the target-side code is globally unique.
  3. Field mapping: In Qeasy's field mapping canvas, map the fields from the table above one by one. Apply cleansing operators on name fields (trim, full/half-width) and enum conversion on status fields.
  4. Centralized code mapping management: Place cross-system inconsistent codes such as categories and groups in the platform's "code mapping" maintenance module rather than hard-coding them in strategy scripts.
  5. Deduplication and idempotency: Enable dedup logic with "supplier code" as the unique key. If it exists, take the update branch; if not, take the insert branch to avoid duplicate creation.
  6. Null and exception handling: For exceptions such as empty names or invalid phone formats, Qeasy supports two handling modes: "skip and log" or "abort strategy." We recommend the former to prevent a single dirty record from blocking the entire batch.

Implementation Steps

Supplier master data synchronization is best rolled out in three phases—the safe approach is "full baseline first, then switch to incremental":

  • Phase 1: Full baseline (before go-live) Manually trigger a full sync on Qeasy to write all currently enabled suppliers from Kingdee Cloud Cosmic to Jushuitan at once. The purpose of this step is not daily operation but to give the target system a "baseline archive." After the full sync, verify counts on both sides and spot-check 20 key field records.

  • Phase 2: Incremental starting point (cutover day) The incremental starting point field is usually "last modified time" in Kingdee Cloud Cosmic. We recommend suspending batch maintenance of supplier records in the source system the day before cutover, taking a clean cutoff timestamp as the incremental starting point to prevent historical dirty data from mixing into the incremental flow.

  • Phase 3: Scheduling frequency (steady-state operation) Supplier master data does not change frequently; an incremental sync every 15-30 minutes is usually sufficient. Qeasy's scheduler supports cron expressions. Common patterns at customer sites are */15 * * * * (every 15 minutes) or hourly at the top of the hour. At night, the frequency can be reduced to hourly and restored during business hours.

Another point that is easy to overlook: split header and body into separate phases. Suppliers are basic data with only a "header" and no body; but if the same integration scenario includes item sync with multiple units of measure, be sure to roll out in two phases—header first, then body. Otherwise, missing units of measure will prevent purchase orders in the front-end from being saved.

Lessons Learned

  1. Reverse synchronization of status fields: A typical mistake is treating "unapproved" in Kingdee Cloud Cosmic as "enabled" when syncing to Jushuitan. The result is that unapproved suppliers can still receive purchase orders. The safe approach is to map clearly to sync only records that are "approved and not disabled."
  2. Code modification causes inconsistency between sides: ERP allows supplier code modification, but once a supplier code in Jushuitan is referenced by a purchase order, it cannot be changed. We recommend treating the code as an "immutable primary key" in Qeasy; code changes should follow a "create new + disable old" flow rather than a direct update.
  3. Incremental starting point polluted by historical data backfill: An ops engineer performs a batch maintenance in the source system, causing the last modified time to roll back. The incremental strategy will miss real additions. The safe approach is to maintain an "incremental starting point blacklist field" list in Qeasy. When such a field is backfilled, raise an alert and let a human re-run the missing data after manual confirmation.
  4. Jushuitan code uniqueness conflict: Jushuitan's supplier_code is globally unique; it cannot duplicate across stores or books. We recommend prepending a source identifier (such as JD_xxx) to the code on the Qeasy side, preserving traceability while avoiding conflicts.
  5. Skip exception vs. abort strategy: During initial debugging, we recommend "abort" to quickly surface dirty data; after steady-state go-live, switch to "skip and log." Otherwise, a single supplier with an empty name will cause the entire batch to fail, impacting downstream purchase orders.

Applicable and Non-Applicable Scenarios

Applicable: Supplier master data is managed centrally in an ERP (similar to Kingdee Cloud Cosmic), the front-end business system (similar to Jushuitan) only reads it, and archive consistency is critical—this fits retail, e-commerce, wholesale, and distribution enterprises. Not applicable: Scenarios where supplier records are independently maintained in the front-end system, bidirectional sync is required, or the business requires managing supplier records separately per store or subsidiary.

Original content. Please credit the source when reposting: https://www.qeasy.cloud/insights/solutions/strat-jushuitan-kingdee-cloud-1725-nc3962f46-ad0ef198

Comments