Qeasy Cloud
Get Started

Purchase Inbound Sync Strategy: Chanjet T+ to Jushuitan Other Inbound

· 谢锴斌· Integration Solutions· 21 views· 4 min read
畅捷通T+Jushuitan供应链集成Inventory Sync轻易云采购入库单

What this strategy solves

In a typical retail or manufacturing operation, purchase inbound orders first land in Chanjet T+ for financial posting and inventory accounting, while physical stocktaking is handled by the Jushuitan WMS. The two systems speak different "inbound" languages: T+ treats inbound as a financial document with supplier, tax, and price, while Jushuitan cares about physical flow and warehouse stock. Manual double entry is slow, and after about three months the two ledgers drift apart. This sync strategy pushes T+ inbound facts into Jushuitan as "other inbound orders," so both systems see the same physical inbound event.

Data flow and field mapping

The flow is one-way: T+ (source) → Qeasy middle layer → Jushuitan (target). The T+ document header carries the master info, the body carries line items, and the middle layer handles field conversion and code mapping.

Key field mapping (representative items, real mapping depends on the actual interfaces):

DimensionChanjet T+ (source)Middle layerJushuitan (target)
Doc typePurchase inboundFixed mappingOther inbound
Doc numberSource doc no.Pass-throughExternal doc no.
SupplierPartner codeCode mapSupplier code
WarehouseWarehouse codeCode mapWarehouse code
Item codeInventory codeCode map (centralized)SKU code
QuantityInbound qtyUnit conversionInbound qty
Price / taxTax-inclusive price, taxPer target rulesUsually optional
Business dateDoc datePass-throughInbound date

Centralized code mapping is the most common pattern among Qeasy customers: keep T+ inventory, supplier, and warehouse codes in one mapping table, so when a new item is added you change it once, not across many strategies.

How to configure it on Qeasy

Typical configuration points on the Qeasy integration platform:

  1. Source pull: use T+'s purchase inbound API, filter on modification time for deltas, and de-duplicate by document number to avoid re-sends.
  2. Field transformation: in the middle layer, run code mapping, unit conversion, and decide which price fields to keep. Tax-related fields are typically dropped per target requirements.
  3. Target write: call Jushuitan's "other inbound order" create API, putting the T+ document number into the external document number field for traceability and reversal.
  4. Exception handling: missing code mapping, empty required fields, or target-side validation failures should go into a retry queue or human task list, not be silently dropped.
  5. Logging and traceability: every sync records the source doc no., target doc no., processing time, and operator (if any), so reconciliation is straightforward.

Implementation steps

We usually run a "full backfill first, then steady-state delta" rollout:

  • Phase 1: full backfill (go-live day)

    • Push all unsynced historical purchase inbound orders to Jushuitan in time-windowed batches (for example, the last 90 days), to avoid overwhelming the target.
    • After the run, do a manual sample reconciliation to confirm both ledgers agree.
  • Phase 2: delta start point

    • Use the last successful timestamp from Phase 1 as the delta start point, then switch to incremental mode.
    • A typical incremental schedule is every 5–15 minutes; high-volume businesses can go sub-minute, low-volume ones every 30 minutes.
    • On the scheduling side, headers and bodies are submitted in phases: lock the header first (doc no., supplier, warehouse), then submit line items in batches, so failures can roll back cleanly.
  • Phase 3: ops and monitoring

    • Configure a daily reconciliation job to compare inbound quantities and document status on both sides.
    • Set alerts on failed tasks so you never get "silent failure."

Pitfalls revisited

  1. Code mapping not centralized. The classic crash: every strategy maintains its own mapping, and three months later a single SKU change requires a platform-wide search. The safe move is one shared mapping table consumed by all related strategies.
  2. No source-side de-duplication. T+ inbound orders may go through multiple versions during approval. Without de-duplication by doc number, you get duplicate inbound orders in Jushuitan.
  3. Force-writing tax/price into the target. Jushuitan's "other inbound" does not always require price and tax. Pushing them anyway triggers target validation failures. Drop them in the middle layer per the agreed rules.
  4. Wrong delta start point. A common mistake is using the max timestamp from the full backfill day as the delta start, which leaves boundary documents behind. The safe move is to compare the full backfill cutoff time with the delta start time and verify no gap before switching.
  5. Unbounded retries. Network blips can cause the same document to block the queue repeatedly. Always set a max retry count and a backoff policy.

When to use it, when not to

Use it when: the source of truth is T+ for finance, Jushuitan owns physical stock, and both sides want inventory ledgers to agree; inbound volume is moderate; field mapping is reasonably stable.

Don't use it when: the target system needs strict purchase-cost accounting (in that case keep the loop inside T+); or the source-side master data changes weekly and field mapping rules keep moving. Fix the master data first, then talk about sync.

Original content. Please credit the source when reposting: https://www.qeasy.cloud/insights/solutions/strat-p9210a3-jushuitan-9687-ne28d2008-2127a208

Comments