주문부터 대금 회수까지, 청구 및 송장 발행 데이터 템플릿

The following configuration outlines the data entities and field mappings required for the export project within the Dynamics 365 Data Management Framework. A post-processing script or tool is required to union these outputs into a single event log.

**1. Activity: Sales Order Fulfilled**
*   **Entity Name:** `Sales packing slip headers`
*   **Mapping:**
    *   `InvoiceNumber`: Use `SalesOrderNumber` as a key to later join to the invoice data and find the corresponding `InvoiceNumber`.
    *   `ActivityName`: 'Sales Order Fulfilled' (Hardcoded Value)
    *   `EventTime`: `DeliveryDate`
    *   `UserResponsible`: `CreatedBy`

**2. Activity: Invoice Generated**
*   **Entity Name:** `Sales invoice headers V2`
*   **Filter:** `InvoiceType` is not 'CreditNote'.
*   **Mapping:**
    *   `InvoiceNumber`: `InvoiceNumber`
    *   `ActivityName`: 'Invoice Generated' (Hardcoded Value)
    *   `EventTime`: `InvoiceDate` and `CreatedTime`
    *   `InvoiceAmount`: `TotalInvoiceAmount`
    *   `CustomerName`: `InvoiceCustomerName`
    *   `UserResponsible`: `CreatedBy`
    *   `PaymentDueDate`: `DueDate`
    *   `InvoiceStatus`: `DocumentState`

**3. Activities: Invoice Submitted for Approval & Invoice Approved**
*   **Entity Name:** `Workflow tracking status table`
*   **Filter:** `ContextTableName` is 'CustInvoiceJour'.
*   **Mapping:**
    *   `InvoiceNumber`: Join `ContextRecId` to `RecId` from `Sales invoice headers V2` to get the `InvoiceNumber`.
    *   `ActivityName`: Use a conditional rule, if `TrackingStatus` is 'Submitted', set to 'Invoice Submitted for Approval'; if 'Approved', set to 'Invoice Approved'.
    *   `EventTime`: `CreatedDateTime`
    *   `UserResponsible`: `User`

**4. Activity: Invoice Sent to Customer**
*   **Note:** This is an inferred event. The posting date is used as a proxy.
*   **Entity Name:** `Sales invoice headers V2`
*   **Mapping:**
    *   `InvoiceNumber`: `InvoiceNumber`
    *   `ActivityName`: 'Invoice Sent to Customer' (Hardcoded Value)
    *   `EventTime`: `PostedDateTime` (or equivalent posting timestamp)

**5. Activity: Payment Due Date Reached**
*   **Note:** This is a calculated event generated during post-processing.
*   **Source Entity:** `Sales invoice headers V2`
*   **Logic:** For each invoice where the `DueDate` has passed and the invoice is not fully settled.
*   **Mapping:**
    *   `InvoiceNumber`: `InvoiceNumber`
    *   `ActivityName`: 'Payment Due Date Reached' (Hardcoded Value)
    *   `EventTime`: `DueDate`

**6. Activity: Payment Reminder Issued**
*   **Entity Name:** `Dunning history letters`
*   **Mapping:**
    *   `InvoiceNumber`: Join from the dunning history to the customer transactions to find the related `InvoiceNumber`.
    *   `ActivityName`: 'Payment Reminder Issued' (Hardcoded Value)
    *   `EventTime`: `DunningLetterDate`
    *   `UserResponsible`: `CreatedBy`

**7. Activity: Customer Payment Received**
*   **Entity Name:** `CustPaymentJournalLineEntity`
*   **Mapping:**
    *   `InvoiceNumber`: The `Invoice` field may be populated, but often settlements are needed to link payments to invoices. This might require staging the payment data first.
    *   `ActivityName`: 'Customer Payment Received' (Hardcoded Value)
    *   `EventTime`: `ReceivedDate` or `DocumentDate`
    *   `UserResponsible`: `CreatedBy`

**8. Activity: Cash Applied to Invoice**
*   **Entity Name:** `Customer settlement`
*   **Mapping:**
    *   `InvoiceNumber`: `OffsetInvoice`
    *   `ActivityName`: 'Cash Applied to Invoice' (Hardcoded Value)
    *   `EventTime`: `TransDate`
    *   `UserResponsible`: `CreatedBy`

**9. Activity: Payment Reversed**
*   **Source Entity:** `Customer settlement`
*   **Logic:** Filter for transactions where a settlement was later reversed. This may require looking for cancellation markers.
*   **Mapping:**
    *   `InvoiceNumber`: `OffsetInvoice`
    *   `ActivityName`: 'Payment Reversed' (Hardcoded Value)
    *   `EventTime`: Reversal transaction timestamp.

**10. Activity: Credit Note Issued**
*   **Entity Name:** `Sales invoice headers V2`
*   **Filter:** `InvoiceType` is 'CreditNote'.
*   **Mapping:**
    *   `InvoiceNumber`: `InvoiceNumber`
    *   `ActivityName`: 'Credit Note Issued' (Hardcoded Value)
    *   `EventTime`: `InvoiceDate`
    *   `InvoiceAmount`: `TotalInvoiceAmount` (will be negative)
    *   `CustomerName`: `InvoiceCustomerName`
    *   `UserResponsible`: `CreatedBy`

**11. Activity: Invoice Closed**
*   **Note:** This is an inferred event based on settlement data.
*   **Source Entity:** `Customer settlement`
*   **Logic:** Use the latest `TransDate` for an invoice from the settlement data where the invoice balance becomes zero.
*   **Mapping:**
    *   `InvoiceNumber`: `OffsetInvoice`
    *   `ActivityName`: 'Invoice Closed' (Hardcoded Value)
    *   `EventTime`: The latest `TransDate` that resulted in the invoice being fully settled.

import requests
import pandas as pd
from datetime import datetime, timezone

class=class="hl-string">"hl-comment"># --- Configuration ---
d365_url = class="hl-string">"https://[your-d365-instance].operations.dynamics.com" class=class="hl-string">"hl-comment"># Your D365 F&O URL
tenant_id = class="hl-string">"[Your Tenant ID]"
client_id = class="hl-string">"[Your Client ID]"
client_secret = class="hl-string">"[Your Client Secret]"
company = class="hl-string">"[Your Company DataAreaId, e.g., USMF]"
start_date = class="hl-string">"2023-01-01"
end_date = class="hl-string">"2023-12-31"

class=class="hl-string">"hl-comment"># --- Authentication ---
def get_access_token():
    url = fclass="hl-string">"https://login.microsoftonline.com/{tenant_id}/oauth2/v2.0/token"
    payload = {
        class="hl-string">'grant_type': class="hl-string">'client_credentials',
        class="hl-string">'client_id': client_id,
        class="hl-string">'client_secret': client_secret,
        class="hl-string">'scope': fclass="hl-string">"{d365_url}/.default"
    }
    response = requests.post(url, data=payload)
    response.raise_for_status()
    return response.json()[class="hl-string">'access_token']

class=class="hl-string">"hl-comment"># --- OData Helper ---
def fetch_all_data(access_token, resource_path):
    headers = {
        class="hl-string">'Authorization': fclass="hl-string">'Bearer {access_token}',
        class="hl-string">'Content-Type': class="hl-string">'application/json'
    }
    full_url = fclass="hl-string">"{d365_url}/data/{resource_path}"
    results = []
    while full_url:
        response = requests.get(full_url, headers=headers)
        response.raise_for_status()
        data = response.json()
        results.extend(data[class="hl-string">'value'])
        full_url = data.get(class="hl-string">'@odata.nextLink')
    return results

class=class="hl-string">"hl-comment"># --- Main Extraction Logic ---
def main():
    print(class="hl-string">"Acquiring access token...")
    token = get_access_token()
    event_log = []
    today = datetime.now(timezone.utc).date()

    print(class="hl-string">"Fetching Invoices and Credit Notes...")
    invoice_filter = fclass="hl-string">"$filter=InvoiceDate ge {start_date} and InvoiceDate le {end_date}"
    invoice_fields = class="hl-string">"$select=InvoiceNumber,InvoiceDate,SalesOrderNumber,InvoiceAmount,CustomerAccount,Due_Date,WorkerSalespersonName,createdDateTime,RecId,SalesInvoiceHeader_SalesStatus"
    invoices = fetch_all_data(token, fclass="hl-string">"CustInvoiceJours?cross-company=true&{invoice_filter}&{invoice_fields}")

    class=class="hl-string">"hl-comment"># Create a lookup for invoice RecIds for workflow queries
    invoice_recid_map = {inv[class="hl-string">'RecId']: inv[class="hl-string">'InvoiceNumber'] for inv in invoices}

    for inv in invoices:
        case_id = inv[class="hl-string">'InvoiceNumber']
        invoice_amount = inv[class="hl-string">'InvoiceAmount']
        customer_name = inv.get(class="hl-string">'CustomerAccount', class="hl-string">'') class=class="hl-string">"hl-comment"># Placeholder, join to CustTable for name
        user_responsible = inv.get(class="hl-string">'WorkerSalespersonName', class="hl-string">'System')
        due_date = inv.get(class="hl-string">'Due_Date')
        invoice_status = inv.get(class="hl-string">'SalesInvoiceHeader_SalesStatus', class="hl-string">'Open')

        class=class="hl-string">"hl-comment"># Activity: Invoice Generated / Credit Note Issued
        activity_name = class="hl-string">'Credit Note Issued' if invoice_amount < 0 else class="hl-string">'Invoice Generated'
        event_log.append({
            class="hl-string">'InvoiceNumber': case_id,
            class="hl-string">'ActivityName': activity_name,
            class="hl-string">'EventTime': inv[class="hl-string">'createdDateTime'],
            class="hl-string">'InvoiceAmount': invoice_amount,
            class="hl-string">'CustomerName': customer_name,
            class="hl-string">'UserResponsible': user_responsible,
            class="hl-string">'Department': class="hl-string">'[Configure based on your system]',
            class="hl-string">'PaymentDueDate': due_date,
            class="hl-string">'InvoiceStatus': invoice_status
        })

        class=class="hl-string">"hl-comment"># Activity: Invoice Sent to Customer (proxy)
        event_log.append({
            class="hl-string">'InvoiceNumber': case_id,
            class="hl-string">'ActivityName': class="hl-string">'Invoice Sent to Customer',
            class="hl-string">'EventTime': inv[class="hl-string">'InvoiceDate'],
            class="hl-string">'InvoiceAmount': invoice_amount,
            class="hl-string">'CustomerName': customer_name,
            class="hl-string">'UserResponsible': user_responsible,
            class="hl-string">'Department': class="hl-string">'[Configure based on your system]',
            class="hl-string">'PaymentDueDate': due_date,
            class="hl-string">'InvoiceStatus': invoice_status
        })
        
        class=class="hl-string">"hl-comment"># Activity: Sales Order Fulfilled (proxy via invoice date, complex to get actual packing slip date)
        class=class="hl-string">"hl-comment"># A more accurate query would trace SalesInvoiceLinesV2 -> SalesOrderLines -> CustPackingSlipJour
        event_log.append({
            class="hl-string">'InvoiceNumber': case_id,
            class="hl-string">'ActivityName': class="hl-string">'Sales Order Fulfilled',
            class="hl-string">'EventTime': inv[class="hl-string">'InvoiceDate'], class=class="hl-string">"hl-comment"># Using InvoiceDate as a proxy
            class="hl-string">'InvoiceAmount': invoice_amount,
            class="hl-string">'CustomerName': customer_name,
            class="hl-string">'UserResponsible': user_responsible,
            class="hl-string">'Department': class="hl-string">'[Configure based on your system]',
            class="hl-string">'PaymentDueDate': due_date,
            class="hl-string">'InvoiceStatus': invoice_status
        })

        class=class="hl-string">"hl-comment"># Activity: Payment Due Date Reached (calculated)
        if due_date and pd.to_datetime(due_date).date() < today and invoice_status != class="hl-string">'Paid':
            event_log.append({
                class="hl-string">'InvoiceNumber': case_id,
                class="hl-string">'ActivityName': class="hl-string">'Payment Due Date Reached',
                class="hl-string">'EventTime': due_date,
                class="hl-string">'InvoiceAmount': invoice_amount,
                class="hl-string">'CustomerName': customer_name,
                class="hl-string">'UserResponsible': class="hl-string">'System',
                class="hl-string">'Department': class="hl-string">'Finance',
                class="hl-string">'PaymentDueDate': due_date,
                class="hl-string">'InvoiceStatus': invoice_status
            })

    print(class="hl-string">"Fetching Settlements for Cash Applied and Invoice Closed events...")
    settlement_filter = fclass="hl-string">"$filter=InvoiceDate ge {start_date} and InvoiceDate le {end_date}"
    settlement_fields = class="hl-string">"$select=InvoiceNumber,SettlementDate,OffsetTransactionVoucher,OffsetRecId,CanBeReversed"
    settlements = fetch_all_data(token, fclass="hl-string">"CustSettlements?cross-company=true&{settlement_filter}&{settlement_fields}")

    class=class="hl-string">"hl-comment"># Group settlements by invoice
    settlements_by_invoice = pd.DataFrame(settlements).groupby(class="hl-string">'InvoiceNumber')

    for case_id, group in settlements_by_invoice:
        last_settlement = group.sort_values(class="hl-string">'SettlementDate').iloc[-1]

        for _, settlement in group.iterrows():
            class=class="hl-string">"hl-comment"># Activity: Cash Applied to Invoice
            event_log.append({
                class="hl-string">'InvoiceNumber': case_id,
                class="hl-string">'ActivityName': class="hl-string">'Cash Applied to Invoice',
                class="hl-string">'EventTime': settlement[class="hl-string">'SettlementDate'],
                class="hl-string">'UserResponsible': class="hl-string">'System',
                class=class="hl-string">"hl-comment"># Add other attributes from original invoice if needed
            })

            class=class="hl-string">"hl-comment"># Activity: Customer Payment Received (date of original payment)
            class=class="hl-string">"hl-comment"># This requires a second hop to CustTrans using OffsetRecId
            class=class="hl-string">"hl-comment"># For simplicity, we use settlement date as a proxy. A full implementation would query CustTrans.
            event_log.append({
                class="hl-string">'InvoiceNumber': case_id,
                class="hl-string">'ActivityName': class="hl-string">'Customer Payment Received',
                class="hl-string">'EventTime': settlement[class="hl-string">'SettlementDate'], class=class="hl-string">"hl-comment"># Proxy timestamp
                class="hl-string">'UserResponsible': class="hl-string">'System',
            })

            class=class="hl-string">"hl-comment"># Activity: Payment Reversed (if applicable)
            if settlement[class="hl-string">'CanBeReversed']: class=class="hl-string">"hl-comment"># A simplistic check, real logic is more complex
                 event_log.append({
                    class="hl-string">'InvoiceNumber': case_id,
                    class="hl-string">'ActivityName': class="hl-string">'Payment Reversed',
                    class="hl-string">'EventTime': settlement[class="hl-string">'SettlementDate'], class=class="hl-string">"hl-comment"># Reversal date would be different in reality
                    class="hl-string">'UserResponsible': class="hl-string">'System',
                 })
        
        class=class="hl-string">"hl-comment"># Activity: Invoice Closed (inferred from last settlement)
        class=class="hl-string">"hl-comment"># This logic assumes the invoice is fully paid by the last settlement. A robust check would sum amounts.
        event_log.append({
            class="hl-string">'InvoiceNumber': case_id,
            class="hl-string">'ActivityName': class="hl-string">'Invoice Closed',
            class="hl-string">'EventTime': last_settlement[class="hl-string">'SettlementDate'],
            class="hl-string">'UserResponsible': class="hl-string">'System',
        })

    print(class="hl-string">"Fetching Payment Reminders...")
    reminder_filter = fclass="hl-string">"$filter=CollectionLetterDate ge {start_date} and CollectionLetterDate le {end_date}"
    reminders = fetch_all_data(token, fclass="hl-string">"CustCollectionLetterJours?cross-company=true&{reminder_filter}")
    for rem in reminders:
        class=class="hl-string">"hl-comment"># Linking reminders to invoices can be complex. This is a simplified approach.
        if rem.get(class="hl-string">'InvoiceNum'):
            event_log.append({
                class="hl-string">'InvoiceNumber': rem[class="hl-string">'InvoiceNum'],
                class="hl-string">'ActivityName': class="hl-string">'Payment Reminder Issued',
                class="hl-string">'EventTime': rem[class="hl-string">'CollectionLetterDate'],
                class="hl-string">'UserResponsible': rem.get(class="hl-string">'UserId', class="hl-string">'System'),
            })

    print(class="hl-string">"Fetching Workflow History...")
    class=class="hl-string">"hl-comment"># This is a complex query. It requires knowing the TableId for CustInvoiceJour (e.g., 523)
    class=class="hl-string">"hl-comment"># The filter below is illustrative and needs to be adapted for your specific workflow setup.
    workflow_filter = fclass="hl-string">"$filter=createdDateTime ge {start_date}T00:00:00Z"
    wf_history = fetch_all_data(token, fclass="hl-string">"WorkflowTrackingStatusTables?cross-company=true&{workflow_filter}")
    for item in wf_history:
        if item.get(class="hl-string">'ContextRecId') in invoice_recid_map:
            case_id = invoice_recid_map[item[class="hl-string">'ContextRecId']]
            activity_name = None
            if item[class="hl-string">'TrackingStatus'] == class="hl-string">'Submitted':
                activity_name = class="hl-string">'Invoice Submitted for Approval'
            elif item[class="hl-string">'TrackingStatus'] == class="hl-string">'Approved':
                activity_name = class="hl-string">'Invoice Approved'
            
            if activity_name:
                event_log.append({
                    class="hl-string">'InvoiceNumber': case_id,
                    class="hl-string">'ActivityName': activity_name,
                    class="hl-string">'EventTime': item[class="hl-string">'createdDateTime'],
                    class="hl-string">'UserResponsible': item.get(class="hl-string">'User', class="hl-string">'System'),
                })

    print(fclass="hl-string">"Total events extracted: {len(event_log)}")
    if not event_log:
        print(class="hl-string">"No events found for the given criteria.")
        return

    df = pd.DataFrame(event_log)
    df = df.sort_values(by=[class="hl-string">'InvoiceNumber', class="hl-string">'EventTime'])
    df.to_csv(class="hl-string">'d365_billing_event_log.csv', index=False)
    print(class="hl-string">"Event log saved to d365_billing_event_log.csv")

if __name__ == class="hl-string">"__main__":
    main()