arazzo: 1.0.1
info:
  title: Microsoft Excel Delete a Table Row by Index
  summary: Read a table's rows, delete one by its zero-based index, and confirm the removal.
  description: >-
    Row deletion in the Graph Excel API is by position, not by key, which makes
    a read-before-delete mandatory: the caller must see the current rows to know
    which index actually holds the record it means to remove, and indexes shift
    as soon as a row is deleted. This flow reads the rows, deletes the target
    index, and re-reads to confirm the new row set. Every step spells out its
    request inline so the flow can be read and executed without opening the
    underlying OpenAPI.
  version: 1.0.0
sourceDescriptions:
- name: excelGraphApi
  url: ../openapi/microsoft-excel-graph-api.yaml
  type: openapi
workflows:
- workflowId: prune-table-row
  summary: Safely remove a single row from an Excel table by index.
  description: >-
    Snapshots the table rows so the caller can confirm the target index, deletes
    that row, and re-reads the rows to verify the deletion.
  inputs:
    type: object
    required:
    - itemId
    - worksheetIdOrName
    - rowIndex
    properties:
      itemId:
        type: string
        description: The drive item id of the workbook file.
      worksheetIdOrName:
        type: string
        description: The worksheet holding the target table.
      rowIndex:
        type: integer
        description: >-
          The zero-based index of the row to delete, verified against the
          snapshot taken by this flow before the delete is issued.
  steps:
  - stepId: openSession
    description: Open a persisting session so the deletion is committed to the stored workbook.
    operationId: createWorkbookSession
    parameters:
    - name: item-id
      in: path
      value: $inputs.itemId
    requestBody:
      contentType: application/json
      payload:
        persistChanges: true
    successCriteria:
    - condition: $statusCode == 201
    outputs:
      sessionId: $response.body#/id
  - stepId: resolveTable
    description: Resolve the table on the worksheet that the row will be deleted from.
    operationId: listTables
    parameters:
    - name: item-id
      in: path
      value: $inputs.itemId
    - name: worksheet-id-or-name
      in: path
      value: $inputs.worksheetIdOrName
    - name: workbook-session-id
      in: header
      value: $steps.openSession.outputs.sessionId
    successCriteria:
    - condition: $statusCode == 200
    - context: $response.body
      condition: $.value.length > 0
      type: jsonpath
    outputs:
      tableId: $response.body#/value/0/id
  - stepId: snapshotRows
    description: >-
      Read every row before deleting so the caller holds a copy of the row about
      to be destroyed and can confirm the index points at the intended record.
    operationId: listTableRows
    parameters:
    - name: item-id
      in: path
      value: $inputs.itemId
    - name: table-id
      in: path
      value: $steps.resolveTable.outputs.tableId
    - name: workbook-session-id
      in: header
      value: $steps.openSession.outputs.sessionId
    successCriteria:
    - condition: $statusCode == 200
    - context: $response.body
      condition: $.value.length > 0
      type: jsonpath
    outputs:
      rowsBefore: $response.body#/value
  - stepId: deleteRow
    description: >-
      Delete the row at the supplied index. This is destructive, and every row
      after it shifts up by one, so any cached indexes are invalid afterwards.
    operationId: deleteTableRow
    parameters:
    - name: item-id
      in: path
      value: $inputs.itemId
    - name: table-id
      in: path
      value: $steps.resolveTable.outputs.tableId
    - name: row-index
      in: path
      value: $inputs.rowIndex
    - name: workbook-session-id
      in: header
      value: $steps.openSession.outputs.sessionId
    successCriteria:
    - condition: $statusCode == 204
  - stepId: verifyDeletion
    description: >-
      Re-read the rows to confirm the record is gone and to hand the caller the
      re-indexed row set for any further deletions.
    operationId: listTableRows
    parameters:
    - name: item-id
      in: path
      value: $inputs.itemId
    - name: table-id
      in: path
      value: $steps.resolveTable.outputs.tableId
    - name: workbook-session-id
      in: header
      value: $steps.openSession.outputs.sessionId
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      rowsAfter: $response.body#/value
  outputs:
    tableId: $steps.resolveTable.outputs.tableId
    rowsBefore: $steps.snapshotRows.outputs.rowsBefore
    rowsAfter: $steps.verifyDeletion.outputs.rowsAfter