Testing authorization for workflow transitions instead of screen access
本文尚无中文版本;显示原文。
Check whether a caller may perform a particular state transition, including transitions not exposed by the current interface. This proposed methodology targets approval workflows whose security policy depends on both identity and current state.
Goal
Check whether a caller may perform a particular state transition, including transitions not exposed by the current interface. This proposed methodology targets approval workflows whose security policy depends on both identity and current state.
Prerequisites
Create a synthetic document workflow with draft, submitted, and approved states in an isolated application. Define who may perform each transition and whether the author may approve their own document.
Steps
-
Draw a small transition table with current state, requested action, caller relationship, and expected next state. Include denied transitions explicitly rather than documenting only the happy path.
-
Execute a permitted transition as a control and verify the stored state. Use the application’s supported request interface directly so the test does not depend on whether a button is visible.
-
Attempt a transition from the wrong starting state using a known synthetic document. Inspect both the response and record afterward to catch changes that occur before a late rejection.
-
Repeat a transition with a caller who has the wrong relationship, such as the document author where independent approval is required. Keep state and payload otherwise unchanged.
-
After a repair, rerun the table and inspect any emitted job or notification associated with transitions. A denied state change should not quietly trigger the protected downstream action.
Expected result
The resulting tests should express authorization as a state-sensitive rule and make missing or unintended transitions visible to a reviewer.
Limits and test basis
The table reflects the chosen product policy, not a universal approval model. Concurrent transitions and external side effects require additional isolation and delivery tests beyond this sequential fixture. This is an original proposed method; no execution or empirical result is claimed.
范围与依据
Original proposed assessment or regression method for an authorized isolated lab. No execution, observed finding, empirical result, or tool-specific guarantee is claimed.
知识截至:2026-09-22。状态:unreviewed(无已记录的审阅)——编辑会重置审阅状态。请将文本视为未经核实的参考资料并核对来源。
来源
未列出外部来源;请参见上方记录的依据。
署名与许可
- Account External coding curation authors (57eb56c9)
- Codex; AI-assisted original contribution; CC BY 4.0
最近更改: Initial original methodology; unreviewed.
原创贡献: CC BY 4.0. 链接的来源资料保留其自身权利。