[Summary view] [Print] [Text view]

   1  @title Differential User Guide: Test Plans
   2  @group userguide
   3  
   4  This document describes things you should think about when developing a test
   5  plan.
   6  
   7  = Overview =
   8  
   9  When you send a revision for review in Differential you must include a test plan
  10  (this can be disabled or made optional in the config). A test plan is a
  11  repeatable list of steps which document what you have done to verify the
  12  behavior of a change. A good test plan convinces a reviewer that you have been
  13  thorough in making sure your change works as intended and has enough detail to
  14  allow someone unfamiliar with your change to verify its behavior.
  15  
  16  This document has some common things to think about when developing or reviewing
  17  a test plan. Some of these suggestions might not be applicable to the software
  18  you are writing; they are adapted from Facebook's internal documentation.
  19  
  20  = All Changes =
  21  
  22    - **Error Handling:** Are errors detected and handled properly? How does your
  23      change deal with error cases? Did you test them and make sure you got the
  24      right error messages and the right behavior? It's important that you test
  25      what happens when things go wrong, not just that the change works if
  26      everything else also works.
  27    - **Service Impact:** How does your change affect services like memcache,
  28      thrift, or databases? Are you adding new cachekeys or queries? Will
  29      this change add a lot of load to services?
  30    - **Performance:** How does your change affect performance? **NOTE**: If
  31      your change is a performance-motivated change, you should include
  32      measurements, profiles or other data in your test plan proving that you have
  33      improved performance.
  34    - **Unit Tests:** Is your change adequately covered by unit tests? Could you
  35      improve test coverage? If you're fixing a bug, did you add a test to prevent
  36      it from happening again? Are the unit tests testing just the code in
  37      question, or would a failure of a database or network service cause your
  38      test to fail?
  39    - **Concurrent Change Robustness:** If you're making a refactoring change, is
  40      it robust against people introducing new calls between the time you started
  41      the change and when you commit it? For example, if you change the parameter
  42      order of some function from ##f(a, b)## to ##f(b, a)## and a new callsite is
  43      introduced in the meantime, could it go unnoticed? How bad would that be?
  44      (Because of this risk, you should almost never make parameter order
  45      changes in weakly typed languages like PHP and Javascript.)
  46    - **Revert Plan:** If your change needs to be reverted and you aren't around,
  47      are any special steps or considerations that the reverter needs to know
  48      about? If there are, make sure they're adequately described in the "Revert
  49      Plan" field so someone without any knowledge of your patch (but with a
  50      general knowledge of the system) can successfully revert your change.
  51    - **Security:** Is your change robust against XSS, CSRF, and injection
  52      attacks? Are you verifying the user has the right capabilities or
  53      permissions? Are you consistently treating user data as untrustworthy? Are
  54      you escaping data properly, and using dangerous functions only
  55      when they are strictly necessary?
  56    - **Architecture:** Is this the right change? Could there be a better way to
  57      solve the problem? Have you talked to (or added as reviewers) domain experts
  58      if you aren't one yourself? What are the limitations of this solution? What
  59      tradeoffs did you make, and why?
  60  
  61  = Frontend / User-Facing Changes =
  62  
  63    - **Static Resources:** Will your change cause the application to serve more
  64      JS or CSS? Can you use less JS/CSS, or reuse more?
  65    - **Browsers:** Have you tested your change in multiple browsers?