[
https://issues.apache.org/jira/browse/NUMBERS-100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Heinrich Bohne resolved NUMBERS-100.
------------------------------------
Resolution: Fixed
> Code in file FractionTest.java is unsatisfactory
> ------------------------------------------------
>
> Key: NUMBERS-100
> URL: https://issues.apache.org/jira/browse/NUMBERS-100
> Project: Commons Numbers
> Issue Type: Improvement
> Components: fraction
> Reporter: Heinrich Bohne
> Priority: Trivial
> Time Spent: 1h 50m
> Remaining Estimate: 0h
>
> The following characteristics of the file {{FractionTest.java}} can be
> improved:
> * The second-to-last try-catch-block in the method {{testAdd()}} is a
> duplicate of the preceding try-catch-block and is therefore redundant.
> * In the method {{testPow()}}, the conditions {{assertFraction(9, 49,
> a.pow(2))}} and {{assertFraction(49, 9, a.pow(-2))}} are tested twice each
> (once in the block after {{a}}'s declaration, and a second time in the block
> after {{b}}'s declaration. This is probably a typo.
> * The last two assertions in the method {{testGetReducedFraction()}} pass the
> parameters to the method {{Assert.assertEquals(long, long)}} in the wrong
> order (the expected value should go first).
> * Several methods in this class contain a number of tests that use shared
> local variables but are completely independent of each other because these
> local variables get assigned new values at the beginning of a test. The fact
> that the scope of these local variables encompasses all those independent
> tests makes the code look more confusing than necessary.
> * Except for the method {{testGoldenRatio()}}, the throwing of an exception
> is tested with a construct involving the swallowing of an exception, rather
> than an explicit syntax.
> * The helper method {{assertFraction(int, int, Fraction)}} is neglected
> throughout large sections of the class in favor of
> {{Assert.assertEquals(long, long)}} pairs, increasing the amount of code
> duplication.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
