[
https://issues.apache.org/jira/browse/MATH-1341?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gilles updated MATH-1341:
-------------------------
Attachment: RandomUtils.java
I attached a new {{RandomUtils}} class which I propose as a replacement for
{{RandomDataGenerator}}.
Having removed all methods that are _trivial_ syntactic sugar for the sampler
API in {{o.a.c.m.distribution}}, the number of lines in {{RandomUtils}} is ~60%
of the old {{RandomDataGenerator}}.
The "nextSecureXxx" methods were also removed as most were identical to their
"non-secure" counterpart, except for the underlying RNG (secure vs not). In
{{RandomUtils}}, secure versions are thus "automatically" achieved by passing a
secure RNG to the class's constructor.
Methods {{nextHexString}} and {{nextSecureHexString}} were different in that
the latter uses a {{MessageDigest}} object in its processing.
The two codes were merged in a single method and a boolean argument selects one
or the other behaviour.
{{RandomUtils}} also obsoletes class {{RandomGeneratorFactory}} whose
functionality is replaced by the following factory method:
{noformat}
public static UniformRandomProvider asUniformRandomProvider(final Random rng) {
// ....
}
{noformat}
Old "features" intentionally left out:
* Lazy initialization (if you don't need the RNG, you don't need an instance of
this class)
* Reseeding of the RNG (if necessary - although a bad idea - the caller can
hold a reference to the {{Random}} object being wrapped, and call "setSeed" on
it)
Please have a look, and let me know of any functionality I may have missed to
port to the new class.
> "RandomDataGenerator" is brittle
> --------------------------------
>
> Key: MATH-1341
> URL: https://issues.apache.org/jira/browse/MATH-1341
> Project: Commons Math
> Issue Type: Bug
> Reporter: Gilles
> Priority: Minor
> Labels: API, cleanup, deprecation, thread-safety
> Fix For: 4.0
>
> Attachments: RandomUtils.java
>
>
> Class {{RandomDataGenerator}} can easily be misused as it advertizes a method
> to access its internal RNG (which is _not_ thread-safe).
> The class is also a mixed bag of "data generators" that are either "secure"
> or not.
> Moreover it uses the "lazy initialization" pattern (for the RNG instance)
> solely because of this duality; otherwise users that need one or the other
> form of data generation will obviously always use the RNG since all data
> generation methods need it.
> This entails also a performance hit (albeit tiny) as each call checks whether
> the RNG has been initialized already.
> The clean solution would be to separate the two types of data generation
> (secure vs not) into different classes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
