[
https://issues.apache.org/jira/browse/LANG-808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16024699#comment-16024699
]
Pascal Schumacher edited comment on LANG-808 at 5/25/17 12:58 PM:
------------------------------------------------------------------
RandomStringUtils will be deprecated in commons-lang 3.6 in favor of
RandomStringGenerator from commons-text. Please use
[RandomStringGenerator|https://commons.apache.org/proper/commons-text/javadocs/api-release/org/apache/commons/text/RandomStringGenerator.html]
instead. Thanks!
was (Author: pascalschumacher):
RandomStringUtils will be deprecated in commons-lang 3.6 in favor of
RandomStringGenerator from commons-text. Please use
[RandomStringGenerator|https://commons.apache.org/proper/commons-text/javadocs/api-release/org/apache/commons/text/RandomStringGenerator.html]
instead.
> RandomStringUtils inefficient init of start and end for letters and numbers;
> misleading Javadoc
> -----------------------------------------------------------------------------------------------
>
> Key: LANG-808
> URL: https://issues.apache.org/jira/browse/LANG-808
> Project: Commons Lang
> Issue Type: Bug
> Components: lang.*
> Affects Versions: 2.6, 3.1
> Reporter: Sebb
> Assignee: Pascal Schumacher
>
> If either letters or numbers is true, RandomStringUtils(count, 0, 0, letters,
> numbers, ...) sets the default start index to space, intending to support the
> set of printable characters (according to the Javadoc).
> However, the code later uses Character.isLetter and Character.isDigit, which
> exclude space and some other printable characters.
> There is currently no direct support for generating printable characters;
> this would probably require another flag.
> The current Javadoc is misleading, and the current code is inefficient
> because some characters cannot currently be used.
> The lowest char which is either digit or letter is '0' (zero).
> The lowest char which is a letter is 'A' (upper case a).
> The Javadoc for the randomNumeric(int) method says:
> bq. Characters will be chosen from the set of numeric characters.
> This does not make clear that the numerics are restricted to the ASCII set
> (because of the way start=end=0 is handled).
> Similarly for randomAlphabetic and randomAlphanumeric.
> Is it the intention that randomAlphabetic etc. be restricted to ASCII?
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
