directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Emmanuel Lecharny (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FC-28) The safeText method is probably not useful, and can be faster
Date Tue, 25 Nov 2014 17:09:13 GMT

    [ https://issues.apache.org/jira/browse/FC-28?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14224846#comment-14224846
] 

Emmanuel Lecharny commented on FC-28:
-------------------------------------

May be it would be better to have it set to optional by default. We also might want those
checks to be relative to the type of backend (it's not necessarily LDAP)

> The safeText method is probably not useful, and can be faster
> -------------------------------------------------------------
>
>                 Key: FC-28
>                 URL: https://issues.apache.org/jira/browse/FC-28
>             Project: FORTRESS-CORE
>          Issue Type: Improvement
>    Affects Versions: 1.0-RC35
>            Reporter: Emmanuel Lecharny
>             Fix For: 1.1.0
>
>
> When we want to validate an entity before injecting it in the server, we call the RegExUtil.safeText()
method.
> It's checking something the server will already check (each value injected in the server
will be controlled). Of course, we can set limits like the max length, but this can be done
when defining the AttributeType (using the
> {NNN}
> notation).
> Otherwise, the way the regexp is evaluated is costly : we compile the gexep every time,
when it could be done only once.
> Pattern safeTextPattern = Pattern.compile(safeTextPatternStr);
> can be done globally.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message