[
https://issues.apache.org/jira/browse/MSHARED-953?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Elliotte Rusty Harold updated MSHARED-953:
------------------------------------------
Description:
The trim parameter in buildDom methods in Xpp3DomBuilder is based on a serious
misunderstanding of how the characters method in SAX works. As currently
written, trim can potentially remove all whitespace in text content from a
document. A little more commonly, it will randomly remove or not remove
whitespace depending on how the SAX parser chooses to split text nodes when it
reports them.
It might be possible to implement this correctly at the cost of a lot of code
complexity if anyone is using this. More easily, we could make the parameter a
no-op and never trim.
was:
The trim parameter in buildDom methods in Xpp3DomBuilder is base don a serious
misunderstanding of how the characters method in SAX works. As currently
written, trim can potentially remove all whitespace in text content from a
document. A little more commonly, it will randomly remove or not remove
whitespace depending on how the SAX parser chooses to split text nodes when it
reports them.
It might be possible to implement this correctly at the cost of a lot of code
complexity if anyone is using this. More easily, we could make the parameter a
no-op and never trim.
> trimming can remove all whitespace
> ----------------------------------
>
> Key: MSHARED-953
> URL: https://issues.apache.org/jira/browse/MSHARED-953
> Project: Maven Shared Components
> Issue Type: Bug
> Components: maven-shared-utils
> Reporter: Elliotte Rusty Harold
> Priority: Critical
>
> The trim parameter in buildDom methods in Xpp3DomBuilder is based on a
> serious misunderstanding of how the characters method in SAX works. As
> currently written, trim can potentially remove all whitespace in text content
> from a document. A little more commonly, it will randomly remove or not
> remove whitespace depending on how the SAX parser chooses to split text nodes
> when it reports them.
> It might be possible to implement this correctly at the cost of a lot of code
> complexity if anyone is using this. More easily, we could make the parameter
> a no-op and never trim.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
