[ 
https://issues.apache.org/jira/browse/PDFBOX-5862?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17871018#comment-17871018
 ] 

Michael Klink commented on PDFBOX-5862:
---------------------------------------

Looking at a number of methods of {{COSArray}} it seems that the intention is 
to allow actual {{null}} elements in the underlying list. Maybe some hardening 
in some methods would be appropriate.

When doing {{COSArray}} cleanups, one should probably also look at other 
details. For example, the typed getters (like {{getName}}) with default value 
parameter claim to return that default value if the entry in question is 
{{null}}; actually, though, they also return it if the entry is of a different 
type {like not a {{COSName}} in {{getName}}}.

> is o.a.p.cos.COSArray supposed to contain nulls?
> ------------------------------------------------
>
>                 Key: PDFBOX-5862
>                 URL: https://issues.apache.org/jira/browse/PDFBOX-5862
>             Project: PDFBox
>          Issue Type: Improvement
>          Components: Documentation
>    Affects Versions: 3.0.2 PDFBox
>            Reporter: Dieter von Holten
>            Priority: Minor
>
> inspection by spotbugs shows a warning at line 573, which references a null 
> from the list and compares it with the given value. When the list contains a 
> null, and a null is passed by the caller, this is considered a match and the 
> index is returned to the caller.
> However, other locations in COSArray cannot deal with null entries in the 
> list.
> i propose:
> 1.) update the document to state whether null is a valid list-entry
> 2.) consider the use of COSNull when the caller tries to insert null. in this 
> case, the users (those who get() values out) of COSArray need to be aware of 
> the change
>  
>  
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to