[ 
https://issues.apache.org/jira/browse/PDFBOX-4471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16776775#comment-16776775
 ] 

Michael Klink commented on PDFBOX-4471:
---------------------------------------

Ah, I see your spec quote is missing a bit of context, the paragraph it is in 
starts with 

{quote}
To update an existing appearance stream to reflect a new field value,...
{quote}

This can be interpreted to imply that one has determined before that _updating_ 
an _existing_ appearance stream in the case at hand is the appropriate measure 
_to reflect a new field value_. One might have determined this by actually 
verifying that the appearance stream in question is only referenced by 
annotations that have to be updated in the same way. Or one might have 
determined this by knowing that one oneself has created the stream and no tool 
had the chance in the time between to re-use the appearance.

I don't want to say that the author of that paragraph has considered this 
option, merely the wording is open to this interpretation.

<sarcasm on>Isn't it great to work on the base of a specification which at 
heart still is a non-normative reference...<sarcasm off>

> Different fields look alike if they have the same /AP/N before value change
> ---------------------------------------------------------------------------
>
>                 Key: PDFBOX-4471
>                 URL: https://issues.apache.org/jira/browse/PDFBOX-4471
>             Project: PDFBox
>          Issue Type: Bug
>          Components: AcroForm
>    Affects Versions: 2.0.13
>            Reporter: Tilman Hausherr
>            Priority: Major
>         Attachments: csOnePage.pdf
>
>
> See attached SO answer - in the attached file, different fields point to the 
> same /AP/N object. I think this could be fixed by changing 
> {{AppearanceGeneratorHelper.setAppearanceValue()}} by removing the "if" 
> segment in {{if (appearance != null && appearance.isStream())}} and keep the 
> "else" part only.
> I'm wondering why we are reusing the existing appearance stream, is there are 
> use case to this? I ran the tests after making the change and there was a 
> difference but it was invisible.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

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

Reply via email to