[
https://issues.apache.org/jira/browse/PDFBOX-6166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18072988#comment-18072988
]
Andreas Lehmkühler commented on PDFBOX-6166:
--------------------------------------------
{quote}
Wouldn't it be better to return size + available as size are the bytes already
being consumed and available bytes consumable (although there might be more).
In case all bytes have been consumed size is the total as available is 0 so
that's correct by then. And before it's a better estimate.
{quote}
Yes, of course. I didn't take into account the fact that the return value of
available is a moving target if the inputstream is processed.
> Improve some implementation flaws in NonSeekableRandomAccessReadInputStream
> ---------------------------------------------------------------------------
>
> Key: PDFBOX-6166
> URL: https://issues.apache.org/jira/browse/PDFBOX-6166
> Project: PDFBox
> Issue Type: Improvement
> Components: IO, Parsing
> Affects Versions: 3.0.6 PDFBox, 4.0.0
> Reporter: Andreas Lehmkühler
> Assignee: Andreas Lehmkühler
> Priority: Major
>
> [~tilman] found some flaws in the implementation of
> NonSeekableRandomAccessReadInputStream based on a posting from Daniel Gredler
> on the users mailinglist
> As NonSeekableRandomAccessReadInputStream is backed by an InputStream we have
> to think about the implementation of some methods as the length of the given
> inputStream isn't available at the beginning
> * length: the current implementation relies on the internal member "size"
> * available: the current default implementation relies on length()
> * readFully: the current default implementation relies on length()
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
