[ 
https://issues.apache.org/jira/browse/PDFBOX-3069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14978890#comment-14978890
 ] 

Tilman Hausherr commented on PDFBOX-3069:
-----------------------------------------

Preferably duplicate code, because the full Apache Imaging would be too big. 
Please
- put the svn URL, date and revision number in the javadoc 
- make the API similar to existing API
- compression should always be G4, lets not bother with anything less
- accept only bitonal images, fail gracefully if somebody brings a different 
image
- create a test that reads from a TIFF image into a BufferedImage, pass this to 
the new method, and them compare the result (copy from existing tests)

The only thing I'm not sure is whether all agree to do this before the 2.0 
release, or rather wait for after the release. From me, a "weak +1" for adding 
it.

> Add code for CCITT compression
> ------------------------------
>
>                 Key: PDFBOX-3069
>                 URL: https://issues.apache.org/jira/browse/PDFBOX-3069
>             Project: PDFBox
>          Issue Type: Improvement
>          Components: PDModel
>            Reporter: Christian Gawron
>            Priority: Minor
>              Labels: ccitt, compression
>
> The class org.apache.pdfbox.pdmodel.graphics.image.CCITTFactory contains code 
> to create a PDImageXObject from an existing TIFF file. Unlike the 
> corresponding JPEGFactory it does *not* provide any code to create a 
> PDImageXObject from a BufferedImage. The main reason is that pdfbox contains 
> only code to decompress CCITT group 3/4 files, code for compression is 
> missing.
> Based on the compression code in Apache Commons Imaging I implemented the 
> missing method and I'm willing to contribute it.
> I can provide a patch, but I would need advise whether it's preferable to 
> duplicate code from Apache Commons Imaging or to create a dependency on it.



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

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

Reply via email to