[
https://issues.apache.org/jira/browse/IMAGING-264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17180558#comment-17180558
]
Bruno P. Kinoshita commented on IMAGING-264:
--------------------------------------------
Sounds sensible to me. Had a look around to see if I could find anything in the
specifications about rounding resolution (which is a LONG), but had no luck.
For the sake of being consistent, I think it's OK that we change, as
JpegImageParser is doing it too. Quickly created a BMP in GIMP, with dpi=72.6,
reproduced the issue where it was rounded down to 72. Added the suggested
`match.round`, and it produced 73.
> BMP Parser physicalWidthDpi and physicalHeightDpi truncated before rounding
> off.
> --------------------------------------------------------------------------------
>
> Key: IMAGING-264
> URL: https://issues.apache.org/jira/browse/IMAGING-264
> Project: Commons Imaging
> Issue Type: Bug
> Components: Format: BMP
> Affects Versions: 1.0-alpha2
> Reporter: Ram
> Assignee: Bruno P. Kinoshita
> Priority: Minor
> Fix For: 1.0-alpha3
>
>
> In BMPImageParser class, the values are typecasted to int before rounding off
> so output varies(other parsers has this logic correct).
> int physicalWidthDpi = (int)((double)bhi.hResolution * 0.0254D);
> int physicalHeightDpi = (int)((double)bhi.vResolution * 0.0254D);
>
> Expected :
> int physicalWidthDpi = (int)*Math.round*((double)bhi.hResolution * 0.0254D);
> int physicalHeightDpi = (int)*Math.round*((double)bhi.vResolution * 0.0254D);
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
