[Nepomuk] Re: Review Request: Fixed range of nfo:interlaceMode in the PNG analyzer

[Sebastian Trueg]

> On June 21, 2011, 7:36 a.m., Jos van den Oever wrote:
> > This patch gives less information out and changes the way of reporting and even the type of the reported triple.
> > 
> > "This patch fixes the range in the PNG analyzer which till now used a string naming the algorithm name."
> > 
> > What does 'fixes the range' mean? Is this the only possible workaround?
> >

In fact there is not less information as PNG only uses a single interlace algorithm: Adam7. However, as mentioned above I see the ugliness in the boolean. That is why I proposed to introduce a second property down the line which states the algorithm. For the time being, however, a boolean is much more useful for queries than a string is.
Apart from that Strigi does use the Nepomuk ontologies. Thus, it should also honour the ranges defined in them. In this case nfo:interlaceMode has a boolean range. An alternative would be to patch the ontology. IMHO that would be overkill at this point though.


- Sebastian


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/101704/#review4050
-----------------------------------------------------------


On June 20, 2011, 2:18 p.m., Sebastian Trueg wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/101704/
> -----------------------------------------------------------
> 
> (Updated June 20, 2011, 2:18 p.m.)
> 
> 
> Review request for Nepomuk and Strigi.
> 
> 
> Summary
> -------
> 
> nfo:interlaceMode has the range xsd:boolean. This might be a bit weird but in the end makes sense since we mostly want to know if an image is interlaced or not. (Down the road we might think of adding another property which holds the name of the interlacing algorithm used.)
> This patch fixes the range in the PNG analyzer which till now used a string naming the algorithm name.
> 
> 
> Diffs
> -----
> 
>   lib/endanalyzers/pngendanalyzer.cpp 804bd5f 
> 
> Diff: http://git.reviewboard.kde.org/r/101704/diff
> 
> 
> Testing
> -------
> 
> Tested on a few PNGs. The value is correct.
> 
> 
> Thanks,
> 
> Sebastian
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.kde.org/pipermail/nepomuk/attachments/20110621/a6dc489e/attachment.htm