[Nepomuk] Cannot set values for abstract property 'diff.stats.hunk_count'

[Weng Xuetian]

On Thu, Apr 5, 2012 at 9:21 PM, Sebastian Trüg <sebastian at trueg.de> wrote:
> i think you are right. :)
>
> On 04/05/2012 03:15 PM, Vishesh Handa wrote:
>>
>>
>> On Thu, Apr 5, 2012 at 6:31 PM, Sebastian Trüg <sebastian at trueg.de
>> <mailto:sebastian at trueg.de>> wrote:
>>
>>     the question is: does anyone need diff metadata?
>>
>>
>> People don't need a lot of things. It's still nice to have at those rare
>> occasion, though honestly, I cannot think of any.
>>
>> I'm guessing from your question, that you do not think people need it.
>> So, why don't we just scrap it? I'm not willing to put in the effort of
>> fixing it, and there really isn't anyone hacking on Strigi these days.
>>
>>
>>
>>     On 04/05/2012 02:53 PM, Vishesh Handa wrote:
>>     >
>>     >
>>     > On Thu, Apr 5, 2012 at 5:54 PM, Sebastian Trüg <sebastian at trueg.de
>>     <mailto:sebastian at trueg.de>
>>     > <mailto:sebastian at trueg.de <mailto:sebastian at trueg.de>>> wrote:
>>     >
>>     >     simply a buggy strigi analyzer.
>>     >
>>     >
>>     > Actually, it's a buggy strigi analyzer for which there is so simple
>>     > solution.
>>     >
>>     > We can either -
>>     >
>>     > a) Throw it away
>>     > b) Write an ontology for handling diff files.
>>     >
>>     > I'm more inclined towards the first option. Mostly cause formalizing
>>     > ontologies is a long process.
>>     >
>>     >
>>     >
>>     >     On 04/05/2012 12:45 PM, Martin Steigerwald wrote:
>>     >     > Hi!
>>     >     >
>>     >     > After restarting to track / fix another unrelated issue I found
>>     >     >
>>     >     > "/usr/bin/nepomukservicestub(2046)" Soprano: "Invalid
>>     argument (1)":
>>     >     > "Cannot set values for abstract property
>>     'diff.stats.hunk_count'."
>>     >     > [/usr/bin/nepomukservicestub]
>>     "/usr/bin/nepomukservicestub(2046)"
>>     >     Soprano:
>>     >     > "Invalid argument (1)": "Cannot set values for abstract property
>>     >     > 'diff.stats.hunk_count'."
>>     >     > "/usr/bin/nepomukservicestub(2046)" Soprano: "Invalid
>>     argument (1)":
>>     >     > "Cannot set values for abstract property
>>     'diff.stats.hunk_count'."
>>     >     > [/usr/bin/nepomukservicestub]
>>     "/usr/bin/nepomukservicestub(2046)"
>>     >     Soprano:
>>     >     > "Invalid argument (1)": "Cannot set values for abstract property
>>     >     > 'diff.stats.hunk_count'."
>>     >     >
>>     >     > in ~/.xsession-errors.
>>     >     >
>>     >     > Worth a bug report or already known?
>>     >     >
>>     >     > Thanks,
>>     >     _______________________________________________
>>     >     Nepomuk mailing list
>>     >     Nepomuk at kde.org <mailto:Nepomuk at kde.org>
>>     <mailto:Nepomuk at kde.org <mailto:Nepomuk at kde.org>>
>>     >     https://mail.kde.org/mailman/listinfo/nepomuk
>>     >
>>     >
>>     >
>>     >
>>     > --
>>     > Vishesh Handa
>>     >
>>
>>
>>
>>
>> --
>> Vishesh Handa
>>
> _______________________________________________
> Nepomuk mailing list
> Nepomuk at kde.org
> https://mail.kde.org/mailman/listinfo/nepomuk

There are several case that strigi analyzer is not used in the right way in KDE.

For example, lokalize seems depends on strigi analyzer for PO files,
but the property of this analyzer is similar to this case, like
"translate.blablabla", not standard ontology.

If I remember correct, kompare requires "diff_stat"?

So there might need to be a way to let program to use strigi internal,
but not put their plugin into index service.

(sorry for not send to the list at first)