Re: Oracle Text, PDF files

Since the DLL name hasn't changed, would you need to do anything besides copying? (I'm not familiar with the process of unregistering/registering).

-Randy

"Khalid Eidoo" <khalid-eidoo_at_rogers.com> wrote in message news:JM0ia.17348$Xw1.15100_at_news01.bloor.is.net.cable.rogers.com...
> It should be possible to just replace the old filter with the new one. In
> the *NIX versions, we can simply replace the ctxhx executable with
another,
> and we're in business. Presumably the same can be done with the DLL -
> hopefully its just a matter of unregistering the old filter and
registering
> the new DLL.
>
> Khalid.
>
>
> "Randy Nichols" <randynichols_at_yahoo.com> wrote in message
> news:YKZha.1305$ey1.116231_at_newsread1.prod.itd.earthlink.net...
> > We just upgraded to 9.2.0.2 and in agreement with you we are getting
> better
> > results. We can now index some problem PDF files, and we also use the
> > markup. The markup is fairly primitive in terms of preserving original
> doc
> > structure, but we are happy to have the capability at all.
> >
> > I am wondering if it is possible to upgrade just the INSO filter for the
> > older Oracle versions, possibly by copying the oractxx9.dll file?
> Probably
> > a long shot, but would be useful if true.
> >
> > -Randy
> >
> > "Khalid Eidoo" <khalid.eidoo_at_utoronto> wrote in message
> > news:YA7ha.34323$KlE.1199_at_news04.bloor.is.net.cable.rogers.com...
> > > You didn't mention which version of Oracle you were using. We recently
> > > upgraded to 9.2.0.2 on Linux, which specifically had some updates to
the
> > > INSO filter. We find that indexing PDFs using the new filter provides
> > > marginally better results.
> > >
> > > We primarily perform gists in our application, and the fact that there
> is
> > > less garbage in them indicates to us that the INSO filter is doing a
> > better
> > > job. We haven't tried highlighting yet though.
> > >
> > > Khalid.
> > >
> > > "Randy Nichols" <randynichols_at_yahoo.com> wrote in message
> > > news:NSGga.22238$jA2.1996308_at_newsread2.prod.itd.earthlink.net...
> > > > Oracle Text apparently has problems indexing and highlighting
certain
> > PDF
> > > > files.
> > > >
> > > > Is anyone aware of any good solutions for making Oracle Text more
> robust
> > > > with respect to PDF files?
> > > >
> > > > Is there a service pack for Oracle Text that fixes some problems
> > > concerning
> > > > PDF files?
> > > >
> > > > Anyone have experience with third-party filters that do a good job
> > > indexing
> > > > and highlighting (markup) of PDF files with Oracle Text?
> > > >
> > > > Thanks,
> > > >
> > > > R. Nichols
> > > >
> > > >
> > >
> > >
> > >
> >
> >
> >
>
>
>
Received on Mon Mar 31 2003 - 17:33:38 CST