Charlie,
Sure... Now that I know what you're looking for, I can explain it in more
details.
First of all, you do need to download and install Adobe's PDF IFilter and
ensure that it has installed correctly. One quick way to verify this is to
use the Indexing Service and setup a catalog to index a folder with pdf
files and use IS to search on unique keywords in the pdf files. If you get
the correct hits, then you know that the PDF IFilter has installed
correctly.
As for SQL Server 2000 and Full-Text Search of pdf files, there's a bit more
setup required. First of all, you have to store the PDF files in a column
defined with an IMAGE datatype and also have a column that defines the "file
extension" to be bound to the IMAGE column. Specifically, this "file
extension" column must use a datatype of char(3) or varchar(4) or sysname in
order for the "Microsoft Search" service to correctly recognize the file
time and launch the correct IFilter. Additionally, how you import or insert
the document (pdf file) into your SQL table is important as well. Using
TextCopy.exe or ADO Stream are both methods that work successfully. Once you
have this completed, you then should start a Full Population on your FT
enable table and then review the Application event log for "Microsoft
Search" source events for any errors and/or messages of a successful
population. Finally, you can issue your SQL FTS contains query, where your
column "content" is the IMAGE column and test your query using a unique
keyword from the PDF file:
Select * from Docs where CONTAINS(Content, 'teststring')
FYI, the Registry key, you should focus on and can edit is:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\ContentIndex
DLLsToRegister -- your PDFFilt.dll should be listed here.
Regards,
John
[quoted text, click to view] "Charlie" <charles.vandusen@[NO SPAM]comcast.net> wrote in message
news:OSXXD81sDHA.2340@TK2MSFTNGP12.phx.gbl...
> Hi John
>
> Can you be more specific?
>
> I am trying to use the indexing and search functionality from Windows
> Sharepoint Services, but clearly WSS is using the fulltext features of SQL
> 2000 to get there.
>
> I have a SQL table with a fulltext index. If I add an Office document
(.doc,
> .xls, etc.) I can search it successfully using sql like
>
> Select * from Docs where CONTAINS(Content, 'teststring')
> WHERE:
> Docs is the table included in the fulltext index
> Content is an Image column
> 'teststring' is a string which is in the .doc file
>
> If I add a row to the Docs table and put a .pdf file into the Content
image
> field, I can search forever and never find the pdf record.
>
> One of the registry keys that I have been focussed on is
> HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Search\1.0\Gathering
> Manager\DLLsToRegister. The key on my server does not show the pdffilt.dll
> in its list. It also happens that it is a binary key. Can you confirm the
> same?
>
> Any other hints you can offer?
>
>
> "John Kane" <jt-kane@comcast.net> wrote in message
> news:urXPjyusDHA.980@TK2MSFTNGP10.phx.gbl...
> > Charlie,
> > That's what I'm using and it works great on Win2K and Win2003.
> >
> > John
> >
> >
> > "Charlie" <charles.vandusen@[NO SPAM]comcast.net> wrote in message
> > news:uoRra7tsDHA.1680@TK2MSFTNGP12.phx.gbl...
> > > Greetings folks
> > >
> > > What are people using to index and search PDF files. Is PDFFilt
version
> > 5.0
> > > still the latest and greatest.
> > >
> > > TIA
> > >
> > > Charlie
> > >
> > >
> >
> >
>
>