[pylucene-dev] Index Fields

Thapar, Ashwin akthapar at wharton.upenn.edu
Tue May 23 10:40:02 PDT 2006


Hi,

I have successfully got pyLucene running, and it works great at indexing
text files in a folder and then searching for simple strings within the
index.

I want to take my system a step further and add more specific fields,
than the basic "contents" field to the index (so that I can later sort
by these fields)

I have structured text files, and I want to extract specific pieces of
information from each text file, and store it in a separate field in the
index.

For example, each text file contains a line starting with Date Added:,
which is followed by a date. What I would like to do, is somehow extract
this date from each text file, and store it in a field called "date" for
each  record in the index.

Is there any way to do this?

Thanks.

Regards,

Ashwin

________________________________

From: pylucene-dev-bounces at osafoundation.org
[mailto:pylucene-dev-bounces at osafoundation.org] On Behalf Of Taquin Ho
Sent: Tuesday, May 23, 2006 12:38 PM
To: pylucene-dev at osafoundation.org
Subject: [pylucene-dev] Does PyLucene Index with temp files?

 

I have a straightforward installation of pyLucene 9.1 and am calling it
through an apache server with the help of a python script.

The apache server is not running as the local system account but a
windows user which has access to the necessary shares where the indexes
are stored.

A problem occurs when the index is to be created and I am guessing it is
because PyLucene attempts to create the index in some temporary file or
directory that is does not have access to.

  File "core\src\bs_search.py", line 64, in __init__ 
    self.analyzer, True) 
  File "C:\programs\python24\lib\site-packages\PyLucene.py", line 2261,
in __init__ 
    newobj = _PyLucene.new_IndexWriter(*args) 
JavaError: java.io.IOException: CreateFile failed 

I need to know if this is the case, and if so, where the location is, so
I can give the particular windows user access to the location.

If its not the case, does anyone have any idea what the problem is? 

:-) 

Taquin Ho 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osafoundation.org/pipermail/pylucene-dev/attachments/20060523/c67c6682/attachment.htm


More information about the pylucene-dev mailing list