lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jack Krupansky" <j...@basetechnology.com>
Subject Re: Trying to search java.lang.NullPointerException in log file.
Date Mon, 22 Jul 2013 14:42:50 GMT
Standard anlyzer/tokenizer will use white space and other punctuation to 
delimit tokens. The rules are a little complicated (although I tried to 
summarize them for Solr in my book) - the same rules apply for Lucene.

Verify that you are properly constructing a PhraseQuery from your analyzed 
text at query time. What is the exact query text and what are the exact 
analyzer tokens for that query text and how many are there?

-- Jack Krupansky

-----Original Message----- 
From: Ankit Murarka
Sent: Monday, July 22, 2013 10:29 AM
To: java-user@lucene.apache.org
Subject: Re: Trying to search java.lang.NullPointerException in log file.

First thing first : Same analyzer is being used to index and to search.

Now, I am not using any custom analyzer to split the string and get the
tokens. I was assuming StandardAnalyzer might be using whitespaces to
split the content. If that is not the case then I must have got it
completely wrong.

So for searching "java.lang.NullPointer" how should I proceed? This
string might be present after : like ":java.lang.NullPointer" . In both
cases I want to search for "java.lang.NullPointer" only.


On 7/22/2013 7:51 PM, Jack Krupansky wrote:
> "This is because the StandardAnalyzer must be splitting the words on 
> "SPACES" and since there is no space present here. The entire string is 
> converted into 1 token."
>
> Those statements are inconsistent! I mean, what code is converting the 
> entire string to 1 token and eliminating white space? Is that your own 
> code before you hand the string to the standard analyzer??? That makes no 
> sense. I mean, the standard analyzer is using the standard tokenizer that 
> doesn't do that!
>
> Are you applying the same analyzer at query time as you do at index time? 
> It is not uncommon for Lucene users to forget to do that. If you don't, 
> then you will have to hand-analyze the query string and simulate exactly 
> what the standard analyzer did at index time.
>
> So, please clarify your situation.
>
>
> -- Jack Krupansky
>
> -----Original Message----- From: Ankit Murarka
> Sent: Monday, July 22, 2013 6:24 AM
> To: java-user@lucene.apache.org
> Subject: Trying to search java.lang.NullPointerException in log file.
>
> Hello. I am trying to search java.lang.NullPointerException in a log
> file. The log file is huge.
>
> However I am unable to search it. This is because the StandardAnalyzer
> must be splitting the words on "SPACES" and since there is no space
> present here. The entire string is converted into 1 token.
>
> What can be a possible way of finding
> "Exception:java.lang.NullPointerException" in a log file.
>
> The string may be different also. Suppose "Exception:
> java.lang.NullPointerException error occured"
>
> I am trying to use Phrase Query but I am not sure if that will serve the
> purpose.
>
> Can please someone suggest.
>


-- 
Regards

Ankit Murarka

"Peace is found not in what surrounds us, but in what we hold within."


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org 


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org


Mime
View raw message