lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Arpad KATONA" <>
Subject RERE: RERE: Highlighting, startOffset, endOffset
Date Wed, 14 May 2003 12:33:20 GMT
Hi Shoba,

do you know the medieval song :

Orientis partibus
adventavit asinus

i really deserve to be called an "asinus", it is hard to do but i must APOLOGIZE to all the
Lucene-Users for my mistake, i remember tavarish Otis Gospodnetic's e-mail on Wed, 2 Apr 2003
08:34:30 -0800 (PST) as he wrote : "Funny how this is the outcome of 90% of the problems people
have with software - their own mistakes"...

In fact, the code i sent to the list runs fine, as you say. We are using here some "FrenchAnalyzer"
and in the code i had run here on my computer there was that FrenchAnalyzer in stead of the
StandardAnalyzer. In order to avoid embarrassing questions about "what is that french-thing
you use there?" (nevertheless, even embarrassing questions might be sometimes helpful) i replaced
"french" by "standard" in the sample code without testing the behavior of that piece of prog
with StandardAnalyzer.

The fault is in our french-specific filters, as i began to look at the java source files (a
work of an ex-Collegue of us), i found that in same places the tokens are created in this
way e.g. :
  Token tok = new Token(sTxt, 0, sTxt.length(), "word");
it is not surprising if Token.startOffset() returns then always zero...

Well, one more funny anecdote on the Lucene Users Mailig List.



-----Message d'origine-----
De : Shoba Ramachandran []
Envoyé : mardi 13 mai 2003 17:16
À : Lucene Users List
Objet : Re: RERE: Highlighting, startOffset, endOffset

Hi Arpad,

Tested your codes with this example. Seems to run fine....

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message