Woops, you are correct, the file formats doc is out of date.
It should be #1.
I'll fix it -- thank you for raising it!
Mike
Ivan Vasilev wrote:
Hi Guys,
In the File Formats web page (http://lucene.apache.org/java/2_3_0/
fileformats.html) there is section describing Segments File, where
we read:
Segments --> Format, Version, NameCounter, ...
...
Format is -1 as of Lucene 1.4 and -3
(SemgentInfos.FORMAT_SINGLE_NORM_FILE) as of Lucene 2.1.
...
On my opinion the last sentence is not completely right. I mean it
should contain addition somthing like this:
1) "and -4 (SemgentInfos.FORMAT_SHARED_DOC_STORE) as of Lucene 2.3."
or like this:
2) "and as of Lucene 2.3 -4 (SemgentInfos.FORMAT_SHARED_DOC_STORE)
in cases when shared stored fields and term vectors exist or -3
(SemgentInfos.FORMAT_SINGLE_NORM_FILE) when no sharing exists."
So my question is what of the two suggestions is correct?
According to my tests with our IndexRecoverer tool the 1) is
correct. Here I will have to mention what does this tool. It
creates segments file out of given set of segments. I did not made
a lot of tests yet but in one single case when I created segments
file for a segments set that does not contain any shared data when
I use Format=SemgentInfos.FORMAT_SINGLE_NORM_FILE Luke can not open
it using Lucene 2.3, but when
Format=SemgentInfos.FORMAT_SHARED_DOC_STORE it opens it correctly.
Best Regards,
Ivan
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
