[
https://issues.apache.org/jira/browse/LUCENE-1794?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12742974#action_12742974
]
Shai Erera commented on LUCENE-1794:
------------------------------------
Robert, what I meant about pulling SavedStreams up to Analyzer (few comments
above) was to do something like this:
{code}
class Analyzer {
protected static class Streams {
public Tokenizer tokenizer;
public TokenStream tokenStream;
}
...
}
class MyAnalyzer extends Analyzer {
public reusableTokenStream() {
Streams streams = getPrevTS();
if (streams == null) {
streams = new Streams();
streams.tokenizer = new Tokenizer();
streams.tokenStream = new TokenStream();
setPrevTS(streams);
} else {
streams.tokenizer.reset(reader);
streams.tokenStream.reset();
}
return streams.tokenStream;
}
{code}
This will just save the declaration of SavedStreams or Streams in all
sub-classes. In addition we can do the following:
# Define reset(String, Reader) on Streams, so that everyone just calls
streams.reset(), instead of resetting tokenizer and tokenStream. Streams will
do that internally.
# Define a protected abstract getTokenizer() on Analyzer that all Analyzers
implement. (due to back-compat, this can throw UOE - let's leave it for now).
# Have Analyzer's reusableTokenStream look like the following:
{code}
public TokenStream reusableTokenStream(String field, Reader reader) {
Streams streams = getPreviousTokenStream();
if (streams == null) {
streams = new Streams();
streams.tokenizer = getTokenizer(field, reader);
streams.tokenStream = tokenStream();
setPrevTS(streams);
} else {
streams.reset(field, reader);
}
return streams.tokenStream;
}
{code}
And that can be even more simplified, by having Streams define a ctor which
accepts Tokenizer and TokenStream. We can also instead of doing "new Streams()"
call a method newStreams() so that sublcasses can override if they want to
provide a different Streams impl. Not a must, and we might even consider the
whole thing final (Streams, reusableTokenStream ? etc.)
That will save some code in that patch I believe. What do you think?
I haven't touched the back-compat issues yet - let's discuss the idea first.
> implement reusableTokenStream for all contrib analyzers
> -------------------------------------------------------
>
> Key: LUCENE-1794
> URL: https://issues.apache.org/jira/browse/LUCENE-1794
> Project: Lucene - Java
> Issue Type: Improvement
> Components: contrib/analyzers
> Reporter: Robert Muir
> Assignee: Robert Muir
> Priority: Minor
> Fix For: 2.9
>
> Attachments: LUCENE-1794.patch, LUCENE-1794.patch, LUCENE-1794.patch,
> LUCENE-1794.patch, LUCENE-1794.patch
>
>
> most contrib analyzers do not have an impl for reusableTokenStream
> regardless of how expensive the back compat reflection is for indexing speed,
> I think we should do this to mitigate any performance costs. hey, overall it
> might even be an improvement!
> the back compat code for non-final analyzers is already in place so this is
> easy money in my opinion.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
