lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Toke Eskildsen (JIRA)" <j...@apache.org>
Subject [jira] Updated: (LUCENE-1990) Add unsigned packed int impls in oal.util
Date Tue, 23 Feb 2010 15:35:27 GMT

     [ https://issues.apache.org/jira/browse/LUCENE-1990?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Toke Eskildsen updated LUCENE-1990:
-----------------------------------

    Attachment: LUCENE-1990-te20100223.patch

I've renamed most of the classes to short form, as the "Packed"-prefix did was not that descriptive
and fixed some bugs. Still pending is the mutable writer and a bug in persistence for aligned64.
Good news (for Lucene at least) is that an airplane blocking snowdrift means that I have time
this week for continued hacking.

{quote}
But, now that we have getMutable, can we make the concrete impls
package private? Javadocs for Mutable.set should note that the size
is fixed once you allocate it.
{quote}

The implementations are now package private, but I only put the note about fixed size on the
getMutable-method. There's nothing wrong with creating a custom auto growing Mutable.

{quote}
We should state clearly that these are all unsigned ints storage.
{quote}

Done.

{quote}
Maybe rename PackedDirectInt to PackedDirect32 (and Short to 16,
Byte to 8).
{quote}

Done (Direct8, Direct16, Direct32 and Direct64).

{quote}
The @see in the new IndexInput.readShort is wrong (referencing
writeInt).
{quote}

Fixed.

{quote}
Can you add @lucene.internal to the javadocs?
{quote}

Should this also be applied to package private classes? Marking those as internal seems redundant.

{quote}
Seems like once we stomp the bugs, beef up the tests, and merge
PRIORITY and BLOCK_PREFERENCE (into maybe STORAGE?) for
the public API, we are nearly done?
{quote}

I've removed BLOCK_PREFERENCE from the API. It's still used internally, mainly to do controlled
testing. Tests are beefed up (and currently fails for aligned, so clearly beefing worked).

> Add unsigned packed int impls in oal.util
> -----------------------------------------
>
>                 Key: LUCENE-1990
>                 URL: https://issues.apache.org/jira/browse/LUCENE-1990
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: Index
>            Reporter: Michael McCandless
>            Priority: Minor
>         Attachments: LUCENE-1990-te20100122.patch, LUCENE-1990-te20100210.patch, LUCENE-1990-te20100212.patch,
LUCENE-1990-te20100223.patch, LUCENE-1990.patch, LUCENE-1990_PerformanceMeasurements20100104.zip
>
>
> There are various places in Lucene that could take advantage of an
> efficient packed unsigned int/long impl.  EG the terms dict index in
> the standard codec in LUCENE-1458 could subsantially reduce it's RAM
> usage.  FieldCache.StringIndex could as well.  And I think "load into
> RAM" codecs like the one in TestExternalCodecs could use this too.
> I'm picturing something very basic like:
> {code}
> interface PackedUnsignedLongs  {
>   long get(long index);
>   void set(long index, long value);
> }
> {code}
> Plus maybe an iterator for getting and maybe also for setting.  If it
> helps, most of the usages of this inside Lucene will be "write once"
> so eg the set could make that an assumption/requirement.
> And a factory somewhere:
> {code}
>   PackedUnsignedLongs create(int count, long maxValue);
> {code}
> I think we should simply autogen the code (we can start from the
> autogen code in LUCENE-1410), or, if there is an good existing impl
> that has a compatible license that'd be great.
> I don't have time near-term to do this... so if anyone has the itch,
> please jump!

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


Mime
View raw message