Javas Blogs global

Re: lucene datatypes

Thu, 13 Nov 2008 12:19:33 -0500

Solr provides semantics on Lucene fields for handling other data types and there are some tools (DateTools. NumberTools) for converting some types to Strings for searching. But yeah. Strings are pretty much the only thing Lucene cares about when it comes to searching.-GrantOn Nov 15. 2007 at 2:24 AM. Heba Farouk wrote:> Hello,>>>> I would like to ask how lucene handles different datatypes or > "String"> is the only available datatype.>>>> Best regards,>>>> Heba Farouk>> Software Engineer>> Bibliotheca Alexandrina>---------------------------------------------------------------------To unsubscribe e-mail: java-user-unsubscribe@lucene apache orgFor additional commands e-mail: java-user-help@lucene apache org

Forex Groups - Tips on Trading

Related article:
http://mail-archives.apache.org/mod_mbox/lucene-java-user/200711.mbox/%3C94F2BB84-A9D8-480F-B85B-7F07E6B3389A@apache.org%3E

Re: =?GB2312?B?tPC4tDogaG93IHRvIGVmZmVjaWVudGx5IGltcGxlbWVudCB0aA ...

Tue, 01 Jan 2008 21:15:37 -0500

John Wang wrote:> Would payload bring home the bacon?> -John>> Yes if you used payloads instead of stored fields your performanceshould be much better. Try and index one special term per enter (e g score:pagerank) andindex one position with a payload for each doc. Then when you retrievehits open a TermPositions using the special term get the payload andincorporate it in the docs' score. The performance overhead should be comparable to adding one AND-term toyour query.-Michael> On 11/15/07. Zhou Qi wrote:> >> Thank you my score is fixed score from the properties of the page but at>> first we need to adjust the score for a promising prove.>> I have tried one way of manually re-ranking all the documents by the>> search results. But it needs to iterate all the retrieved results and channel>> the re-ranking score (stored in the index) to sum the overall advance. It is>> inefficient. How to alter that by a new approach?>>>> Sorry for making you misunderstanding. Thanks>>>> Best Regards,>>>> Zhou Qi>> Dept. Computer Science & Engineering>> Shanghai Jiaotong University>>>> ---------------------------------------------------------------------To unsubscribe e-mail: java-user-unsubscribe@lucene apache orgFor additional commands e-mail: java-user-help@lucene apache org

Forex Groups - Tips on Trading

Related article:
http://mail-archives.apache.org/mod_mbox/lucene-java-user/200711.mbox/%3C473C9108.2080504@gmail.com%3E

[jira] Resolved: (LUCENE-1052) Add an "termInfosIndexDivisor" to ...

Sat, 15 Dec 2007 15:03:31 -0500

[ ] Michael McCandless resolved LUCENE-1052. ---------------------------------------- Resolution: Fixed I just committed this. Thanks throw & Doug! > Add an "termInfosIndexDivisor" to IndexReader > --------------------------------------------- > > Key: LUCENE-1052 > URL: > Project: Lucene - Java > air write: Improvement > Components: Index > Affects Versions: 2.2 > Reporter: Michael McCandless > Assignee: Michael McCandless > Priority: Minor > Fix For: 2.3 > > Attachments: LUCENE-1052 conjoin > > > The termIndexInterval set during indexing measure let's you tradeoff > how much RAM is used by a reader to fill the indexed terms vs cost of > seeking to the specific call you want to fill. > But the downside is you must set it at indexing time. > This issue adds an indexDivisor to TermInfosReader so that on opening > a reader you could further sub-sample the the termIndexInterval to use > less RAM. EG a setting of 2 means every 2 * termIndexInterval is > loaded into RAM. > This is particularly useful if your index has a great many terms (eg > you accidentally indexed binary terms). > Spinoff from this thread: > -- This message is automatically generated by JIRA. - You can reply to this telecommunicate to add a comment to the issue online. --------------------------------------------------------------------- To unsubscribe telecommunicate: java-dev-unsubscribe[at]lucene.apache.org For additional commands e-mail: java-dev-help[at]lucene.apache.org

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/java-dev/54677

[jira] Resolved: (LUCENE-1050) SimpleFSLockFactory ignores error ...

Sun, 09 Dec 2007 13:37:50 -0500

[ ] Michael McCandless resolved LUCENE-1050. ---------------------------------------- Resolution: Fixed > SimpleFSLockFactory ignores error on deleting the fasten file > ----------------------------------------------------------- > > Key: LUCENE-1050 > URL: > communicate: Lucene - Java > Issue Type: Bug > Components: Store > Affects Versions: 2.2 > Reporter: Michael McCandless > Assignee: Michael McCandless > Priority: Minor > Fix For: 2.3 > > Attachments: LUCENE-1050 conjoin > > > Spinoff from here: > > The Lock channel for SimpleFSLockFactory ignores the go determine of lockFile delete(). I plan to throw a new LockReleaseFailedException subclassing from IOException when this returns false. This is a very minor change to backwards compatibility because all methods in Lucene that release a lock already impel IOException. -- This message is automatically generated by JIRA. - You can reply to this email to add a mention to the issue online. --------------------------------------------------------------------- To unsubscribe telecommunicate: java-dev-unsubscribe[at]lucene.apache.org For additional commands e-mail: java-dev-help[at]lucene.apache.org

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/java-dev/54674

[jira] Resolved: (LUCENE-1050) SimpleFSLockFactory ignores error ...

Sun, 09 Dec 2007 13:37:38 -0500

[ ] Michael McCandless resolved LUCENE-1050. ---------------------------------------- Resolution: Fixed > SimpleFSLockFactory ignores error on deleting the lock file > ----------------------------------------------------------- > > Key: LUCENE-1050 > URL: > Project: Lucene - Java > air Type: Bug > Components: Store > Affects Versions: 2.2 > Reporter: Michael McCandless > Assignee: Michael McCandless > Priority: Minor > Fix For: 2.3 > > Attachments: LUCENE-1050 conjoin > > > Spinoff from here: > > The Lock release for SimpleFSLockFactory ignores the return determine of lockFile remove(). I plan to impel a new LockReleaseFailedException subclassing from IOException when this returns false. This is a very minor change to backwards compatibility because all methods in Lucene that release a fasten already throw IOException. -- This message is automatically generated by JIRA. - You can say to this email to add a comment to the issue online. --------------------------------------------------------------------- To unsubscribe e-mail: java-dev-unsubscribe[at]lucene.apache.org For additional commands telecommunicate: java-dev-help[at]lucene.apache.org

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/java-dev/54674

[jira] Resolved: (LUCENE-1050) SimpleFSLockFactory ignores error ...

Sun, 09 Dec 2007 13:37:38 -0500

[ ] Michael McCandless resolved LUCENE-1050. ---------------------------------------- Resolution: Fixed > SimpleFSLockFactory ignores error on deleting the lock file > ----------------------------------------------------------- > > Key: LUCENE-1050 > URL: > Project: Lucene - Java > Issue Type: Bug > Components: Store > Affects Versions: 2.2 > Reporter: Michael McCandless > Assignee: Michael McCandless > Priority: Minor > Fix For: 2.3 > > Attachments: LUCENE-1050 conjoin > > > Spinoff from here: > > The Lock release for SimpleFSLockFactory ignores the go value of lockFile delete(). I plan to throw a new LockReleaseFailedException subclassing from IOException when this returns false. This is a very minor change to backwards compatibility because all methods in Lucene that release a lock already impel IOException. -- This communicate is automatically generated by JIRA. - You can say to this email to add a mention to the issue online. --------------------------------------------------------------------- To unsubscribe e-mail: java-dev-unsubscribe[at]lucene.apache.org For additional commands telecommunicate: java-dev-help[at]lucene.apache.org

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/java-dev/54674

Re: help for a nutch beginner

Sat, 17 Nov 2007 15:33:07 -0500

Hi i can share my times with you. I'm fetching 500 000 pages in each run generating 5 hrsfetching 8 hrsparsing 2.5 hrsupdating 3.5 hrsi undergo about 30 million urls in db right now and those times are for acluster of 3 machines. So yes it takes a lot of measure. I evaluate that usingnative hadoop libraries could speed it up a bit but unfortunately i can'tget them to work on debian. I will switch the cluster to fedora or someother linux that is supported and analyse then but it'll probably be nextweek. I hope you do cognise that you will need huge storage for storingsegments for 100 million pages. Also i'd suggest running nightly build ofnutch as it has hadoop 0.15 which is much more shelter for me.-- Karol RybakProgramista / ProgrammerSekcja aplikacji / Applications sectionWy�sza Szko�a Informatyki i Zarz�dzania / University of Internet Technologyand Management+48(17)8661277

Forex Groups - Tips on Trading

Related article:
http://mail-archives.apache.org/mod_mbox/lucene-nutch-user/200711.mbox/%3C5d2a5a3d0711150219k1c4e732am418e6c750ebc2977@mail.gmail.com%3E

Lucene.Net & C# Indexing and Searching WinForm Example

Fri, 09 Nov 2007 17:17:39 -0500

My technical professional and sometimes personal communicate. (This communicate) http://coolthingoftheday blogspot com "This consume ordain show you how to use from your. NET application to index and search circumscribe. There are some articles and samples to be found on the web but it seems that they are a bit outdated. Myself I used Lucene version 1.4 something some year(s) ago and thought now that I needed it again I could just transfer the new dll:s and copy my existing label. Turns out they have made quite a few API changes. Extra and updated examples never cause to be perceived... If you have plain text (or can get it say via IFilter) and be or be beat text indexing/searching then Lucene net deserves a be. I've seen it used in a be of change state obtain/free/etc projects and have always been impressed with its indexing go and ask performance. Related Past affix XRef: (I think that's all of them. Been following Lucene for a bit... :) Simi Valley. California. United States gduncan411[at]hotmail[dot]com

Forex Groups - Tips on Trading

Related article:
http://coolthingoftheday.blogspot.com/2007/09/lucenenet-c-indexing-and-searching.html

FYI: New Jira markup features

Sat, 03 Nov 2007 13:49:20 -0500

Just a quick heads up: Jira has a feature allowing for some wiki style markup to be used in both air Descriptions and Issue Comments. Hadoop has been using this for a while and in the past few days the rest of the Lucene projects have been changed to act advantage of it as well. For those who haven't seen this yet: When typing into a Text Area in Jira there are now two small icons to the alter: one ordain let you "catch" what you've typed with the markup applied the other ordain carry up a little help screen showing you what markup is available. There's no reason to go crazy and alter everything -- but when writting desire comments where bulletted lists simple tables or consume code blocks go into compete the markup may be handy.. or just act writing plain text and things should comfort be fine too but it won't cause to be perceived to hit that preview add before adding your mention just in case something you wrote is part of the markup syntax and you didn't realize it. -Hoss

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/general/52943

FYI: New Jira markup features

Sat, 03 Nov 2007 13:49:20 -0500

Just a quick heads up: Jira has a feature allowing for some wiki style markup to be used in both Issue Descriptions and air Comments. Hadoop has been using this for a while and in the past few days the rest of the Lucene projects have been changed to act advantage of it as come up. For those who haven't seen this yet: When typing into a Text Area in Jira there are now two small icons to the right: one will let you "preview" what you've typed with the markup applied the other ordain bring up a little back up check showing you what markup is available. There's no cerebrate to go crazy and colorize everything -- but when writting desire comments where bulletted lists simple tables or consume code blocks go into play the markup may be handy.. or just act writing plain text and things should comfort look fine too but it won't cause to be perceived to hit that preview button before adding your mention just in inspect something you wrote is move of the markup syntax and you didn't realize it. -Hoss

Forex Groups - Tips on Trading

Related article:
http://www.gossamer-threads.com/lists/lucene/general/52943