manifoldcf-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erik Hatcher <erik.hatc...@gmail.com>
Subject Re: Other document data.
Date Tue, 15 Jun 2010 11:35:22 GMT
Note that lowernames=true will take care of the case change on the  
Solr side anyway.  But "literal." does need to be prepended, and  
doesn't really make sense for the SQL query to "as" rename it with  
that prefix.  LCF should prepend it automatically.

	Erik

On Jun 15, 2010, at 7:31 AM, <karl.wright@nokia.com> wrote:

> When you write the query, you have complete control over the column  
> names.
> We can certainly always attach "literal."  to the front of each  
> metadata parameter, which I'm happy to do, but mapping case etc I  
> think should be under the user's control.
>
> So, your query should look something like this:
>
> SELECT DDOCAUTHOR as ddocauthor, ... FROM ...
>
> Does this work for you?
>
> Karl
>
>
> -----Original Message-----
> From: ext Rohan.GPatil@cognizant.com [mailto:Rohan.GPatil@cognizant.com 
> ]
> Sent: Tuesday, June 15, 2010 7:05 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
>
> Hi,
>
> Yes the metadata is sent to solr. But it is being sent like this.
>
> webapp=/solr path=/update/extract  
> params 
> ={fmap.content=text&DDOCAUTHOR=sysadmin&uprefix=attr_&literal.id=/ 
> idc/groups/public/documents/sunil_layout/ 
> mfe001251.xml&lowernames=true&captureAttr=true}
>
> as you can see DDOCAUTHOR is sent in the params part which is a  
> custom metadata.
>
> But to add custom data as metadata in Solr we should pass in the  
> literal parameter along with the file.
>
> So, there will be a small change in the code.
>
> Instead of sending the column name in the params, we should send
>
> literal.<<columnname>> (columnnames in small case)
>
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com
>
>
> -----Original Message-----
> From: karl.wright@nokia.com [mailto:karl.wright@nokia.com]
> Sent: Tuesday, June 15, 2010 4:03 PM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> That's hard to determine without looking at the solr logs.  I am not  
> familiar with the log options available, but unless I'm mistaken the  
> default configuration should be dumping every request to standard out.
>
> Karl
> ________________________________________
> From: ext Rohan.GPatil@cognizant.com [Rohan.GPatil@cognizant.com]
> Sent: Tuesday, June 15, 2010 5:23 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> Hi,
>
> Yes I get it. Thanks for the clarification.
>
> I was doing the similar thing before and it used to run, now it  
> didn't. So I got confused.
>
> Is there any way to check if metadata is actually sent to solr?  
> Because I am experiencing some problem there and I don't seem to  
> figure out where it is going wrong.
>
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com
>
> -----Original Message-----
> From: karl.wright@nokia.com [mailto:karl.wright@nokia.com]
> Sent: Tuesday, June 15, 2010 2:13 PM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> LCF is an incremental crawler.  The version query is used to  
> determine whether data needs to be refetched and reindexed.  If it  
> returns the same thing each time the document is examined, the data  
> query will not be run the second time.  I therefore suggest either  
> the following:
>
> (1) Supply no version query at all.  That signals to the connector  
> that there is no version information and the data must be reindexed  
> on every job run.
> (2) Supply a version query that properly reflects changes to the  
> data.  For instance, if there's a timestamp in each record, you can  
> use that by itself ONLY if any metadata changes also are associated  
> with a change in that timestamp.  If not, you will need to glom the  
> metadata into the version string as well as the timestamp.  Is this  
> understood?
>
> If you want to FORCE a reindex, there is a link in the crawler-ui  
> for the output connection which allows you to force reindexing of  
> all data associated with that connection.
>
> If this still doesn't seem to describe what you are seeing, please  
> clarify further.
>
> Thanks,
> Karl
>
> ________________________________________
> From: ext Rohan.GPatil@cognizant.com [Rohan.GPatil@cognizant.com]
> Sent: Tuesday, June 15, 2010 12:51 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> Hi,
>
> When we specify the metadata content, It runs fine the first time,  
> The second time it doesn't run the data query at all. What must be  
> the problem ?
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com
>
>
> -----Original Message-----
> From: karl.wright@nokia.com [mailto:karl.wright@nokia.com]
> Sent: Sunday, June 13, 2010 7:04 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> No.  The data query (the same one that returns the blob info) can  
> now include additional columns.  These columns will be sent to Solr  
> as metadata fields.
>
> Karl
>
> ________________________________________
> From: ext Rohan.GPatil@cognizant.com [Rohan.GPatil@cognizant.com]
> Sent: Friday, June 11, 2010 2:28 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> Hi,
>
> I see that the issue is resolved.
>
> Now is there a new query where in we can specify the metadata fields ?
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com
>
>
> -----Original Message-----
> From: karl.wright@nokia.com [mailto:karl.wright@nokia.com]
> Sent: Thursday, June 10, 2010 4:12 PM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> It is not possible to properly glom other fields onto a BLOB unless  
> you know that the blob's contents are always encoded text.  So I  
> suggest you create a jira enhancement request in the Lucene  
> Connector Framework project to describe this enhancement (adding  
> metadata support to JDBC connector).
>
> The url is: http://issues.apache.org/jira
>
> You may need to create an account if you don't already have one.   
> Let me know if you have any difficulties.
>
> Thanks,
> Karl
>
>
> -----Original Message-----
> From: ext Rohan.GPatil@cognizant.com [mailto:Rohan.GPatil@cognizant.com 
> ]
> Sent: Thursday, June 10, 2010 6:39 AM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
>
> Hi,
>
> Using solution 1 was not a bad idea, but the problem is the content  
> is stored as BLOB in the database and gluing other fields with BLOB  
> is not possible (Is it ?) .
>
> Regarding 2 : Yes I guess I can do that modification, and anyway it  
> all depends on how we show it to the user.
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com
>
> -----Original Message-----
> From: karl.wright@nokia.com [mailto:karl.wright@nokia.com]
> Sent: Thursday, June 10, 2010 3:19 PM
> To: connectors-user@incubator.apache.org
> Subject: RE: Other document data.
>
> (1) The JDBC connector is currently relatively primitive and does  
> not have any support for "document metadata" at this time.  You can,  
> of course, glom together multiple fields into the content field with  
> it, but that's pretty crude.
> (2) The LCF convention for how to identify documents uniquely in the  
> target index is to use the URL of the document.  All documents  
> indexed with LCF have such a URL and it is likely to be both useful  
> and unique.  This url is how LCF requests deletion of the document  
> from the index, if necessary, and also overwrites the document.  So  
> it maps pretty precisely to literal.id for the basic solr setup.   
> Now, it may be that this is too tied to the example, and that the  
> solr connector should have a configuration setting to allow the name  
> of the id field used to be changed - that sounds like a reasonable  
> modification that would not be too difficult to do.  Is this  
> something you are looking for?
>
> Karl
> ________________________________________
> From: ext Rohan.GPatil@cognizant.com [Rohan.GPatil@cognizant.com]
> Sent: Thursday, June 10, 2010 4:52 AM
> To: connectors-user@incubator.apache.org
> Subject: Other document data.
>
> I am using JDBC connection to search for the documents in the  
> database.
>
> The issue is  some document data(Check in date etc) is present in  
> the other columns. How to send this data to Solr so as to index it.
>
> Why is the URL of the file taken as ID in Solr.
>
> Thanks & Regards,
> Rohan G Patil
> Cognizant  Programmer Analyst Trainee,Bangalore || Mob # +91  
> 9535577001
> Rohan.GPatil@cognizant.com<mailto:Rohan.GPatil@cognizant.com>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>
>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>
>
> This e-mail and any files transmitted with it are for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information.
> If you are not the intended recipient, please contact the sender by
> reply e-mail and destroy all copies of the original message.
> Any unauthorized review, use, disclosure, dissemination, forwarding,
> printing or copying of this email or any action taken in reliance on  
> this
> e-mail is strictly prohibited and may be unlawful.
>
>


Mime
View raw message