From jporter at redhat.com  Tue Sep 23 11:55:53 2014
From: jporter at redhat.com (Jason Porter)
Date: Tue, 23 Sep 2014 11:55:53 -0400 (EDT)
Subject: [searchisko-dev] Reusing an indexer from a river
In-Reply-To: <2118946081.28108514.1411487706270.JavaMail.zimbra@redhat.com>
Message-ID: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>

Hey all, we're looking at a better way for indexing articles and solutions from access.redhat.com. I'm wondering if we can reuse the river index for this, or if we need to create a new one. If we can copy and paste (and slightly modify) the current index that works too. Thanks!

From velias at redhat.com  Thu Sep 25 08:02:05 2014
From: velias at redhat.com (Vlastimil Elias)
Date: Thu, 25 Sep 2014 14:02:05 +0200
Subject: [searchisko-dev] Reusing an indexer from a river
In-Reply-To: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>
References: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>
Message-ID: <5424043D.90109@redhat.com>

Hi Jason,

if I understand correctly, you can stop to use river running inside DCP
for indexing articles and solutions from access.redhat.com, and start to
push/delete this content directly from your side/build over REST API.
 
In this case you have to move content definition for these types from
provider called 'rht' under you provider (jboss-developer). As we use
naming conventions for ES index names it also means you have to create
new indexes, but you are right that you can simply take current
definitions and modify them.
During deployment of your new definitions we also have to remove old
'rht' provider and both rivers related to it.

Cheers

Vlastimil

On 23.9.2014 17:55, Jason Porter wrote:
> Hey all, we're looking at a better way for indexing articles and solutions from access.redhat.com. I'm wondering if we can reuse the river index for this, or if we need to create a new one. If we can copy and paste (and slightly modify) the current index that works too. Thanks!
> _______________________________________________
> searchisko-dev mailing list
> searchisko-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/searchisko-dev

-- 
Vlastimil Elias
Principal Software Engineer
jboss.org Development Team




From paul.robinson at redhat.com  Thu Sep 25 08:06:08 2014
From: paul.robinson at redhat.com (Paul Robinson)
Date: Thu, 25 Sep 2014 13:06:08 +0100
Subject: [searchisko-dev] Reusing an indexer from a river
In-Reply-To: <5424043D.90109@redhat.com>
References: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>
	<5424043D.90109@redhat.com>
Message-ID: <F9A11DEE-C687-4971-80F2-D8E19F7BB362@redhat.com>

Is there any code re-use that Jason can do here? 

On 25 Sep 2014, at 13:02, Vlastimil Elias <velias at redhat.com> wrote:

> Hi Jason,
> 
> if I understand correctly, you can stop to use river running inside DCP
> for indexing articles and solutions from access.redhat.com, and start to
> push/delete this content directly from your side/build over REST API.
> 
> In this case you have to move content definition for these types from
> provider called 'rht' under you provider (jboss-developer). As we use
> naming conventions for ES index names it also means you have to create
> new indexes, but you are right that you can simply take current
> definitions and modify them.
> During deployment of your new definitions we also have to remove old
> 'rht' provider and both rivers related to it.
> 
> Cheers
> 
> Vlastimil
> 
> On 23.9.2014 17:55, Jason Porter wrote:
>> Hey all, we're looking at a better way for indexing articles and solutions from access.redhat.com. I'm wondering if we can reuse the river index for this, or if we need to create a new one. If we can copy and paste (and slightly modify) the current index that works too. Thanks!
>> _______________________________________________
>> searchisko-dev mailing list
>> searchisko-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/searchisko-dev
> 
> -- 
> Vlastimil Elias
> Principal Software Engineer
> jboss.org Development Team
> 
> 


Paul.

-- 
Paul Robinson
JBoss Developer Team Lead (www.jboss.org)

JBoss, a Division of Red Hat
Registered in England and Wales under Company Registration No. 03798903
Directors:Michael Cunningham (US), Charles Peters (US), Matt Parson (US), Paul Hickey (Ireland)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/searchisko-dev/attachments/20140925/ab75a5d6/attachment-0001.html 

From velias at redhat.com  Thu Sep 25 08:09:13 2014
From: velias at redhat.com (Vlastimil Elias)
Date: Thu, 25 Sep 2014 14:09:13 +0200
Subject: [searchisko-dev] Reusing an indexer from a river
In-Reply-To: <5E119274-9BC8-4C7A-AB47-32D46B344714@redhat.com>
References: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>
	<5424043D.90109@redhat.com>
	<5E119274-9BC8-4C7A-AB47-32D46B344714@redhat.com>
Message-ID: <542405E9.4070300@redhat.com>

What do you mean by code reuse? ;-)

And BTW why do you want to abandon use of the rivers?
These rivers are very flexible, maybe we can only configure them better
to do what you need, without any coding necessary.

Vl.


On 25.9.2014 14:04, Paul Robinson wrote:
> Is there any code re-use that Jason can do here? 
>
> On 25 Sep 2014, at 13:02, Vlastimil Elias <velias at redhat.com
> <mailto:velias at redhat.com>> wrote:
>
>> Hi Jason,
>>
>> if I understand correctly, you can stop to use river running inside DCP
>> for indexing articles and solutions from access.redhat.com
>> <http://access.redhat.com>, and start to
>> push/delete this content directly from your side/build over REST API.
>>
>> In this case you have to move content definition for these types from
>> provider called 'rht' under you provider (jboss-developer). As we use
>> naming conventions for ES index names it also means you have to create
>> new indexes, but you are right that you can simply take current
>> definitions and modify them.
>> During deployment of your new definitions we also have to remove old
>> 'rht' provider and both rivers related to it.
>>
>> Cheers
>>
>> Vlastimil
>>
>> On 23.9.2014 17:55, Jason Porter wrote:
>>> Hey all, we're looking at a better way for indexing articles and
>>> solutions from access.redhat.com <http://access.redhat.com>. I'm
>>> wondering if we can reuse the river index for this, or if we need to
>>> create a new one. If we can copy and paste (and slightly modify) the
>>> current index that works too. Thanks!
>>> _______________________________________________
>>> searchisko-dev mailing list
>>> searchisko-dev at lists.jboss.org <mailto:searchisko-dev at lists.jboss.org>
>>> https://lists.jboss.org/mailman/listinfo/searchisko-dev
>>
>> -- 
>> Vlastimil Elias
>> Principal Software Engineer
>> jboss.org <http://jboss.org> Development Team
>>
>>
>
>
> Paul.
>
> -- 
> Paul Robinson
> JBoss Developer Team Lead (www.jboss.org <http://www.jboss.org>)
>
> JBoss, a Division of Red Hat
> Registered in England and Wales under Company Registration No. 03798903
> Directors:Michael Cunningham (US), Charles Peters (US), Matt Parson
> (US), Paul Hickey (Ireland)
>

-- 
Vlastimil Elias
Principal Software Engineer
jboss.org Development Team

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/searchisko-dev/attachments/20140925/928634c3/attachment.html 

From jporter at redhat.com  Thu Sep 25 19:37:07 2014
From: jporter at redhat.com (Jason Porter)
Date: Thu, 25 Sep 2014 19:37:07 -0400 (EDT)
Subject: [searchisko-dev] Reusing an indexer from a river
In-Reply-To: <5424043D.90109@redhat.com>
References: <856689674.28109214.1411487753030.JavaMail.zimbra@redhat.com>
	<5424043D.90109@redhat.com>
Message-ID: <980834955.29991449.1411688227890.JavaMail.zimbra@redhat.com>

Paul and I were talking about this earlier today, is there a reason we couldn't use the strata search to find all articles and solutions for a product and fetch those returned URLs? If we were to do this, we'd cut out all the middle work and go straight from one system to the next. What are your thoughts?

----- Original Message -----
> From: "Vlastimil Elias" <velias at redhat.com>
> To: searchisko-dev at lists.jboss.org, "Paul Robinson" <paul.robinson at redhat.com>
> Sent: Thursday, September 25, 2014 6:02:05 AM
> Subject: Re: [searchisko-dev] Reusing an indexer from a river
> 
> Hi Jason,
> 
> if I understand correctly, you can stop to use river running inside DCP
> for indexing articles and solutions from access.redhat.com, and start to
> push/delete this content directly from your side/build over REST API.
>  
> In this case you have to move content definition for these types from
> provider called 'rht' under you provider (jboss-developer). As we use
> naming conventions for ES index names it also means you have to create
> new indexes, but you are right that you can simply take current
> definitions and modify them.
> During deployment of your new definitions we also have to remove old
> 'rht' provider and both rivers related to it.
> 
> Cheers
> 
> Vlastimil
> 
> On 23.9.2014 17:55, Jason Porter wrote:
> > Hey all, we're looking at a better way for indexing articles and solutions
> > from access.redhat.com. I'm wondering if we can reuse the river index for
> > this, or if we need to create a new one. If we can copy and paste (and
> > slightly modify) the current index that works too. Thanks!
> > _______________________________________________
> > searchisko-dev mailing list
> > searchisko-dev at lists.jboss.org
> > https://lists.jboss.org/mailman/listinfo/searchisko-dev
> 
> --
> Vlastimil Elias
> Principal Software Engineer
> jboss.org Development Team
> 
> 
> 
> _______________________________________________
> searchisko-dev mailing list
> searchisko-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/searchisko-dev
>