Does AEM Query builder API allow searching the Content Fragment Core component's content included within a Page?
Sorry for a typo in my previous post. Here is the query, that works on my test env and returns a page where a Content Fragment includes a word "winter":
path=/content/we-retail/us
type= cq:Page
group.p.and
group.1_fulltext=winter
group.2_fulltext=contentfragment
berliant Thanks again! But the updated query provided also does not seem to work (no search results yet).
I have authored a test page under "/content/we-retail/us" containing Core Content Fragment component which includes a word "career". Are you using a proxied Core "Content Fragment" component or out of the box (General) "Content Fragment" component?
Please note that the search works appropriately with OOB (General) Content Fragment already. My question is specific to the latest Core Content Fragment Component.
I am currently using AEM 6.3.3.0 with AEM Core Components v2.5.0 installed. Could you please let me know which version of AEM Core components has been installed in your test environment and the AEM version being used.
My test was done on the same version.
As you have "contentfragment" in JCR I don't think it matters what component you are extending.
It searches on the keyword and the fact that the page has CF.
berliant Really appreciate your prompt response.
Sorry, I forgot to mention that I have been trying to search for "Structured Content Fragments" created using custom Content Fragment Model. The search had been working appropriately for a simple content fragment but does not seem to work with the structured content fragments.
Find below the structured content fragment included on a test page. The main difference I could see is that the "text" property ( which is created automatically for simple fragments) containing the fragment data/text is not being created within component node as multiple model elements are being referenced in this case under "master" variation. The search may not be working as the text is not attached to the page node and is just being referenced through "fragmentPath" property.
Find below the structured content fragment created with a new Content Fragment Model containing "name", "age", "sport", "about" elements.
Please review and advise if Query builder API supports searching the structured content fragments as well or just the simple content fragments.
You need to index your custom properties: "about", "name", "age" etc.
The query that I provided executes the plan:
PLAN: [cq:Page] as [a] /* lucene:cqPageLucene(/oak:index/cqPageLucene) +(+:fulltext:winter +:fulltext:contentfragment) +:ancestors:/content/we-retail/us ft:("winter" "contentfragment") where (contains([a].[*], 'winter')) and (contains([a].[*], 'contentfragment')) and (isdescendantnode([a], [/content/we-retail/us])) */
It means that you need to edit cqPageLucene index by adding a dedicated property. For example, for "about" it should be something like:
Make sure that you re-index /oak:index/cqPageLucene prior of testing the query.
berliant Thanks again for your response.
I have added the "about" property as specified under "cqPageLucene" and also re-indexed the "/oak:index/cqPageLucene" index. However, the debugger does not return any results yet. I have also tried restarting the AEM instance and re-indexing again.
Could you please review and let me know if any additional configurations are needed.
Find below the "Query Explanation" details.
Indexes Used
cqPageLucene(/oak:index/cqPageLucene)
Execution Time
Total time: 2 ms
Query execution time: 1 ms
Get nodes time: 1 ms
Execution Plan
[cq:Page] as [a] /* lucene:cqPageLucene(/oak:index/cqPageLucene) +(+:fulltext:career +:fulltext:contentfragment) +:ancestors:/content/we-retail/us ft:("career" "contentfragment") where (contains([a].[*], 'career')) and (contains([a].[*], 'contentfragment')) and (isdescendantnode([a], [/content/we-retail/us])) */
Logs
Parsing xpath statement: explain /jcr:root/content/we-retail/us//element(*, cq:Page)[(jcr:contains(., 'career') and jcr:contains(., 'contentfragment'))]
XPath > SQL2: explain select [jcr:path], [jcr:score], * from [cq:Page] as a where contains(*, 'career') and contains(*, 'contentfragment') and isdescendantnode(a, '/content/we-retail/us') /* xpath: /jcr:root/content/we-retail/us//element(*, cq:Page)[(jcr:contains(., 'career') and jcr:contains(., 'contentfragment'))] */
Literal used
Attempting optimisation
cost using filter Filter(query=explain select [jcr:path], [jcr:score], * from [cq:Page] as a where contains(*, 'career') and contains(*, 'contentfragment') and isdescendantnode(a, '/content/we-retail/us') /* xpath: /jcr:root/content/we-retail/us//element(*, cq:Page)[(jcr:contains(., 'career') and jcr:contains(., 'contentfragment'))] */ fullText="career" "contentfragment", path=/content/we-retail/us//*)
cost for reference is Infinity
cost for property is Infinity
cost for nodeType is Infinity
Evaluating plan with index definition Lucene Index : /oak:index/slingeventJob
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/versionStoreIndex
Evaluating plan with index definition Lucene Index : /oak:index/cqTagLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/workflowDataLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/socialLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /content/oak:index/enablementResourceName
Evaluating plan with index definition Lucene Index : /oak:index/authorizables
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/ntBaseLucene
Evaluating plan with index definition Lucene Index : /oak:index/lucene-spellcheck
Applicable IndexingRule found IndexRule: nt:base
Evaluating plan with index definition Lucene Index : /oak:index/lucene
Applicable IndexingRule found IndexRule: nt:base
Evaluating plan with index definition Lucene Index : /oak:index/commerceLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/cqProjectLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/cqMobileAppLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/damAssetLucene
No applicable IndexingRule found for any of the superTypes [nt:hierarchyNode, cq:Page, nt:base, mix:created]
Evaluating plan with index definition Lucene Index : /oak:index/cqPageLucene
Applicable IndexingRule found IndexRule: cq:Page
cost for lucene-property[/oak:index/cqPageLucene] is 8608.0
Full-text index without plan: "career" "contentfragment"
cost for aggregate lucene is Infinity
looking for plans for paths : []
cost for solr is Infinity
cost for traverse is Infinity
No alternatives found. Query: select [a].[jcr:path] as [jcr:path], [a].[jcr:score] as [jcr:score], [a].[jcr:primaryType] as [jcr:primaryType], [a].[jcr:createdBy] as [jcr:createdBy], [a].[jcr:created] as [jcr:created] from [cq:Page] as [a] where (contains([a].[*], 'career')) and (contains([a].[*], 'contentfragment')) and (isdescendantnode([a], [/content/we-retail/us]))
I don't see the "Attach" option to upload the package even in the advanced editor. Is there any other option to share the package?
Assuming that we have a custom index that supports CF model's property, we need a union query similar to:
path=/content
1_group.1_type=nt:unstructured
1_group.1_property=name
1_group.1_property.value=Jason Chester
group.p.and
group.2_type= cq:Page
group.2_fulltext=contentfragment
Unfortunately, here it reaches a limitation of AEM 6.3 oak engine for union queries support.
If a search of CF with Multiple Elements is a firm business requirement, you can explicitly add all required element:
with that configuration you can use the query, that I suggested earlier:
path=/content/we-retail/us
type= cq:Page
group.p.and
group.1_fulltext=winter
group.2_fulltext=contentfragment
berliant Thank you!
I tried adding all the required elements within CF as recommended. But it did not resolve the search issue even with the custom index.
The business requirement is to implement search capability for multiple elements of structured content fragments included on a page. We are trying to replace existing components with core content fragment components to enable content reuse across different projects/channels. But after implementing the change in a test environment, the search does not seem to work and no longer returning appropriate search results as the CF data is not available on the page.
The search does not seem to work with the Core Content Fragment component as it does not automatically add the "text" property containing selected Structured Content Fragment elements data to the component node unlike the OOB Content Fragment Component which adds the "text" property. Search works appropriately with the OOB Content Fragment Component as the CF data is available within Page structure under respective component.
Find below the "Core Content Fragment Component" node data for a test page without the "text" property.
Find below the "OOB Content Fragment Component" node data containing "text" property for the same test page.
Could you please let me know how the automatic addition of "text" property works for OOB component and if the same capability may be activated for the Core component to enable search functionality.
I used your package for the test.
1. I modified the Test Page by editing the CF component. My changes implied adding two components:
2. It means that now the node structure includes two additional properties: elementNames and text:
3. Now we can use full-text search capabilities and jcr:contains function:
path=/content/we-retail/us
type= cq:Page
group.p.and
group.1_fulltext=Jason
group.2_fulltext=contentfragment
Note, that the query is using oak:index/cqPageLucene
For a full-text search, you don't need to edit indexes.
Make sure, that you have the same configuration and the query execution plan is also similar to what I see on my test machine. If this is not a case, it might be some bigger issue related to an oak/indexing on your AEM.
At that point, I suggest opening a support case and have a screen session with a support engineer.
Community
- Community home
- Guidelines
- Community Advisors
- Advertising Cloud
- Analytics
- Audience Manager
- Campaign Classic v7 & Campaign v8
- Campaign Standard
- Developer Cloud Extensibility
- Experience Cloud
- Experience Manager Sites & More
- Experience Platform
- Journey Optimizer
- Target
- Creative Cloud
- Document Cloud
- Commerce
- Marketo Engage
- Workfront
Resources
Adobe account
Change language
Copyright © 2022 Adobe. All rights reserved.
