diff --git a/solr/solr-ref-guide/src/blockjoin-faceting.adoc b/solr/solr-ref-guide/src/blockjoin-faceting.adoc index 18a74084f38d..94b51108f4de 100644 --- a/solr/solr-ref-guide/src/blockjoin-faceting.adoc +++ b/solr/solr-ref-guide/src/blockjoin-faceting.adoc @@ -41,7 +41,7 @@ This example shows how you could add this search components to `solrconfig.xml` This component can be added into any search request handler. This component work with distributed search in SolrCloud mode. -Documents should be added in children-parent blocks as described in <>. Examples: +Documents should be added in children-parent blocks as described in <>. Examples: .Sample document [source,xml] diff --git a/solr/solr-ref-guide/src/indexing-and-basic-data-operations.adoc b/solr/solr-ref-guide/src/indexing-and-basic-data-operations.adoc index 40b5f3f57825..e145a71986f0 100644 --- a/solr/solr-ref-guide/src/indexing-and-basic-data-operations.adoc +++ b/solr/solr-ref-guide/src/indexing-and-basic-data-operations.adoc @@ -1,5 +1,5 @@ = Indexing and Basic Data Operations -:page-children: introduction-to-solr-indexing, post-tool, uploading-data-with-index-handlers, uploading-data-with-solr-cell-using-apache-tika, uploading-structured-data-store-data-with-the-data-import-handler, updating-parts-of-documents, detecting-languages-during-indexing, de-duplication, content-streams +:page-children: introduction-to-solr-indexing, post-tool, uploading-data-with-index-handlers, indexing-nested-documents, uploading-data-with-solr-cell-using-apache-tika, uploading-structured-data-store-data-with-the-data-import-handler, updating-parts-of-documents, detecting-languages-during-indexing, de-duplication, content-streams // Licensed to the Apache Software Foundation (ASF) under one // or more contributor license agreements. See the NOTICE file // distributed with this work for additional information @@ -27,6 +27,8 @@ This section describes how Solr adds data to its index. It covers the following * *<>*: Index any JSON of your choice +* *<>*: Detailed information about indexing and schema configuration for nested documents. + * *<>*: Information about using the Solr Cell framework to upload data for indexing. * *<>*: Information about uploading and indexing data from a structured data store. diff --git a/solr/solr-ref-guide/src/indexing-nested-documents.adoc b/solr/solr-ref-guide/src/indexing-nested-documents.adoc new file mode 100644 index 000000000000..ce07a3126356 --- /dev/null +++ b/solr/solr-ref-guide/src/indexing-nested-documents.adoc @@ -0,0 +1,152 @@ += Indexing Nested Child Documents +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +Solr supports indexing nested documents for creating stronger bonds and relationships between documents, +to be used for updates and <>. + +Nested documents in Solr can be used to bind a blog post parent document and comments as child documents +-- or products as parent documents and sizes, colors, or other variations as child documents. + +The parent with all children is referred to as a "block" and it explains some of the nomenclature of related features. +At query time, the <> can search these relationships, + and the `<>` Document Transformer can attach child documents to the result documents. +In terms of performance, indexing the relationships between documents usually yields much faster queries than an equivalent "query time join", + since the relationships are already stored in the index and do not need to be computed. +However, nested documents are less flexible than query time joins as it imposes rules that some applications may not be able to accept. +Nested documents may be indexed via either the XML or JSON data syntax, and is also supported by <> with javabin. + +[NOTE] +==== +A big limitation is that the whole block of parent-children documents must be updated or deleted together, not separately. +In other words, even if a single child document or the parent document is changed, the whole block of parent-child documents must be indexed together. +_Solr does not enforce this rule_; if it's violated, you may get sporadic query failures or incorrect results. +==== + +== Schema Configuration + + * The schema must include indexed field `\_root_`. The value of that field is populated automatically and is the same for all documents in the block, regardless of the inheritance depth. The ID of the top document in every nested hierarchy is populated in this field. + + `` + * `\_nest_path_` is used to store the path of the document in the hierarchy. This field is optional. + + ` + ` + * `\_nest_parent_` is used to store the `ID` of the parent in the previous level. This field is optional. + + `` + * Nested documents are very much documents in their own right even if certain nested documents hold different information from the parent. + Therefore: + ** the schema must be able to represent the fields of any document + ** it may be infeasible to use `required` + ** even child documents need a unique `ID` + * If you associate a child document as a field (e.g., comment), that field need not be defined in the schema, and probably + shouldn't be as it would be confusing. There is no child document field type. + +== Rudimentary Root-only schemas + + * These schemas do not contain any other nested related fields apart from `\_root_`. + + In this mode relationship types(field names) between parents and their children are not saved. + + In this case <> transformer returns all children under the `\_childDocuments_` field. + * Typically you should have a field that differentiates a root doc from any nested children. However this isn't strictly necessary; so long as it's possible to write a query that can select only root documents somehow. Such a query is needed for the <> and <> doc transformer to function. + +=== XML Examples + +For example, here are two documents and their child documents. +It illustrates two styles of adding child documents; the first is associated via a field "comment" (preferred), +and the second is done in the classic way now referred to as an "anonymous" or "unlabelled" child document. +This field label relationship is available to the URP chain in Solr but is ultimately discarded. +Solr 8 will save the relationship. + +[source,xml] +---- + + + 1 + Solr adds block join support + parentDocument + + + 2 + SolrCloud supports it too! + + + + + 3 + New Lucene and Solr release is out + parentDocument + + 4 + Lots of new features + + + +---- + +In this example, we have indexed the parent documents with the field `content_type`, which has the value "parentDocument". +We could have also used a boolean field, such as `isParent`, with a value of "true", or any other similar approach. + +=== JSON Examples + +This example is equivalent to the XML example above. +Again, the field labelled relationship is preferred. +The labelled relationship here is one child document but could have been wrapped in array brackets. +For the anonymous relationship, note the special `\_childDocuments_` key whose contents must be an array of child documents. + +[source,json] +---- +[ + { + "ID": "1", + "title": "Solr adds block join support", + "content_type": "parentDocument", + "comments": [{ + "ID": "2", + "content": "SolrCloud supports it too!" + }, + { + "ID": "3", + "content": "New filter syntax" + } + ] + }, + { + "ID": "4", + "title": "New Lucene and Solr release is out", + "content_type": "parentDocument", + "_childDocuments_": [ + { + "ID": "5", + "comments": "Lots of new features" + } + ] + } +] +---- + +.Root-Only Mode +[NOTE] + In Root-only schemas, these two documents will result in the same docs being indexed (Root-only schemas do not honor nested relationships). + When queried, child docs will be appended to _childDocuments_ key. + +== Updating Nested Documents + +Currently Solr supports updating whole hierarchies using atomic updates. Documents should be updated by the Root (top) +document's ID, and the update should contain all its children. This is needed considering Solr deletes the old hierarchy, +since the update term is `\_root_:id`. In case some child documents are omitted from the update command, +said documents will be deleted from the index. + +.Updating By a Child Document's ID +[NOTE] + An update by ID to a child document will index a new document with the same ID as the one in the nested hierarchy, + yet the new document will not be indexed as a child, but rather as a new document outside of the nested hierarchy. \ No newline at end of file diff --git a/solr/solr-ref-guide/src/json-facet-api.adoc b/solr/solr-ref-guide/src/json-facet-api.adoc index fbccf9112a88..61b96f4cce67 100644 --- a/solr/solr-ref-guide/src/json-facet-api.adoc +++ b/solr/solr-ref-guide/src/json-facet-api.adoc @@ -757,7 +757,7 @@ Most stat facet functions (`avg`, `sumsq`, etc.) allow users to perform math com === uniqueBlock() and Block Join Counts -When a collection contains <>, the `blockChildren` and `blockParent` <> can be useful when searching for parent documents and you want to compute stats against all of the affected children documents (or vice versa). +When a collection contains <>, the `blockChildren` and `blockParent` <> can be useful when searching for parent documents and you want to compute stats against all of the affected children documents (or vice versa). But if you only need to know the _count_ of all the blocks that exist in the current domain, a more efficient option is the `uniqueBlock()` aggregate function. Suppose we have products with multiple SKUs, and we want to count products for each color. diff --git a/solr/solr-ref-guide/src/json-faceting-domain-changes.adoc b/solr/solr-ref-guide/src/json-faceting-domain-changes.adoc index 6f57a695a5e9..60a842ed8c7e 100644 --- a/solr/solr-ref-guide/src/json-faceting-domain-changes.adoc +++ b/solr/solr-ref-guide/src/json-faceting-domain-changes.adoc @@ -177,7 +177,7 @@ NOTE: While a `query` domain can be combined with an additional domain `filter`, == Block Join Domain Changes -When a collection contains <>, the `blockChildren` or `blockParent` domain options can be used transform an existing domain containing one type of document, into a domain containing the documents with the specified relationship (child or parent of) to the documents from the original domain. +When a collection contains <>, the `blockChildren` or `blockParent` domain options can be used transform an existing domain containing one type of document, into a domain containing the documents with the specified relationship (child or parent of) to the documents from the original domain. Both of these options work similarly to the corresponding <> by taking in a single String query that exclusively matches all parent documents in the collection. If `blockParent` is used, then the resulting domain will contain all parent documents of the children from the original domain. If `blockChildren` is used, then the resulting domain will contain all child documents of the parents from the original domain. diff --git a/solr/solr-ref-guide/src/other-parsers.adoc b/solr/solr-ref-guide/src/other-parsers.adoc index f22dcdc0599e..266080c8087a 100644 --- a/solr/solr-ref-guide/src/other-parsers.adoc +++ b/solr/solr-ref-guide/src/other-parsers.adoc @@ -24,7 +24,7 @@ Many of these parsers are expressed the same way as <>. +There are two query parsers that support block joins. These parsers allow indexing and searching for relational content that has been <>. The example usage of the query parsers below assumes these two documents and each of their child documents have been indexed: diff --git a/solr/solr-ref-guide/src/searching-nested-documents.adoc b/solr/solr-ref-guide/src/searching-nested-documents.adoc new file mode 100644 index 000000000000..27e394a07f79 --- /dev/null +++ b/solr/solr-ref-guide/src/searching-nested-documents.adoc @@ -0,0 +1,202 @@ += Searching Nested Child Documents +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +This section exposes potential techniques which can be used for searching deeply nested documents, +show casing how more complex queries can be constructed using some of Solr's query parsers and Doc Transformers. +These features require `\_root_`, `\_nest_path_` to be declared in the index's schema. + +Please refer to the <> +section for more details about schema and index configuration. + + +[NOTE] +This section does not show case faceting on nested documents. For nested document faceting, please refer to the +<> section. + +== Query Examples + +For the upcoming examples, assume the following documents have been indexed: + +[source,json] +---- +[ + { + "ID": "1", + "title": "Cooking Recommendations", + "tags": ["cooking", "meetup"], + "posts": [{ + "ID": "2", + "title": "Cookies", + "comments": [{ + "ID": "3", + "content": "Lovely recipe" + }, + { + "ID": "4", + "content": "A-" + } + ] + }, + { + "ID": "5", + "title": "Cakes" + } + ] + }, + { + "ID": "6", + "title": "For Hire", + "tags": ["professional", "jobs"], + "posts": [{ + "ID": "7", + "title": "Search Engineer", + "comments": [{ + "ID": "8", + "content": "I am interested" + }, + { + "ID": "9", + "content": "How large is the team?" + } + ] + }, + { + "ID": "10", + "title": "Low level Engineer" + } + ] + } +] +---- + +=== Child Doc Transformer + +Can be used enrich query results with the documents' descendants. + +For a detailed explanation of this transformer, see the section <>. + +For example, let us examine this query: +`q=ID:1, +fl=ID,[child childFilter=/comments/content:recipe]`. + +The Child Doc Transformer can be used to enrich matching docs with comments that match a particular filter. + +In this particular query, the child Filter will only match the first comment of doc(ID:1), +therefore only that particular comment will be appended to the result. + +[source,json] +---- + { "response":{"numFound":1,"start":0,"docs":[ + { + "ID": "1", + "title": "Cooking Recommendations", + "tags": ["cooking", "meetup"], + "posts": [{ + "ID": "2", + "title": "Cookies", + "comments": [{ + "ID": "3", + "content": "Lovely recipe" + }] + }] + }] + } + } +---- + +=== Children Query Parser + +Can be used to retrieve children of a matching document. + +For a detailed explanation of this parser, see the section <>. + +For example, let us examine this query: +`q={!child of='_nest_path_:/posts}content:"Search Engineer"`. + +The `'of'` filter returns all posts. This is used to filter out all documents in a particular path of the hierarchy(all parents). +The second part of the query is a filter for some parents, which we wish to return their children. + +In this example, all comments of posts which had "Search Engineer in their `content` field will be returned. + +[source,json] +---- + { "response":{"numFound":2,"start":0,"docs":[ + { + "ID": "8", + "content": "I am interested" + }, + { + "ID": "9", + "content": "How large is the team?" + } + ]} + } +---- + +=== Parents Query Parser + +Can be used to retrieve parents of a child document. + +For a detailed explanation of this parser, see the section <>. + +For example, let us examine this query: +`q={!parent which='-_nest_path_:* \*:*'}title:"Search Engineer"`. + +The `'which'` filter returns all root documents. +The second part of this query is a filter to match some child documents. +This query returns the parent at the root(since all parents filter returns root documents) of each +matching child document. In this case, all child documents which had `Search Engineer` in their `title` field. + +[source,json] +---- + { "response":{"numFound":1,"start":0,"docs":[{ + "ID": "6", + "title": "For Hire", + "tags": ["professional", "jobs"] + } + ]} + } +---- + +=== Combining Block Join Query Parsers with Child Doc Transformer + +The combination of these two features enable seamless creation of powerful queries. + +For example, querying posts which are under a page tagged as a job, contain the words "Search Engineer". +The comments for matching posts can also be fetched, all done in a single Solr Query. + +For example, let us examine this query: +`q=+{!child of='-\_nest_path_:* \*:*'}+tags:"jobs" &fl=*,[child] +&fq=\_nest_path_:/posts`. + +This query returns all posts and their comments, which had "Search Engineer" in their title, +and are indexed under a page tagged with "jobs". +The comments are appended to the matching posts, since the ChildDocTransformer is specified under the `fl` parameter. + +[source,json] +---- + { "response":{"numFound":1,"start":0,"docs":[ + { + "ID": "7", + "title": "Search Engineer", + "comments": [{ + "ID": "8", + "content": "I am interested" + }, + { + "ID": "9", + "content": "How large is the team?" + } + ] + }, + { + "ID": "10", + "title": "Low level Engineer" + }] + } + } +---- diff --git a/solr/solr-ref-guide/src/searching.adoc b/solr/solr-ref-guide/src/searching.adoc index 9fa0e57cae84..8367105a14b7 100644 --- a/solr/solr-ref-guide/src/searching.adoc +++ b/solr/solr-ref-guide/src/searching.adoc @@ -10,6 +10,7 @@ spell-checking, + query-re-ranking, + transforming-result-documents, + + searching-nested-documents, + suggester, + morelikethis, + pagination-of-results, + @@ -68,6 +69,7 @@ This section describes how Solr works with search requests. It covers the follow * <>: Detailed information about re-ranking top scoring documents from simple queries using more complex scores. ** <>: How to use LTR to run machine learned ranking models in Solr. +* <>: Detailed information about constructing nested and hierarchical queries. * <>: Detailed information about using `DocTransformers` to add computed information to individual documents * <>: Detailed information about Solr's powerful autosuggest component. * <>: Detailed information about Solr's similar results query component. diff --git a/solr/solr-ref-guide/src/transforming-and-indexing-custom-json.adoc b/solr/solr-ref-guide/src/transforming-and-indexing-custom-json.adoc index 26bd60bd8458..c2fd06ece877 100644 --- a/solr/solr-ref-guide/src/transforming-and-indexing-custom-json.adoc +++ b/solr/solr-ref-guide/src/transforming-and-indexing-custom-json.adoc @@ -777,106 +777,6 @@ curl 'http://localhost:8983/api/collections/techproducts/update/json' -H 'Conten ==== -- -== Indexing Nested Documents - -The following is an example of indexing nested documents: - -[.dynamic-tabs] --- -[example.tab-pane#v1nested] -==== -[.tab-label]*V1 API* -[source,bash] ----- -curl 'http://localhost:8983/solr/techproducts/update/json/docs?split=/|/orgs'\ - -H 'Content-type:application/json' -d '{ - "name": "Joe Smith", - "phone": 876876687, - "orgs": [ - { - "name": "Microsoft", - "city": "Seattle", - "zip": 98052 - }, - { - "name": "Apple", - "city": "Cupertino", - "zip": 95014 - } - ] -}' ----- -==== - -[example.tab-pane#v2nested] -==== -[.tab-label]*V2 API Standalone Solr* -[source,bash] ----- -curl 'http://localhost:8983/api/cores/techproducts/update/json?split=/|/orgs'\ - -H 'Content-type:application/json' -d '{ - "name": "Joe Smith", - "phone": 876876687, - "orgs": [ - { - "name": "Microsoft", - "city": "Seattle", - "zip": 98052 - }, - { - "name": "Apple", - "city": "Cupertino", - "zip": 95014 - } - ] -}' ----- -==== - -[example.tab-pane#v2nestedcloud] -==== -[.tab-label]*V2 API SolrCloud* -[source,bash] ----- -curl 'http://localhost:8983/api/collections/techproducts/update/json?split=/|/orgs'\ - -H 'Content-type:application/json' -d '{ - "name": "Joe Smith", - "phone": 876876687, - "orgs": [ - { - "name": "Microsoft", - "city": "Seattle", - "zip": 98052 - }, - { - "name": "Apple", - "city": "Cupertino", - "zip": 95014 - } - ] -}' ----- -==== --- - -With this example, the documents indexed would be, as follows: - -[source,json] ----- -{ - "name":"Joe Smith", - "phone":876876687, - "_childDocuments_":[ - { - "name":"Microsoft", - "city":"Seattle", - "zip":98052}, - { - "name":"Apple", - "city":"Cupertino", - "zip":95014}]} ----- - == Tips for Custom JSON Indexing . Schemaless mode: This handles field creation automatically. The field guessing may not be exactly as you expect, but it works. The best thing to do is to setup a local server in schemaless mode, index a few sample docs and create those fields in your real setup with proper field types before indexing diff --git a/solr/solr-ref-guide/src/transforming-result-documents.adoc b/solr/solr-ref-guide/src/transforming-result-documents.adoc index 97917b48dbd4..49d91f808f6c 100644 --- a/solr/solr-ref-guide/src/transforming-result-documents.adoc +++ b/solr/solr-ref-guide/src/transforming-result-documents.adoc @@ -124,7 +124,7 @@ A default style can be configured by specifying an `args` parameter in your `sol === [child] - ChildDocTransformerFactory -This transformer returns all <> of each parent document matching your query in a flat list nested inside the matching parent document. This is useful when you have indexed nested child documents and want to retrieve the child documents for the relevant parent documents for any type of search query. +This transformer returns all <> of each parent document matching your query in a flat list nested inside the matching parent document. This is useful when you have indexed nested child documents and want to retrieve the child documents for the relevant parent documents for any type of search query. [source,plain] ---- diff --git a/solr/solr-ref-guide/src/uploading-data-with-index-handlers.adoc b/solr/solr-ref-guide/src/uploading-data-with-index-handlers.adoc index 3bf9b52a1d5e..1c090a09693e 100644 --- a/solr/solr-ref-guide/src/uploading-data-with-index-handlers.adoc +++ b/solr/solr-ref-guide/src/uploading-data-with-index-handlers.adoc @@ -541,105 +541,3 @@ In addition to the `/update` handler, there is an additional CSV specific reques |=== The `/update/csv` path may be useful for clients sending in CSV formatted update commands from applications where setting the Content-Type proves difficult. - -== Nested Child Documents - -Solr supports indexing nested documents such as a blog post parent document and comments as child documents -- or products as parent documents and sizes, colors, or other variations as child documents. -The parent with all children is referred to as a "block" and it explains some of the nomenclature of related features. -At query time, the <> can search these relationships, - and the `[child]` <> can attach child documents to the result documents. -In terms of performance, indexing the relationships between documents usually yields much faster queries than an equivalent "query time join", - since the relationships are already stored in the index and do not need to be computed. -However, nested documents are less flexible than query time joins as it imposes rules that some applications may not be able to accept. - -.Note -[NOTE] -==== -A big limitation is that the whole block of parent-children documents must be updated or deleted together, not separately. -In other words, even if a single child document or the parent document is changed, the whole block of parent-child documents must be indexed together. -_Solr does not enforce this rule_; if it's violated, you may get sporadic query failures or incorrect results. -==== - -Nested documents may be indexed via either the XML or JSON data syntax, and is also supported by <> with javabin. - -=== Schema Notes - - * The schema must include an indexed, non-stored field `\_root_`. The value of that field is populated automatically and is the same for all documents in the block, regardless of the inheritance depth. - * Nested documents are very much documents in their own right even if certain nested documents hold different information from the parent. - Therefore: - ** the schema must be able to represent the fields of any document - ** it may be infeasible to use `required` - ** even child documents need a unique `id` - * You must include a field that identifies the parent document as a parent; it can be any field that suits this purpose, and it will be used as input for the <>. - * If you associate a child document as a field (e.g., comment), that field need not be defined in the schema, and probably - shouldn't be as it would be confusing. There is no child document field type. - -=== XML Examples - -For example, here are two documents and their child documents. -It illustrates two styles of adding child documents; the first is associated via a field "comment" (preferred), -and the second is done in the classic way now referred to as an "anonymous" or "unlabelled" child document. -This field label relationship is available to the URP chain in Solr but is ultimately discarded. -Solr 8 will save the relationship. - -[source,xml] ----- - - - 1 - Solr adds block join support - parentDocument - - - 2 - SolrCloud supports it too! - - - - - 3 - New Lucene and Solr release is out - parentDocument - - 4 - Lots of new features - - - ----- - -In this example, we have indexed the parent documents with the field `content_type`, which has the value "parentDocument". -We could have also used a boolean field, such as `isParent`, with a value of "true", or any other similar approach. - -=== JSON Examples - -This example is equivalent to the XML example above. -Again, the field labelled relationship is preferred. -The labelled relationship here is one child document but could have been wrapped in array brackets. -For the anonymous relationship, note the special `\_childDocuments_` key whose contents must be an array of child documents. - -[source,json] ----- -[ - { - "id": "1", - "title": "Solr adds block join support", - "content_type": "parentDocument", - "comment": { - "id": "2", - "comments": "SolrCloud supports it too!" - } - }, - { - "id": "3", - "title": "New Lucene and Solr release is out", - "content_type": "parentDocument", - "_childDocuments_": [ - { - "id": "4", - "comments": "Lots of new features" - } - ] - } -] -----