What is difference between xdmp:estimate and fn:count?

 Ans: Both functions are used for counting the fragments in Marklogic. The xdmp:estimate is fast because it measures through the index and fn:count function also counts the item but also validates in the index and document. But it is not guaranteed that xdmp:estimate & fn:count give the same result. Sometimes xdmp:estimate can give a false result.

So fn:count function works slowly as compared to xdmp:estimate.

Let's understand in deep, why fn:count & xdmp:estimate results are not the same. 

Step-1: Upload below book catalog XML in Marklogic DB.

<?xml version="1.0" encoding="UTF-8"?>

<catalog>
<book id="bk101">
<author>
Ralls, Kim
</author>
<title>
XML Developer's Guide
</title>
<genre>
Computer
</genre>
<price>
44.95
</price>
<publish_date>
2000-10-01
</publish_date>
<description>
An in-depth look at creating applications
with XML.
</description>
</book>
<book id="bk102">
<author>
Ralls, Kim
</author>
<title>
Midnight Rain
</title>
<genre>
Fantasy
</genre>
<price>
5.95
</price>
<publish_date>
2000-12-16
</publish_date>
<description>
A former architect battles corporate zombies,
an evil sorceress, and her own childhood to become queen
of the world.
</description>
</book>
<book id="bk103">
<author>
Corets, Eva
</author>
<title>
Maeve Ascendant
</title>
<genre>
Fantasy
</genre>
<price>
5.95
</price>
<publish_date>
2000-11-17
</publish_date>
<description>
After the collapse of a nanotechnology
society in England, the young survivors lay the
foundation for a new society.
</description>
</book>
<book id="bk104">
<author>
Corets, Eva
</author>
<title>
Oberon's Legacy
</title>
<genre>
Fantasy
</genre>
<price>
5.95
</price>
<publish_date>
2001-03-10
</publish_date>
<description>
In post-apocalypse England, the mysterious
agent known only as Oberon helps to create a new life
for the inhabitants of London. Sequel to Maeve
Ascendant.
</description>
</book>
</catalog>

Step-2: Run the below query from the Marklogic QConsole:

xquery version "1.0-ml";

fn:concat("xdmp:estimate Result=", xdmp:estimate(fn:doc('catalog.xml')/catalog/book[author = 'Ralls, Kim']))
,
fn:concat("fn:count Result=", fn:count(fn:doc('catalog.xml')/catalog/book[author = 'Ralls, Kim']))


Results are:



xdmp:estimate Result=1


fn:count Result=2

Because:

If a fragment contains more than one matching item for the XPath specified, xdmp:estimate will undercount these items as a single item whereas fn:count would count them individually.

So here below query return the 2 fragments in both cases 

xquery version "1.0-ml";

fn:doc('catalog.xml')/catalog/book[author = 'Ralls, Kim']

As:

<book id="bk101">

<author>
Ralls, Kim
</author>
<title>
XML Developer's Guide
</title>
<genre>
Computer
</genre>
<price>
44.95
</price>
<publish_date>
2000-10-01
</publish_date>
<description>
An in-depth look at creating applications
with XML.
</description>

</book>

<book id="bk102">

<author>
Ralls, Kim
</author>
<title>
Midnight Rain
</title>
<genre>
Fantasy
</genre>
<price>
5.95
</price>
<publish_date>
2000-12-16
</publish_date>
<description>
A former architect battles corporate zombies,
an evil sorceress, and her own childhood to become queen
of the world.
</description>

</book>

But xdmp:estimate is returning 1 because the author name "Rall, Kim" is same for both fragments so it will count as 1 only.

Therefore, fn:count is not relying on the value of the author's name, it counts the actual fragments and returns 2.

I hope you understood. If not, kindly put a comment with a query I'll reply with the solution.

Comments

Post a Comment

Popular posts from this blog

What is universal index and range index?

  Universal index : The universal index is the indexing process of MarkLogic that Marklogic performs automatically when data is loaded into MarkLogic. By default MarkLogic indexing below universal index: Word Indexing Phrase Indexing Relationship Indexing Value Indexing Word and Phrase Indexing Inverted index: An inverted index inverts the document-word relationship into a word-documents relationship.  Each entry in the inverted index is called a  term list .  Range Index:  Ranges index is also an indexing process in Marklogic but this index is created by the developer based on project requirements. When we need a fast result in the project, we create ranges indexes.   Path range index Element range index Element attribute range index etc...
Read more

What is the basic difference between xdmp:spawn and xdmp:invoke a function in Marklogic.

 Aha..... Before understanding  xdmp:spawn-function and xdmp:invoke-function in Marklogic, we need to understand synchronous & asynchronous terms. Also, we need to know what is the main server, task server, and Transactions in the Marklogic. So I'm posting here the simple definition of each term that I mentioned here.  Synchronous and Asynchronous Task: Synchronous : In synchronous operations tasks are performed one at a time and you can't move the following task until the current task is finished. Asynchronous : In asynchronous operations, you can move to another task before the previous one finishes. Main Server : Task server: Transactions: Now,  We need to know what  asynchronous  functions &  asynchronous  functions in Marklogic are: Asynchronous Functions: xdmp:spawn : Place the specified module/query on the task queue for evaluation. This function places the specified module in the task queue to be processed. The module will be eva...
Read more

MarkLogic server architecture.

Image
There are two layers that work in the complete Marklogic application.  The first is the  Data layer . Second  Query layer .   In the  Data layer : Marklogic stores data (in any format, XML, JSON, text, binary data, semantics ) at the bottom of the data layer. On bottom data, Marklogic creates universal index and range index also if defined. Marklogic also creates a cache and journal in the data layer. In all of this, there is also a transaction controller in the data layers. On came in the second layer.  This is the query layer where we evaluate the x query programs.
Read more