Contact UsGet started
© 2024 Iron Hive LLC Belgrade

XML format of a text search response

Written by
Updated at November 20, 2024

In response to a text search query, Search API can return a UTF-8 encoded XML file with search results. The service returns an XML response for text queries processed both via API v1 and API v2.

Alert

You can get a maximum of 1,000 search results per search query. Depending on the docs-in-group attribute value, each result may contain one to three documents. The maximum number of search results pages depends on the number of groups of documents per page, which is defined by the groups-on-page attribute. For example, if the groups-on-page attribute is set to 10, a maximum of 100 pages with search results can be generated.

Files consist of grouping tags such as request (aggregate information on search query parameters) and response (search query handling results).

Below is the general structure of the resulting XML document with examples of values.

Warning

This structure is provided for information purposes only and contains mutually exclusive elements.

Response structure

<?xml version="1.0" encoding="utf-8"?>
<yandexsearch version="1.0">
<request>
   <query>yandex</query>
   <page>0</page>
   <sortby order="descending" priority="no">rlv</sortby>
   <maxpassages>2</maxpassages>
   <groupings>
      <groupby  attr="d" mode="deep" groups-on-page="10" docs-in-group="3" curcateg="-1" />
   </groupings>
</request>
<response date="20120928T103130">
   <error code="15">No results found for your word combination</error>
   <reqid>1348828873568466-1289158387737177180255457-3-011-XML</reqid>
   <found priority="phrase">206775197</found>
   <found priority="strict">206775197</found>
   <found priority="all">206775197</found>
   <found-human>207 million results found</found-human>
   <misspell>
      <rule>Misspell</rule>
      <source-text>yande<hlword>xx</hlword></source-text>
      <text>yandex</text>
   </misspell>
   <reask>
      <rule>Misspell</rule>
      <source-text><hlword>yn</hlword>dex</source-text>
      <text-to-show>yandex</text-to-show>
      <text>yandex</text>
   </reask>
   <results>
      <grouping attr="d" mode="deep" groups-on-page="10" docs-in-group="3" curcateg="-1">
         <found priority="phrase">45094</found>
         <found priority="strict">45094</found>
         <found priority="all">45094</found>
         <found-docs priority="phrase">192685602</found-docs>
         <found-docs priority="strict">192685602</found-docs>
         <found-docs priority="all">192685602</found-docs>
         <found-human>193 million results found</found-human>
         <page first="1" last="10">0</page>
         <group>
            <categ attr="d" name="UngroupVital223.ru" />
            <doccount>34</doccount>
            <relevance priority="all" />
            <doc id="ZD831E1113BCFDD95">
               <relevance priority="phrase" />
               <url>https://www.yandex.ru/</url>
               <domain>www.yandex.ru</domain>
               <title><hlword>Yandex</hlword>: search engine and internet portal</title>
               <headline>Internet-wide search based on the user's region.</headline>
               <modtime>20060814T040000</modtime>
               <size>26938</size>
               <charset>utf-8</charset>
               <passages>
                  <passage><hlword>Yandex</hlword> is a search engine responding to your query with...</passage>
               </passages>
               <properties>
                   <_PassagesType>0</_PassagesType>
                   <lang>ru</lang>
               </properties>
               <mime-type>text/html</mime-type>
               <saved-copy-url>https://hghltd.yandex.net/yandbtm?text=yandex&amp;url=https%3A%2F%2Fwww.yandex.ru%2F&amp;fmode=inject&amp;mime=html&amp;l10n=ru&amp;sign=e3737561fc3d1105967d1ce6********&amp;keyno=0</saved-copy-url>
            </doc>
         </group>
      </grouping>
   </results>
</response>
</yandexsearch>

Response parameters

The request group provides aggregate information about request parameters. It may be missing if there are errors in the response.

request

Request group tags

Description

Attributes

query

Search query text

None

page

Number of the returned page with search results. Page numbering starts from zero (0 stands for page 1).

None

sortby

Result sorting parameters. The possible values are:

  • rlv: By relevance.
  • tm: By document update time.

  • order: Sorting order. The default value is descending (forward sorting). When sorting by update time, the parameter may take the ascending value (reverse sorting).
  • priority: Service attribute. Takes the no value.

maxpassages

Maximum number of passages per search result

None

groupings

This is a grouping tag.

It contains grouping parameters in the groupby tag.

None

groupby

Parameters for grouping search results found.

  • mode: Grouping method.
  • attr: Service attribute.
  • groups-on-page: Maximum number of groups to return per page with search results.
  • docs-in-group: Maximum number of documents to return per group. Any group may contain fewer documents than set by this parameter.
  • curcateg: Service attribute takes the -1 value.

The example below shows the contents of the request grouping tag returned in response to a query:

https://yandex.com.tr/search/xml?l10n=en&user=xml-search-user&key=03.79031114:b631r9j587dkl4jko987hgg7********&query=%22has%20sample%20applications%20for%20the%20most%20popular%20programming%22&sortby=tm&maxpassages=2&groupby=attr%3Dd.mode%3Ddeep.groups-on-page%3D5.docs-in-group%3D3&maxpassages=3&page=1

<request>
  <query>"has sample applications for the most popular programming"</query>
  <page>1</page>
  <sortby order="descending" priority="no">tm</sortby>
  <maxpassages>2</maxpassages>
  <groupings>
    <groupby attr="d" mode="deep" groups-on-page="5" docs-in-group="3" curcateg="-1"/>
  </groupings>
</request>

response

This group has the results of handling the search query, the information on which is provided in the request child tags.

It contains the date attribute with the query date and time (UTC) in <year><month><day>Т<hour><minute><second> format.

This group consists of the following sections:

General information about search results

The table below lists the tags used in the appropriate section.

Tags for general information about search results

Description

Attributes

error

Error description

Used only if a search query is handled incorrectly (e.g., if the query is empty or parameters are incorrect).

In some cases, it is mutually exclusive with other tags of the response grouping tag.

code: Error code.

reqid

Unique query ID

None

found

Estimate of the number of documents found in response to the query.

priority: Service attribute. The possible values are:

  • phrase
  • strict
  • all.

found-human

String in the language matching the selected search type. It shows the number of found documents and contains related information.

None

misspell / reasksection

This is an optional tag and is only used if a typo is detected (misspell) or fixed (reask) in the query.

The table below lists the tags for this section.

misspell / reask section tags

Description

Attributes

misspell

This is a grouping tag.

It contains information about a possible typo in the search query.

None

reask

This is a grouping tag.

It contains information about a fix made in the original query before searching for documents.

None

rule

Type of error found in the query.

The possible values are:

  • Misspell: Typo.
  • KeyboardLayout: Incorrect keyboard layout.
  • Volapyuk: Query is made in Russian transliterated into English. It is used if (yandex.ru) is selected for search type.Russian

None

source-text

Source query text.

The hlword tag is added to a fragment of a search query that might contain an error.

None

text-to-show

This tag is optional and used only for the reask grouping tag.

It contains the fixed search query text. In most cases, it matches the value provided in the text tag.

None

text

Fixed search query text

None

resultssection

This is an optional tag and only used if any results are found for a query.

The table below lists the tags for this section.

results section tags

Description

Attributes

results

This is a grouping tag. Child tags contain information about search parameters and found documents.

None

grouping

This is a grouping tag. Child tags contain information about search parameters and found documents.

Attributes reflect the grouping rules for documents.

  • mode: Grouping method.
  • attr: Service attribute that depends on the mode attribute value.
  • groups-on-page: Number of groups returned per page with search results.
  • docs-in-group: Number of documents returned per group.
  • curcateg: Service attribute takes the -1 value.

found

Estimate of the number of generated groups

priority: Service attribute. The possible values are:

  • phrase
  • strict
  • all.

found-docs

Estimate of the number of documents found in response to the query.

It is a more accurate estimate as compared to the value provided in the found tag of the section with general information about search results.

priority: Service attribute. The possible values are:

  • phrase
  • strict
  • all.

found-docs-human

A string in the language matching the selected search type. It shows the number of found documents and contains related information.

The value being provided must be used when generating search results.

None

page

Number of the returned page with search results. Page numbering starts from zero (0 stands for page 1).

  • first: Number of the first group with search results displayed on the page.
  • last: Number of the last group with search results displayed on the page.

group

This is a grouping tag.

Each group tag contains information about the found document group.

None

categ

Identification data of the group of found documents

doccount

Estimate of the number of documents used to create the group.

Documents that qualify for inclusion in the group are ranked based on the query criteria provided in the sortby parameter. Depending on the docs-in-group parameter value, from one to three of the first documents are included into the group.

None

relevance

This is a service tag.

priority: Service attribute.

doc

This is a grouping tag.

Each doc tag contains information about the found document.

Depending on the docs-in-group parameter value, each group can contain from one to three doc grouping tags.

id: Unique ID of the found document.

url

Unique ID of the found document

None

domain

Domain hosting the found document

None

title

Title of the found document.

Words included in the search query are highlighted with the hlword tag.

None

headline

This is an optional tag that stands for a document annotation.

It is generated using the meta HTML tag with the name attribute set to description.

None

modtime

Date and time of document update in the following format:

<year><month><day>Т<hour><minute><second>

Warning

This tag is optional and may be missing in some cases.

None

size

Size of the found document, in bytes

None

charset

Encoding of the found document

None

passages

Grouping tag with a list of document passages

None

passage

Passage with a document annotation.

Words included in the search query are highlighted with the hlword tag.

The maximum number of passages provided in a single passages tag depends on the maxpassages parameter value of the search query.

None

mime-type

Document type according to RFC2046

None

properties

Grouping tag with document properties

None

_PassagesType

Passage type. The possible values are:

  • 0: Regular passage (derived from the document text).
  • 1: Passage based on the link text. Used if the document is found via a link.

None

lang

This is an optional tag

It determines the document language.

None

saved-copy-url

URL of the saved document copy

None

See also

Previous
Text search
Next
Quotas and limits