Core fields
These fields exist on every document of a site search index. They are what you search, filter, sort and show when you build your own search page. The types are explained in How text is analysed.
Content
| Field | Type | What it is |
|---|---|---|
id | string | The unique key: the MD5 of the uri (Fields Opensolr adds). |
uri | text_general | The address of the page or file, searchable word by word, highlighted. |
title | text_general | The title. It counts most in the default field weights. Highlighted. |
description | text_general | The short description, shown as the result snippet. Highlighted. |
text | text_general | The full body, the largest field. Highlighted. |
author | text_general | The author, searchable; an exact copy is kept in author_s. |
How much each field counts is set in Field weights.
Metadata, time and numbers
| Field | Type | What it is |
|---|---|---|
content_type | string | The file type: text/html, application/pdf and so on. For facets and filters. |
content_status | string | The HTTP status the page answered with when it was read. |
signature | string | A hash of the content, used to find duplicates. |
category | string | A free category label. For facets and filters. |
og_image | string | The address of the thumbnail. Returned, but not searchable. |
creation_date | pdate | The publishing date. For date ranges and the freshest results first. |
timestamp | plong | The same moment as Unix time, for fast sorting. |
rank | pint | A ranking value of your own, for your boost functions. |
size | pint | The size of the content, 0 by default. |
Coordinates and sentiment
| Field | Type | What it is |
|---|---|---|
coords | location | A latitude,longitude point for distance filters and sorting ({!geofilt}, geodist()). |
lat, lon | pdouble | The same coordinates as plain numbers, 0 by default. |
sent_pos, sent_neu, sent_neg | pdouble | How positive, neutral and negative the text reads, from 0 to 1. |
sent_com | pdouble | The overall sentiment, from -1 to 1. fq=sent_com:[0.5 TO *] keeps clearly positive content. |
Search-only fields (searched, never returned)
| Field | Type | What it is |
|---|---|---|
spell | textSpell | The dictionary behind "Did you mean". |
tags, title_tags | edgy_text_kw | Matching the start of a whole phrase, for suggestions while typing. |
tags_ws, title_tags_ws | edgy_text_ws | Matching the start of each word. |
embeddings | knn_vector | The meaning vector of 1,024 numbers behind search by meaning and hybrid search. Not stored: it is searched, never read back. |
_version_, _root_ | plong, string | Solr's own bookkeeping. Leave them alone. |
What a page read by the crawl also carries
meta_plus the name of every meta tag of the page, for examplemeta_og_site_name, along withmeta_domain,meta_iconandmeta_detected_language.price_fandcurrency_s: the price found on a product page;product_category_sm: its categories (Prices and product categories).- The structured data of the page (JSON-LD, microdata, breadcrumbs), each value in a list field typed by what it holds:
_sm,_fm,_imor_dtm. text_t: the structured data as searchable text;video_duration_iand other video fields on pages with a video (Video details);quality_f: the content quality score.
The exact fields of your index are listed in the Control Panel when you set up Facets.