Query Syntax

Examples

"David cameron" europe tags:blog
Blog articles with "David cameron" AND "europe" in the content. Note the use of quotes to include the space.
headline:"david cameron"
All articles with the prime minister's name in the headline.
headline:(green OR (yellow blue))
returns articles with either "green" in the title, or both "yellow" and "blue".
tags:(grapefruit -citrus)
Find articles tagged grapefruit, but not citrus.
tory
Equivalent to "content: tory". NOTE: this will match both "tory" and "victory". This will likely change in future.
published:[2014-01-01 TO 2014-01-31]
Articles published during January (the range is inclusive - the start/end dates are included)
published:[ TO 2014-05-01]
Articles published before or on 2014-05-01
published:[ 2014-05-01 TO ]
Articles published on or after 2014-05-01
content:ukip
All articles containing ukip in the main text (or just plain "ukip" will do, as content is the default field)
pub:spectator
all articles from the spectator
urls:/commentisfree/ -tags:comment
any article with a url containing /commentisfree/ URL, and isn't tagged as comment.
headline:="weekly roundup"
Match all articles with the exact title "Weekly Roundup" (but not, say, "Bob's weekly roundup"). The equals sign (=) means that the whole field must match.
published:=""
Match all articles with a blank publication date.
kw:"better together"
All articles with a "better together" keyword

Fields

These are the fields you can query:
headline
The title of the article
content
The article text (in html). If no field is specified in a search term, then content is assumed.
authors
Authors (can be more than one, and each author can have multiple fields (name, email, twitter etc)
published
the date (and, optionally, time) that the article was first published
updated
the date (and, optionally, time) that the article was last updated. Might be blank.
tags
a list of tags that have been applied to the article
kw
List of keywords, as scraped from the original article (some sites have good keywords, others have crap)
pub
The publication (sun, guardian etc..)
urls
the URLs of the article (can be more than one)
canonicalurl
the canonical URL of the article (if known)