Searching

Anki's Browse screen and the Filtered Deck feature use a common method of searching for specific cards/notes.

Simple searches

When you type some text into the search box, Anki finds matching notes and displays their cards. Anki searches in all fields of the notes, but does not search for tags (see later in this section for how to search for tags). Some examples:

dog
search for "dog" - will match words like "doggy" and "underdog" too.

dog cat
finds notes that have both "dog" and "cat" on them, such as "raining cats and dogs".

dog or cat
finds notes with either "dog" or "cat".

dog (cat or mouse)
finds notes with dog and cat, or dog and mouse.

-cat
finds notes without the word "cat".

-cat -mouse
finds notes with neither "cat" nor "mouse".

-(cat or mouse)
same as the above.

"a dog"
finds notes with the exact sequence of characters "a dog" on them, such as "atta dog", but not "dog a" or "adog".

-"a dog"
finds notes without the exact phrase "a dog"

d_g
finds notes with d, <a letter>, g, like dog, dig, dug, and so on.

d*g
finds notes with d, <zero or more letters>, g, like dg, dog, dung, etc.

w:dog
search for "dog" on a word boundary - will match "dog", but not "doggy" or "underdog". Requires Anki 2.1.24+ or AnkiMobile 2.1.61+.

w:dog*
will match "dog" and "doggy", but not "underdog".

w:*dog
will match "dog" and "underdog", but not "doggy".

Things to note from the above:

  • Search terms are separated by spaces.

  • When multiple search terms are provided, Anki looks for notes that match all of the terms - an implicit 'and' is inserted between each term. On Anki 2.1.24+ and AnkiMobile 2.0.60+ you can be explicit if you like ("dog and cat" is the same as "dog cat"), but older Anki versions will treat "and" as just another word to search for.

  • You can use "or" if you only need one of the terms to match.

  • You can prepend a minus sign to a term to find notes that don’t match.

  • You can group search terms by placing them in parentheses, as in the dog (cat or mouse) example. This becomes important when combining OR and AND searches — in the example, with the parentheses, it matches either 'dog cat' or 'dog mouse', whereas without them it would match either 'dog and cat' or 'mouse'.

  • Anki is only able to search within formatting in the sort field you’ve configured. For example, if you add "example" to one of your fields, this will not be matched when searching for "example" unless that field is the sort field. If a word is not formatted, or the formatting does not change in the middle of the word, then Anki will be able to find it in any field.

  • Standard searches are case insensitive for Latin characters - a-z will match A-Z, and vice versa. Other characters such as Cyrillic are case sensitive in a standard search, but can be made case insensitive by searching on a word boundary or regular expression (w:, re:).

Limiting to a field

You can also ask Anki to match only if a particular field contains some text. Unlike the searches above, searching on fields requires an 'exact match' by default.

front:dog
find notes with a Front field of exactly "dog". A field that says "a dog" will not match.

front:*dog*
find notes with Front field containing dog somewhere

front:
find notes that have an empty Front field

front:_*
find notes that have a non-empty Front field

front:*
find notes that have a Front field, empty or not

fr*:text find notes in a field starting with "fr". Requires Anki 2.1.24+ or AnkiMobile 2.1.60+.

Tags, decks, cards and notes

tag:animal
find notes with the tag "animal", or subtags like "animal::mammal"

tag:none
find notes with no tags

tag:ani*
find notes with tags starting with ani

deck:french
find cards in a French deck, or subdecks like French::Vocab

deck:french -deck:french::*
find cards in French, but not subdecks

deck:"french vocab"
searching when a deck has a space

"deck:french vocab"
also ok

deck:filtered
filtered decks only

-deck:filtered
normal decks only

card:forward
search for Forward cards

card:1
search for cards by template number - eg, to find the second cloze deletion for a note, you’d use card:2

note:basic
search for cards with a Basic note type

Ignoring accents/combining characters

Requires Anki 2.1.24+ or AnkiMobile 2.0.60+.

You can use nc: to remove combining characters ("no combining"). For example:

nc:uber
matches notes with "uber", "über", "Über" and so on.

nc:は
matches "は", "ば", and "ぱ"

Searches that ignore combining characters are slower than regular searches.

Regular expressions

Anki 2.1.24+ and AnkiMobile 2.0.60+ support searching in notes with "regular expressions", a standard and powerful way of searching in text.

Start a search with re: to search by regular expression. To make things easier, Anki will treat the following as raw input, so bear in mind the rules listed there.

Some examples:

"re:(some|another).*thing"
find notes that have "some" or "another" on them, followed by 0 or more characters, and then "thing"

re:\d{3}
find notes that have 3 digits in a row

Regular expressions can also be limited to a specific field. Please note that unlike the normal searches in a specific field, regular expressions in fields don't require an exact match. Eg:

front:re:[a-c]1
matches uppercase or lowercase a1, B1 or c1 that occurs anywhere in the "Front" field

front:re:^[a-c]1$
like the above, but will not match if any other text falls before or after a1/b1/c1.

You can learn more about regular expressions here: https://regexone.com/lesson/introduction_abcs

Some things to be aware of:

  • The search is case-insensitive by default; use (?-i) at the start to turn on case sensitivity.
  • Some text like spaces and newlines may be represented differently in HTML - you can use the HTML editor in the editing screen to see the underlying HTML contents.
  • For the specifics of Anki's regex support, please see the regex crate documentation: https://docs.rs/regex/1.3.9/regex/#syntax

Card state

is:due
review cards and learning cards waiting to be studied

is:new
new cards

is:learn
cards in learning

is:review
reviews (both due and not due) and lapsed cards

is:suspended
cards that have been manually suspended

is:buried
cards that have been buried, either automatically or manually

Note that with the new scheduler, Anki now distinguishes between manually and automatically buried cards so you can unbury one set without the other.

Cards that have lapsed fall into several of these categories, so it may be useful to combine them to get more precise results:

is:learn is:review
cards that have lapsed and are awaiting relearning

-is:learn is:review
review cards, not including lapsed cards

is:learn -is:review
cards that are in learning for the first time

flag:1
cards with a red flag

flag:2
cards with an orange flag

flag:3
cards with a green flag

flag:4
cards with a blue flag

flag:5
cards with a pink flag

flag:6
cards with a turquoise flag

flag:7
cards with a purple flag

Card properties

prop:ivl>=10
cards with interval of 10 days or more

prop:due=1
cards due tomorrow

prop:due=-1
cards due yesterday that haven’t been answered yet

prop:due>-1 prop:due<1
cards due between yesterday and tomorrow

prop:reps<10
cards that have been answered less than 10 times

prop:lapses>3
cards that have moved into relearning more than 3 times

prop:ease!=2.5
cards easier or harder than default

Note that due only matches review cards and learning cards with an interval of a day or more: cards in learning with small intervals like 10 minutes are not included.

Recent Events

Added

added:1
cards added today

added:7
cards added in last week

The check is made against card creation time rather than note creation time, so cards that were generated within the time frame will be included even if their notes were added a long time ago.

Edited

edited:n
cards where the note text was added/edited in the last n days.

This requires Anki 2.1.28+ / AnkiMobile 2.0.64+.

Answered

rated:1
cards answered today

rated:1:2
cards answered Hard (2) today

rated:7:1
cards answered Again (1) over the last 7 days

rated:31:4
cards answered Easy (4) in the last month

Rating searches had been limited to 31 days before version 2.1.39.

First Answered

On version 2.1.45+, you can also search for the very first review only:

introduced:1
cards answered for the first time today

introduced:365
cards answered for the first time within the last 365 days

Matching special characters

This section was written for Anki 2.1.36+ - earlier versions did not support escaping characters in certain situations.

As shown in the previous section, some characters like *, _ and " have a special meaning in Anki. If you need to locate those characters in a search, you need to tell Anki not to treat them specially.

  • Space
    To match something including spaces, enclose the "entire term" in double quotes. If it is a colon search, you also have the option to only quote the part:"after the colon".

  • ", * and _
    Add a backslash before these characters to treat them literally. For example, _ will match any single character, but \_ matches only an actual underscore.

  • \
    Because a backlash is used to remove the special meaning from other characters, it too is treated specially. If you need to search for an actual backslash, use \\ instead of \.

  • ( and )
    You can search for parentheses either by enclosing the full term in quotes, and/or by using a backslash. That is, "some(text)", some\(text\) and "some\(text\)" are all equivalent, but some(text) is not.

  • -
    Starting a search term with - usually inverts it: -dog matches everything except dog for example. If you instead wish to include an actual hyphen, you can either use a backslash, or include the text in quotes, such as \-.- or "-.-".

  • :
    Colons have to be escaped unless they are preceded by another, unescaped colon. So w:e:b is a word boundary search for e:b, w\:e\:b searches literally for w:e:b and w\:e:b searches the field w:e for b (see field searches).

  • &, <, and >
    &, <, and > are treated as HTML when searching in Anki, and as such searches containing them don't work as expected. However, you can search for them by using their corresponding HTML entity names (&amp; for &, &lt; for <, and &gt; for >). For example, searching &lt;&amp;text&gt; searches for a card with <&text> in a field.

Raw input

Text preceded by certain keywords (like re:) will be treated as raw input. That is, the characters listed above largely lose their special meaning. In such a context, only a minimum of escaping is required to prevent ambiguity:

  • " must be escaped.

  • Spaces and unescaped parentheses require the search term to be quoted.

  • The search term must not end in an odd number of backslashes.

Object IDs

nid:123
all cards of the note with note id 123

cid:123,456,789
all cards with card ids 123, 456 or 789

Note and card IDs can be found in the card info dialog in the browser. These searches may also be helpful when doing add-on development or otherwise working closely with the database.

Object IDs will not work in the mobile clients, and are not intended to be used in filtered decks at the moment.