Struct ImageSearchQuery

pub struct ImageSearchQuery {
    pub q: String,
    pub size: Option<ImageSize>,
    pub page: Option<u32>,
    pub styles: Vec<ImageStyleId>,
    pub age_ranges: Vec<AgeRangeId>,
    pub affiliations: Vec<AffiliationId>,
    pub categories: Vec<CategoryId>,
    pub tags: Vec<ImageTagIndex>,
    pub tags_priority: Vec<ImageTagIndex>,
    pub is_premium: Option<bool>,
    pub is_published: Option<bool>,
    pub page_limit: Option<u32>,
}

Search for images via the given query string.

• kind field must match the case as represented in the returned json body (PascalCase?).
• Vector fields, such as age_ranges should be given as a comma separated vector (CSV).

Fields

q: String

The query string.

size: Option<ImageSize>

Optionally filter by kind

page: Option<u32>

The page number of the images to get.

styles: Vec<ImageStyleId>

Optionally filter by image_styles

age_ranges: Vec<AgeRangeId>

Optionally filter by age_ranges

affiliations: Vec<AffiliationId>

Optionally filter by affiliations

categories: Vec<CategoryId>

Optionally filter by categories

tags: Vec<ImageTagIndex>

Optionally filter by tags

tags_priority: Vec<ImageTagIndex>

Optionally order by tags, given in decreasing priority.

Notes on priority

Consider a request with 4 tags, [clothing, food, red, sports].

“Priority ordering” means that all items tagged as clothing will appear before those without it, and that [clothing, food] will appear before [clothing] or [clothing, red].

Assigning scores

The priority is achieved by using Algolia’s filter scoring feature with "sumOrFiltersScore": true.

Scores are weighted exponentially by a factor of 2. The lowest priority tag is given a score of 1, and the ith highest priority tag is given a score of 2.pow(i). This assignment is provably correct that we get the desired ranking. This can also be interpreted as bit vector with comparison.

NOTE: this means that with i64 range supported by Algolia, we can only assign priority for the first 62 tags. The remaining are all given a score of 1.

Example

For an example request [clothing, food, red, sports], we assign the scores:

tag name	score	(truncated) bit vector score
clothing	8	0b_1000
food	4	0b_0100
red	2	0b_0010
sports	1	0b_0001

This means that the entries will be returned in the following order, based on their tags:

position	entry name	tag names	score	(truncated) bit vector score
0	hat	clothing	8	0b_1000
1	cherry	red, food	6	0b_0110
2	cucumber	green, food	4	0b_0100
3	stop sign	red	2	0b_0010
4	basketball	sports	1	0b_0001
5	wallet	[no tags]	0	0b_0000

is_premium: Option<bool>

Optionally filter by is_premium

is_published: Option<bool>

Optionally filter by is_published

page_limit: Option<u32>

The limit of results per page.

Trait Implementations

impl Clone for ImageSearchQuery

fn clone(&self) -> ImageSearchQuery

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ImageSearchQuery

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for ImageSearchQuery

fn default() -> ImageSearchQuery

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ImageSearchQuery

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ImageSearchQuery

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.