Intro

zyte-parsers provides functions that extract specific data from HTML elements. The input element can be an instance of either parsel.selector.Selector or lxml.html.HtmlElement. Some functions can also take a string with text (e.g. extracted from HTML or JSON) as input.

zyte_parsers.SelectorOrElement: alias of Union[Selector, HtmlElement, HtmlComment]

Parsers

Brand

zyte_parsers.extract_brand_name(node: Selector | HtmlElement | HtmlComment, search_depth: int = 0) → str | None[source]

Extract a brand name from a node that contains it.

It tries element text and image alt and title attributes.

Parameters:

node – Node including the brand name.
search_depth – Max depth for searching images.

Returns:

The brand name or None.

Breadcrumbs

class zyte_parsers.Breadcrumb(name: str | None = None, url: str | None = None)[source]

name: str | None

url: str | None

zyte_parsers.extract_breadcrumbs(node: Selector | HtmlElement | HtmlComment, *, base_url: str | None, max_search_depth: int = 10) → Tuple[Breadcrumb, ...] | None[source]

Extract breadcrumb items from node that represents breadcrumb component.

It finds all anchor elements to specified maximal depth. Anchors are collected in pre-order traversal. Such strategy of traversing supports cases where structure of nodes representing breadcrumbs is flat, which means that breadcrumb’s anchors are on the same depth of HTML structure and where breadcrumb items are nested, which means that element with next item can be a child of element with previous breadcrumb item. It also post-processes extracted breadcrumbs by using semantic markup or the location of breadcrumb separators.

Parameters:

node – Node representing and including breadcrumb component.
base_url – Base URL of site.
max_search_depth – Max depth for searching anchors.

Returns:

Tuple with breadcrumb items.

GTIN

class zyte_parsers.Gtin(type: str, value: str)[source]

type: str

value: str

zyte_parsers.extract_gtin(node: Selector | HtmlElement | HtmlComment | str) → Gtin | None[source]

Extract a GTIN (Global Trade Item Number) from a node or a string that contains its text.

It detects the GTIN type and returns it together with the cleaned GTIN value. The following types are supported: isbn10, isbn13, issn, ismn, upc, gtin8, gtin13, gtin14.

Parameters:: node – A node or a string that includes the GTIN text.
Returns:: A GTIN item.

Price

Extract a price value from a node or a string that contains it.

Parameters:

node – A node or a string that includes the price text.
currency_hint – A string or a node that can contain currency. It will be passed as a hint to price-parser. If currency is present in the price string, it could be preferred over the value extracted from currency_hint.

Returns:

The price value as a price_parser.Price object.

Ratings and review count

class zyte_parsers.AggregateRating(bestRating: float | None = None, ratingValue: float | None = None)[source]

bestRating: float | None

ratingValue: float | None

zyte_parsers.extract_rating(node: Selector | HtmlElement | HtmlComment) → AggregateRating[source]

Extract rating data from a node.

Parameters:: node – Node that includes the rating data.
Returns:: AggregateRating item.

zyte_parsers.extract_rating_stars(node: Selector | HtmlElement | HtmlComment) → float | None[source]

Extract a rating value from a node containing rating stars.

Parameters:: node – Node that includes the rating stars.
Returns:: Rating value as a float or None.

zyte_parsers.extract_review_count(node: Selector | HtmlElement | HtmlComment) → int | None[source]

Extract review count from a node containing it.

Parameters:: node – Node that includes the review count.
Returns:: Review count as an int or None.