Context: I have the following little query in Beautiful Soup, and then build a list comprehension full of tuples from it. It works great:
tags = soup.find_all('span', {'class': 'tags-links'})
title_text_list = [(tag['title'], tag.text) for tag in tags]
Question: Why do we access the title like a dictionary, but the text with object notation? Why not do both with object notation, or both like a dictionary?
The difference in syntax is due to how Beautiful Soup is designed to interact with HTML elements and their attributes.
1. Attributes like title, class, id, etc.: These are accessed using dictionary-like syntax (tag['attribute']) because they are attributes of the HTML tag. In HTML, attributes are always key-value pairs, so Beautiful Soup treats these attributes like key-value pairs in a dictionary, allowing access their values using square brackets. Example:
<span title="some title" class="some-class">Text here</span>
In this example, title and class are attributes of the span tag that are always key-value pairs, and thus are stored and accessed as key-value pairs with dictionary-like syntax.
2. Text content of a tag: This is accessed using .text because it's not an attribute or key-value pair, but rather the text content enclosed within the opening and closing tags. Beautiful Soup provides a .text attribute to directly access this content.
<span>Text here</span>
In this example, "Text here" is the text content of the span tag, and there's no key.
In summary, attributes are accessed like dictionary key-value pairs because HTML attributes are always key-value pairs, while the text content is accessed as an attribute of the Beautiful Soup Tag object since it does not have a key in HTML itself but rather exists between tags. This design choice aims to make it intuitive to access different types of information one might need from an HTML element by following the structure of the HTML itself.