Created
January 31, 2018 19:28
-
-
Save wildmichael/6f05794b02c2ebf386064294b191a037 to your computer and use it in GitHub Desktop.
Small utility to easily create a document tree with BeautifulSoup4
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from bs4 import BeautifulSoup, Tag | |
| class Attribute(object): | |
| def __init__(self, attrname, attrvalue): | |
| 'Initializes a new attribute.' | |
| self._attrname = attrname | |
| self._attrvalue = attrvalue | |
| @property | |
| def name(self): | |
| 'The attribute name' | |
| return self._attrname | |
| @property | |
| def value(self): | |
| 'The attribute value' | |
| return self._attrvalue | |
| class TagFactory(object): | |
| 'A factory for `bs4.Tag` objects' | |
| def __init__(self, soup): | |
| 'Initializes a new factory given a `bs4.BeautifulSoup`.' | |
| self._soup = soup | |
| @property | |
| def soup(self): | |
| 'The `bs4.BeautifulSoup` instance.' | |
| return self._soup | |
| def __call__(self, tagname, *contents, **kwargs): | |
| 'Instantiates a `bs4.BeautifulSoup` instance given contents.' | |
| tag = None | |
| for c in contents: | |
| if type(c) is Attribute: | |
| if tag is not None: | |
| raise ValueError('Attribute cannot follow content') | |
| kwargs[c.name] = c.value | |
| else: | |
| if tag is None: | |
| tag = self.soup.new_tag(tagname, **kwargs) | |
| tag.append(c) | |
| if tag is None: | |
| tag = self.soup.new_tag(tagname, **kwargs) | |
| return tag | |
| soup = BeautifulSoup('') | |
| f = TagFactory(soup) | |
| html = f('html', | |
| f('head', | |
| f('meta', Attribute('foo', 'bar')) | |
| ), | |
| f('body', | |
| f('h1', 'Hello World!'), | |
| f('p', 'This is a test.') | |
| ) | |
| ) | |
| print(html.prettify()) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment