|
| 1 | +import re |
| 2 | +from bs4 import BeautifulSoup |
| 3 | + |
| 4 | +ITEM_HTML = '''<!DOCTYPE html> |
| 5 | +<html lang="en" dir="ltr"> |
| 6 | + <head> |
| 7 | + <meta charset="utf-8"> |
| 8 | + <title></title> |
| 9 | + </head> |
| 10 | + <body> |
| 11 | + <li class="col-xs-6 col-sm-4 col-md-3 col-lg-3"> |
| 12 | +
|
| 13 | + <article class="product_pod"> |
| 14 | +
|
| 15 | + <div class="image_container"> |
| 16 | +
|
| 17 | + <a href="catalogue/a-light-in-the-attic_1000/index.html"><img src="media/cache/2c/da/2cdad67c44b002e7ead0cc35693c0e8b.jpg" alt="A Light in the Attic" class="thumbnail"></a> |
| 18 | + </div> |
| 19 | +
|
| 20 | + <p class="star-rating Three"> |
| 21 | + <i class="icon-star"></i> |
| 22 | + <i class="icon-star"></i> |
| 23 | + <i class="icon-star"></i> |
| 24 | + <i class="icon-star"></i> |
| 25 | + <i class="icon-star"></i> |
| 26 | + </p> |
| 27 | +
|
| 28 | + <h3><a href="catalogue/a-light-in-the-attic_1000/index.html" title="A Light in the Attic">A Light in the ...</a></h3> |
| 29 | +
|
| 30 | + <div class="product_price"> |
| 31 | +
|
| 32 | + <p class="price_color">£51.77</p> |
| 33 | +
|
| 34 | + <p class="instock availability"> |
| 35 | + <i class="icon-ok"></i> |
| 36 | +
|
| 37 | + In stock |
| 38 | + </p> |
| 39 | +
|
| 40 | + <form> |
| 41 | + <button type="submit" class="btn btn-primary btn-block" data-loading-text="Adding...">Add to basket</button> |
| 42 | + </form> |
| 43 | + </div> |
| 44 | + </article> |
| 45 | + </li> |
| 46 | + </body> |
| 47 | +</html> |
| 48 | +''' |
| 49 | + |
| 50 | +soup = BeautifulSoup(ITEM_HTML, 'html.parser') |
| 51 | + |
| 52 | + |
| 53 | +def find_item_name(): |
| 54 | + locator = 'article.product_pod h3 a' # CSS locator, location of what we want |
| 55 | + item_link = soup.select_one(locator) |
| 56 | + item_name = item_link.attrs['title'] |
| 57 | + print(item_name) |
| 58 | + |
| 59 | + |
| 60 | +def find_item_link(): |
| 61 | + locator = 'article.product_pod h3 a' # CSS locator, location of what we want |
| 62 | + item_link = soup.select_one(locator).attrs['href'] # this is a relative link |
| 63 | + print(item_link) |
| 64 | + |
| 65 | + |
| 66 | +''' |
| 67 | +def find_item_price(): |
| 68 | + # locator = 'article.product_pod p.price_color' |
| 69 | + # item_link = soup.select_one(locator) |
| 70 | + prod_price = soup.find('p', {'class': 'price_color'}).string |
| 71 | + print(prod_price[1:]) # this extracts a strong not a number |
| 72 | +''' |
| 73 | + |
| 74 | + |
| 75 | +def find_item_price(): |
| 76 | + locator = 'article.product_pod p.price_color' |
| 77 | + item_price = soup.select_one(locator).string |
| 78 | + |
| 79 | + pattern = '£([0-9]+\.[0-9]+)' |
| 80 | + matcher = re.search(pattern, item_price) |
| 81 | + print(matcher.group(0)) # £51.77 |
| 82 | + print(float(matcher.group(1))) # 51.77 |
| 83 | + |
| 84 | + |
| 85 | +def find_item_rating(): |
| 86 | + locator = 'article.product_pod p.star-rating' |
| 87 | + star_rating_tag = soup.select_one(locator) |
| 88 | + # print(star_rating_tag) |
| 89 | + classes = star_rating_tag.attrs['class'] |
| 90 | + rating_classes = [r for r in classes if r != 'star-rating'] |
| 91 | + print(rating_classes[0]) |
| 92 | + |
| 93 | + |
| 94 | +find_item_name() |
| 95 | +find_item_link() |
| 96 | +find_item_price() |
| 97 | +find_item_rating() |
0 commit comments