1
0
This commit is contained in:
Jon Michael Aanes 2024-07-25 13:05:50 +02:00
parent b8c0c5c52c
commit 9dedb18c4f
3 changed files with 104 additions and 3 deletions

View File

@ -0,0 +1,82 @@
"""Scrapes Tavex.dk prices.
By request of my colleague.
"""
import abc
from decimal import Decimal
import dataclasses
import datetime
import logging
import re
import secrets
import bs4
import personal_data.html_util
import personal_data.parse_util
from personal_data.data import DeduplicateMode, Scraper
def parse_dkk_price(dkk: str) -> Decimal:
print(dkk)
if dkk.strip() == '-':
return None
return Decimal(dkk.removesuffix(' DKK').replace(',','.'))
@dataclasses.dataclass(frozen=True)
class TavexScraperBase(Scraper):
deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
@staticmethod
def requires_cfscrape() -> bool:
return True
@abc.abstractmethod
def page_url() -> str:
pass
def scrape(self):
response = self.session.get(self.page_url())
response.raise_for_status()
NOW = personal_data.parse_util.parse_response_datetime(response)
soup = bs4.BeautifulSoup(response.content, 'lxml')
soup_page = personal_data.html_util.normalize_soup_slightly(
soup,
classes=False,
scripts=True,
)
soup = soup.select_one('.product-poster__box .product-poster__table')
table = []
for soup_row in soup.children:
if isinstance(soup_row, bs4.NavigableString):
continue
table.append([soup_cell.get_text().strip() for soup_cell in soup_row if not isinstance(soup_cell, bs4.NavigableString)])
yield {
'time': NOW,
'buy': parse_dkk_price(table[1][1]),
'sell': parse_dkk_price(table[1][2]),
'spread_percentage': Decimal(table[1][3].removesuffix('%')) if len(table[1]) > 3 else None
}
@dataclasses.dataclass(frozen=True)
class TavexScraperGold(TavexScraperBase):
dataset_name = 'prices_tavex_gold'
deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
@staticmethod
def page_url() -> str:
return 'https://tavex.dk/guld/1oz-canadisk-maple-leaf-guldmont/'
@dataclasses.dataclass(frozen=True)
class TavexScraperSilver(TavexScraperBase):
dataset_name = 'prices_tavex_silver'
deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
@staticmethod
def page_url() -> str:
return 'https://tavex.dk/solv/1-oz-american-eagle-solvmont-tidligere-argange/'

View File

@ -57,8 +57,22 @@ def normalize_soup(soup) -> bytes:
text = normalize_soup_lxml(soup).text_content()
return normalize_text(text)
def data_attributes_of_element(e):
for attr_key in list(e.attrs.keys()):
if attr_key.startswith('data-'):
yield attr_key
def normalize_soup_slightly(soup, classes=True, scripts=True, comments=True):
def has_data_attribute(e) -> bool:
for attr_key in data_attributes_of_element(e):
return True
return False
def normalize_soup_slightly(soup,
classes=True,
scripts=True,
comments=True,
data_attributes=True):
"""Perform soup normalization."""
# Little if any content
for tag in HTML_TAGS_MOSTLY_CONTENTLESS:
for e in soup.select(tag):
@ -85,5 +99,11 @@ def normalize_soup_slightly(soup, classes=True, scripts=True, comments=True):
for c in soup.find_all(string=lambda text: isinstance(text, bs4.Comment)):
c.extract()
if data_attributes:
for e in soup.find_all(has_data_attribute):
for attr_key in data_attributes_of_element(e):
del e[attr_key], attr_key
del e
soup.smooth()
return soup

View File

@ -30,8 +30,7 @@ def parse_duration(text: str) -> datetime.timedelta:
def parse_response_datetime(response) -> datetime.datetime:
return datetime.datetime.strptime(response.headers['Date'], FORMAT_DATE_HEADER)
return datetime.datetime.strptime(response.headers['Date'], FORMAT_DATE_HEADER).replace(tzinfo=datetime.UTC)
def parse_time(text: str) -> datetime.datetime:
text = text.replace('\n', ' ')