56cc29b49d
This way the writing of the JSON file is handled automatically if the update does not fail. It pave the way to further global improvements, such as a better error handling.
20 lines
964 B
Python
20 lines
964 B
Python
from bs4 import BeautifulSoup
|
|
from common import dates, http, releasedata
|
|
|
|
URLS = [
|
|
# Disable, it causes too many timeouts / errors
|
|
# "https://web.archive.org/web/20210123024247/https://www.ibm.com/support/pages/aix-support-lifecycle-information",
|
|
"https://www.ibm.com/support/pages/aix-support-lifecycle-information",
|
|
]
|
|
|
|
with releasedata.ProductData("ibm-aix") as product_data:
|
|
for page in http.fetch_urls(URLS):
|
|
page_soup = BeautifulSoup(page.text, features="html5lib")
|
|
|
|
for release_table in page_soup.find("div", class_="ibm-container-body").find_all("table", class_="ibm-data-table ibm-grid"):
|
|
for row in release_table.find_all("tr")[1:]: # for all rows except the header
|
|
cells = row.find_all("td")
|
|
version = cells[0].text.strip("AIX ").replace(' TL', '.')
|
|
date = dates.parse_month_year_date(cells[1].text)
|
|
product_data.declare_version(version, date)
|