Align custom scripts with generic scripts (#445)

Align custom scripts with generic scripts, making them configurable. This has a few advantages:

- script code is more unified,
- no more hard-coded method names in scripts, which is less error prone and make it easier to rename scripts,
- no more hard coded product names in scripts, which is less error prone and make it easier to rename products,
- less hard-coded URLs and regexes in scripts, which makes auto-configuration more expressive / updatable,

Also added method `endoflife.list_configs_from_argv()` so that it is easier to manipulate scripts arguments.
This commit is contained in:
Marc Wrobel
2025-06-07 12:41:59 +02:00
parent 60a62e4696
commit f404274310
63 changed files with 704 additions and 830 deletions

View File

@@ -1,31 +1,24 @@
import logging
import re
import sys
from bs4 import BeautifulSoup
from common import dates, endoflife, http, releasedata
"""Fetches EOL dates from Atlassian EOL page.
This script takes a single argument which is the product title identifier on the Atlassian EOL page, such as
This script takes a selector argument which is the product title identifier on the Atlassian EOL page, such as
`AtlassianSupportEndofLifePolicy-JiraSoftware`.
"""
METHOD = "atlassian_eol"
REGEX = r"(?P<release>\d+(\.\d+)+) \(EO[SL] date: (?P<date>.+)\).*$"
PATTERN = re.compile(REGEX, re.MULTILINE)
p_filter = sys.argv[1] if len(sys.argv) > 1 else None
m_filter = sys.argv[2] if len(sys.argv) > 2 else None
for config in endoflife.list_configs(p_filter, METHOD, m_filter):
for config in endoflife.list_configs_from_argv():
with releasedata.ProductData(config.product) as product_data:
content = http.fetch_javascript_url('https://confluence.atlassian.com/support/atlassian-support-end-of-life-policy-201851003.html')
content = http.fetch_javascript_url(config.url)
soup = BeautifulSoup(content, features="html5lib")
for li in soup.select(f"#{config.url}+ul li"):
match = PATTERN.match(li.get_text(strip=True))
# Find the section with the EOL dates
for li in soup.select(f"#{config.data.get('selector')}+ul li"):
match = config.first_match(li.get_text(strip=True))
if not match:
logging.warning(f"Failed to parse EOL date from '{li.get_text(strip=True)}'")
logging.warning(f"Skipping '{li.get_text(strip=True)}', no match found")
continue
release_name = match.group("release")