From 1e7d583a9a73ffad4520aa9f0ffcd4321d6c77a3 Mon Sep 17 00:00:00 2001
From: Nemo <commits@captnemo.in>
Date: Tue, 13 Dec 2022 12:39:03 +0530
Subject: [PATCH] [gke] Adds GKE automation

This avoids using the GKE APIs, much cleaner as a result.
The downside is that we need to keep track of all channels
separately. This also tracks the date a version is made available
on a channel, not the date it is made the default. This is being
slightly optimistic, but I think works well
---
 src/gke.py | 57 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 src/gke.py

diff --git a/src/gke.py b/src/gke.py
new file mode 100644
index 00000000..2119eda7
--- /dev/null
+++ b/src/gke.py
@@ -0,0 +1,57 @@
+import urllib.request
+from bs4 import BeautifulSoup
+import re
+import json
+from datetime import datetime
+
+# https://regex101.com/r/zPxBqT/1
+REGEX = r"\d.\d+\.\d+-gke\.\d+"
+
+def fetch_channel(channel):
+    url = "https://cloud.google.com/kubernetes-engine/docs/release-notes-{}".format(channel)
+    with urllib.request.urlopen(url, data=None, timeout=5) as response:
+        return BeautifulSoup(response, features="html5lib")
+
+"""
+Takes soup, and returns a dictionary of versions and their release dates
+"""
+def parse_soup_for_versions(soup):
+    """ Parse the soup """
+    versions = {}
+    for section in soup.find_all('section', class_='releases'):
+        # h2 contains the date, which we parse
+        for h2 in section.find_all('h2'):
+            date = h2.get('data-text')
+            date = datetime.strptime(date, '%B %d, %Y').strftime('%Y-%m-%d')
+            # The div next to the h2 contains the notes about changes made on that date
+            next_div = h2.find_next('div')
+            # New releases are noted in a nested list, so we look for that
+            # and parse it using the version regex
+            for li in next_div.find_all('li'):
+                # If the <li> text contains with "versions are now available:", get the <ul> inside the li
+                if "versions are now available" in li.text:
+                    ul = li.find('ul')
+                    for version in re.findall(REGEX, ul.text):
+                        versions[version] = date
+                        print("%s: %s" % (version, date))
+    return versions
+
+CHANNELS = ['nochannel', 'stable', 'regular', 'rapid']
+
+def main():
+    for channel in CHANNELS:
+        soup = fetch_channel(channel)
+        print("::group::GKE - {}".format(channel))
+        versions = parse_soup_for_versions(soup)
+        for version, date in versions.items():
+            print("{}: {}".format(version, date))
+        if channel == 'nochannel':
+            fn = 'releases/gke.json'
+        else:
+            fn = 'releases/gke-{}.json'.format(channel)
+        with open(fn, "w") as f:
+            f.write(json.dumps(versions, indent=2))
+        print("::endgroup::")
+
+if __name__ == '__main__':
+    main()