Improve and fix haproxy.py

- Get cycles URL by parsing https://www.haproxy.org/download/:
  - This way the script does not have to be updated each time a new minor version is released.
  - The old algorithm was missing some versions (such as 1.6.x).
- Use a GitHub actions log group to wrap the script logs.
- Display version like the other scripts.

src/haproxy.py

@@ -2,21 +2,73 @@ import json
 import re
 import urllib.request
 
+from bs4 import BeautifulSoup
+
+"""Fetch HAProxy versions with their dates from https://www.haproxy.org/download/.
+"""
+
+PRODUCT = "haproxy"
+CYCLE_REGEX = r"^(\d+\.\d+)/$"
 # https://regex101.com/r/1JCnFC/1
-REGEX = r"^(\d{4})\/(\d{2})\/(\d{2})\s+:\s+(\d+\.\d+\.\d.?)$"
+VERSION_REGEX = r"^(\d{4})\/(\d{2})\/(\d{2})\s+:\s+(\d+\.\d+\.\d.?)$"
 
-versions = {}
 
-for i in range(17, 28):
-    url = "https://www.haproxy.org/download/%s/src/CHANGELOG" % (i / 10)
-    print(url)
-    with urllib.request.urlopen(url) as response:
-        for line in response:
-            m = re.match(REGEX, line.decode("utf-8"))
+def fetch_cycles():
+    cycles = []
+
+    print("Fetching cycles")
+    with urllib.request.urlopen(
+        "https://www.haproxy.org/download/") as response:
+        soup = BeautifulSoup(response, features="html5lib")
+        for link in soup.select("a"):
+            m = re.match(CYCLE_REGEX, link.attrs["href"])
             if m:
-                year, month, date, version = m.groups()
-                abs_date = "%s-%s-%s" % (year, month, date)
-                versions[version] = abs_date
+                cycle = m.groups()[0]
+                cycles.append(cycle)
+                print(f"Found {cycle}")
 
-with open("releases/haproxy.json", "w") as f:
-    f.write(json.dumps(versions, indent=2))
+    # No changelog in https://www.haproxy.org/download/1.0/src
+    cycles.remove("1.0")
+
+    return cycles
+
+
+def fetch_releases(cycles):
+    releases = {}
+
+    for cycle in cycles:
+        url = f"https://www.haproxy.org/download/{cycle}/src/CHANGELOG"
+        print(f"Fetching version from {url}")
+        with urllib.request.urlopen(url) as response:
+            for line in response:
+                m = re.match(VERSION_REGEX, line.decode("utf-8"))
+                if m:
+                    year, month, day, version = m.groups()
+                    date = f"{year}-{month}-{day}"
+                    releases[version] = date
+
+    return releases
+
+
+def print_releases(releases):
+    # Do not print versions in fetch_releases because it contains duplicates
+    for version, date in releases.items():
+        print(f"{version} : {date}")
+
+
+def main():
+    print(f"::group::{PRODUCT}")
+    cycles = fetch_cycles()
+    releases = fetch_releases(cycles)
+    print_releases(releases)
+    print("::endgroup::")
+
+    with open(f"releases/{PRODUCT}.json", "w") as f:
+        f.write(json.dumps(dict(
+            # sort by date then version (desc)
+            sorted(releases.items(), key=lambda x: (x[1], x[0]), reverse=True)
+        ), indent=2))
+
+
+if __name__ == '__main__':
+    main()