From 9bc14553325da5f17e4b318c9e6616737ec6fbe8 Mon Sep 17 00:00:00 2001
From: Kingu <twkrol@gmail.com>
Date: Sat, 9 Jan 2021 01:29:47 +0100
Subject: [PATCH] LibreOffice fixed Microsoft Edge added

---
 clients/edge.py        | 62 ++++++++++++++++++++++++++++++++++++++++++
 clients/libreoffice.py | 18 ++++++++----
 start.py               |  2 +-
 3 files changed, 76 insertions(+), 6 deletions(-)
 create mode 100644 clients/edge.py

diff --git a/clients/edge.py b/clients/edge.py
new file mode 100644
index 0000000..531c1c4
--- /dev/null
+++ b/clients/edge.py
@@ -0,0 +1,62 @@
+# -*- encoding: utf-8 -*-
+# Copyright (c) Tomasz Krol tomek@kingu.pl
+"""
+Module to grab actual editions & versions of Microsoft Edge
+"""
+import copy
+import urllib.request
+import re
+from datetime import datetime, date
+from bs4 import BeautifulSoup
+import logging
+from logging import critical, error, info, warning, debug
+
+product = "Microsoft Edge"
+
+def getEditions(template):
+    result = []
+
+    template.product = product
+    template.ends = date.max
+    template.stable = True
+    template.latest = True
+    
+    #Looking for releases
+    body = urllib.request.urlopen("https://raw.githubusercontent.com/MicrosoftDocs/Edge-Enterprise/public/edgeenterprise/microsoft-edge-relnote-stable-channel.md").read()
+    soup = BeautifulSoup(body, "html5lib")
+
+    for i, string in enumerate(soup.get_text().splitlines()):
+        if string.startswith("## Version"):
+            # the version line format is: "## Version 87.0.664.66: December 17"
+            elements = string.split(' ')
+            
+            # read version skipping the ending ":"
+            releaseVersion = elements[2][:-1]
+            
+            # decode date, mind the year which is not provided (yet) and must be guessed
+            releaseDate = datetime.strptime(f"{elements[3]} {elements[4]}, {datetime.now().year}", '%B %d, %Y')
+            if releaseDate.date() > datetime.now().date():
+                releaseDate = datetime.strptime(f"{elements[3]} {elements[4]}, {datetime.now().year-1}", '%B %d, %Y')
+            
+            break
+    
+    debug(f"version:{releaseVersion}")
+    debug(f"releaseDate: {releaseDate}")
+
+    item = copy.copy(template)									#copy of Softver object
+
+    #find release date
+    item.released = releaseDate.date()
+
+    #find version
+    value = releaseVersion
+    item.version = value
+    value = re.search('\d+\.\d+',value)
+    item.edition = value.group()
+    
+    result.append(item)
+
+    return result
+
+# if __name__ == "__main__":
+    # print(getEditions(Softver()))
diff --git a/clients/libreoffice.py b/clients/libreoffice.py
index 8495a34..0b24b08 100644
--- a/clients/libreoffice.py
+++ b/clients/libreoffice.py
@@ -10,12 +10,14 @@
 import vergrabber
 from bs4 import BeautifulSoup
 from datetime import datetime, date
+import logging
+from logging import critical, error, info, warning, debug
 
 product = "LibreOffice"
 user_agent_stable = 'LibreOffice 6.2.1 (c9944f7-48b7ff5-0507789-54a4c8a-8b242a8; Windows; X86_64; )'
 user_agent_fresh = 'LibreOffice 6.3.1 (7074905676c47b82bbcfbea1aeefc84afe1c50e1; Windows; X86_64; )'
 url = 'http://update.libreoffice.org/check.php'
-releases_url = "https://wiki.documentfoundation.org/ReleasePlan/"
+releases_url = "https://wiki.documentfoundation.org/ReleasePlan"
 
 """
 get libreoffice webservice output manually:
@@ -49,7 +51,8 @@
 
 def getReleaseDates(item):
     # Looking for release start and end dates
-    body = urllib.request.urlopen(releases_url+item.edition).read()
+    body = urllib.request.urlopen(releases_url + "/" + item.edition).read()
+    debug(f"Looking for release date for {item.version}, requesting url: {releases_url}/{item.edition}")
     soup = BeautifulSoup(body, "html5lib")
 
     # Get all tables from page
@@ -60,15 +63,18 @@ def getReleaseDates(item):
         # release version info is always in last row
         releaseRow = rows[len(rows)-1]
         ths = releaseRow.find_all('th')
-        releaseString = ths[0].get_text()
+        releaseString = ths[0].get_text().strip(' \t\r\n').replace('  ',' ')    #apply some text clearance
         tds = releaseRow.find_all('td')
         release_date_string = tds[0].get_text()
         
         #releaseString.
-        if releaseString.startswith("Release "+item.version):
+        debug(f"testing for {releaseString} == Release {item.version}")
+        if releaseString.startswith(f"Release {item.version}"):
+            debug(f"Looking for date match in {release_date_string}")
             release_match = re.search('([a-zA-Z]{3} \d{1,2}?, \d\d\d\d)', release_date_string)
             if release_match != None:
                 release_date_substr = release_match.group(1)
+                debug(f"Found match: {release_date_substr}")
                 try:
                     release_date = datetime.strptime(release_date_substr, '%b %d, %Y').date()
                     item.released = release_date
@@ -76,10 +82,12 @@ def getReleaseDates(item):
                      print('ValueError:', e)
             
         if releaseString.startswith("End of"):
+            debug(f"Looking for End of life date match in {release_date_string}")
             try:
-                release_end_date = datetime.strptime(release_date_string, '%B %d, %Y').date()
+                release_end_date = datetime.strptime(release_date_string.strip(' \t\r\n'), '%B %d, %Y').date()
                 item.ends = release_end_date
             except ValueError as e:
+                # None
                 print('ValueError:', e)
 
 def getEditions(template):
diff --git a/start.py b/start.py
index ec3c2df..0ee2add 100644
--- a/start.py
+++ b/start.py
@@ -14,7 +14,7 @@
 import argparse
 import vergrabber
 
-__version__ = '4.1.0'
+__version__ = '4.2.0'
 __author__ = 'Tomasz Krol'
 __author_email__ = 'vergrabber@kingu.pl'