Using real system proxies + rcnamespace + docs + default values

- Adding __init__ docs
- Adding WikimediaAPI class constructor default values for endpoint and
format
- Real usage of rcnamespaces with namespace conversion from string to int

1 files changed, 50 insertions, 14 deletions

diff --git a/src/downloader/__init__.py b/src/downloader/__init__.py
index cfcf20a..4dd6169 100644
--- a/src/downloader/__init__.py
+++ b/src/downloader/__init__.py
@@ -6,15 +6,13 @@ program, including manipulation of the wikimedia API.
 """
 
 import urllib.request
-# For system proxy
-import os
 
 
 class Downloader():
     """Class used to download a given webpage considering system proxy"""
     def __init__(self):
-        self.proxy_address = os.environ.get("HTTP_Proxy")
-        self.proxy = urllib.request.ProxyHandler({'http': self.proxy_address})
+        """ Downloader class constructor """
+        self.proxy = urllib.request.ProxyHandler(urllib.request.getproxies())
         self.opener = urllib.request.build_opener(self.proxy)
         urllib.request.install_opener(self.opener)
 
@@ -31,13 +29,19 @@ class Downloader():
 class WikimediaAPI():
     """
     Class used to generate wikimedia API urls for several uses
-
-    The endpoint for this project should be "http://en.wikipedia.org/w/api.php"
-    but can be other wiki api endpoint made with the Wikimedia software.
-    The return_format can be one of json, php, wddx, xml, yaml, raw, txt, dbg,
-    dump or none.
     """
-    def __init__(self, endpoint, return_format):
+    def __init__(self, endpoint="http://en.wikipedia.org/w/api.php",
+                 return_format="json"):
+        """
+        WikimediaAPI class constructor
+
+        The endpoint for this project should be
+        "http://en.wikipedia.org/w/api.php" but it can be any other wiki
+        api endpoint made with the Wikimedia software.
+
+        The return_format can be one of json, php, wddx, xml, yaml, raw, txt,
+        dbg, dump or none.
+        """
         self.endpoint = endpoint
         self.return_format = return_format
 
@@ -47,8 +51,40 @@ class WikimediaAPI():
         (https://www.mediawiki.org/wiki/API:Recentchanges)
 
         The namespace is used to restrict the results to a certain level. It
-        can be "(Main)" which is the default one, "Wikipedia", "File" or
-        others. See https://meta.wikimedia.org/wiki/Help:Namespace
+        can be (Main) which is the default one, "Wikipedia", "File" or
+        others. It will be converted to an int corresponding to the rcnamespace
+        parameter. See https://meta.wikimedia.org/wiki/Help:Namespace
         """
-        return self.base_url + "?action=query&list=recentchanges&format="\
-            + self.return_format + "&namespace=" + namespace
+        rcnamespaces = {
+            "(Main)": "0",
+            "Talk": "1",
+            "User talk": "2",
+            "Wikipedia": "3",
+            "Wikipedia talk": "4",
+            "File": "5",
+            "File talk": "6",
+            "MediaWiki": "7",
+            "MediaWiki talk": "8",
+            "Template": "9",
+            "Template talk": "10",
+            "Help": "11",
+            "Help talk": "12",
+            "Category": "13",
+            "Category talk": "14",
+            # Custom Wikipedia namespaces
+            "Portal": "100",
+            "Portal talk": "101",
+            "Book": "108",
+            "Book talk": "109",
+            "Draft": "118",
+            "Draft talk": "119",
+            "Education Program": "446",
+            "Education Program talk": "447",
+            "TimedText": "710",
+            "TimedText talk": "711",
+            "Module": "828",
+            "Module talk": "829",
+            "Topic": "2600"
+        }
+        return self.endpoint + "?action=query&list=recentchanges&format="\
+            + self.return_format + "&namespace=" + rcnamespaces[namespace]