wikipedia

wikipedia

@@ -3,26 +3,41 @@ import requests
 import sys
 import textwrap
 
-def connection_error():
+# Fetch JSON from url and run it through transform, pretty printing errors
-    print("Could not contact wikipedia servers.")
+# and the data worked on as exhaustively as possible.
+def json_query(url, transform, params={}):
+    try:
+        result = requests.get(url, params)
+    except ConnectionError:
+        print("Network connection error.")
+        sys.exit(1)
+    try:
+        data = result.json()
+    except JSONDecodeError as err:
+        print('Error when decoding JSON:\nFrom endpoint ' + url + ':\n' + err + '\n' + result + '\n')
+        sys.exit(1)
+    try:
+        return transform(data)
+    except (IndexError, KeyError) as err:
+        print('Error when traversing JSON:\nFrom endpoint ' + url + ':\n' + err)
+        pprint.PrettyPrinter(indent=2).pprint(data)
         sys.exit(1)
 
-def page_search(string):
+# Search wikipedia for string, returning at most max_results results
+# or the empty list if no matches where returned.
+def page_search(string, max_results):
     params = {
             'q' : string,
-            'limit' : 1
+            'limit' : max_results
             }
-    try:
+    return json_query('https://en.wikipedia.org/w/rest.php/v1/search/page', lambda data: data['pages'], params)
-        return requests.get('https://en.wikipedia.org/w/rest.php/v1/search/page', params).json()['pages']
-    except ConnectionError:
-        connection_error()
 
+# Get a JSON object for the titled page, containing page metadata and a text summary.
 def get_page_with_summary(title):
-    try:
+    return json_query('https://en.wikipedia.org/api/rest_v1/page/summary/' + title, lambda data: data)
-        return requests.get('https://en.wikipedia.org/api/rest_v1/page/summary/' + title).json()
-    except ConnectionError:
-        connection_error()
 
+# Get a list of the links from a page. For a disambiguation page, this means
+# a list of the links to individual pages.
 def get_page_links(title):
     params = {
             'action' : 'query',
@@ -30,19 +45,17 @@ def get_page_links(title):
             'prop' : 'links',
             'format' : 'json'
             }
-    try:
+    return json_query('https://en.wikipedia.org/w/api.php', lambda data: list(data['query']['pages'].values())[0]['links'], params)
-        return list(requests.get('https://en.wikipedia.org/w/api.php', params).json()['query']['pages'].values())[0]['links']
-    except ConnectionError:
-        connection_error()
 
 def main():
-    if not sys.argv[1:]:
+    arg = ' '.join(sys.argv[1:])
+    if not arg:
         print("Usage: wikipedia <list of search terms>")
         sys.exit(1)
     else:
-        result = page_search(' '.join(sys.argv[1:]))
+        results = page_search(arg, 1)
-        if result:
+        if results:
-            page = get_page_with_summary(result[0]['title'])
+            page = get_page_with_summary(results[0]['title'])
             if page['type'] == 'disambiguation':
                 print('Ambiguous result, please clarify:\n    ' + '\n    '.join([link['title'] for link in get_page_links(page['title'])]))
             else: