Fixes #464 Add Image search support of Yahoo

.travis.yml

@@ -9,7 +9,7 @@ install:
   - pip install -r requirements.txt
 
 before_script:
-  - flake8 . --count --max-complexity=15 --show-source --statistics
+  - flake8 . --count --max-complexity=16 --show-source --statistics
 script:
   - python -m app.server > /dev/null &
   - pytest --cov=./

app/scrapers/__init__.py

@@ -42,6 +42,8 @@ def feed_gen(query, engine, count=10, qtype=''):
     engine = old_names.get(engine, engine)
     if engine in ('quora', 'youtube'):
         urls = scrapers[engine].search_without_count(query)
+    elif engine in ('yahoo',) and qtype == 'isch':
+        urls = scrapers[engine].image_search_without_count(query)
     else:
         urls = scrapers[engine].search(query, count, qtype)
     return urls

app/scrapers/generalized.py

@@ -84,3 +84,17 @@ def search_without_count(self, query):
         soup = BeautifulSoup(response.text, 'html.parser')
         urls = self.parse_response(soup)
         return urls
+
+    def image_search_without_count(self, query):
+        """
+            Search for the query and return set of urls
+            Returns: list
+        """
+        urls = []
+        if self.name in ['yahoo']:
+            url = self.imageURL
+            payload = {'p': query}
+        response = requests.get(url, headers=self.headers, params=payload)
+        soup = BeautifulSoup(response.text, 'html.parser')
+        urls = self.parse_image_response(soup)
+        return urls

app/scrapers/yahoo.py

@@ -14,6 +14,7 @@ def __init__(self):
         Scraper.__init__(self)
         self.url = 'https://search.yahoo.com/search'
         self.videoURL = 'https://video.search.yahoo.com/search/video'
+        self.imageURL = 'https://images.search.yahoo.com/search/images'
         self.defaultStart = 1
         self.startKey = 'b'
         self.name = 'yahoo'
@@ -63,3 +64,21 @@ def parse_video_response(soup):
         print('Yahoo parsed: ' + str(urls))
 
         return urls
+
+    @staticmethod
+    def parse_image_response(soup):
+        """ Parse response and returns the urls
+
+            Returns: urls (list)
+                    [[url1], [url2], ...]
+        """
+        urls = []
+        for li in soup.findAll('li', attrs={'class': 'ld'}):
+            url = 'https://images.search.yahoo.com' + li.a.get('href')
+            urls.append({
+                'link': url
+            })
+
+        print('Yahoo parsed: ' + str(urls))
+
+        return urls

app/server.py

@@ -80,7 +80,8 @@ def search(search_engine):
             unicode  # unicode is undefined in Python 3 so NameError is raised
             for line in result:
                 line['link'] = line['link'].encode('utf-8')
-                line['title'] = line['title'].encode('utf-8')
+                if 'title' in line:
+                    line['title'] = line['title'].encode('utf-8')
                 if 'desc' in line:
                     line['desc'] = line['desc'].encode('utf-8')
         except NameError: