From 27fe903c511691c078942bef5ee9a05a43b15c8f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jes=C3=BAs?= <heckyel@hyperbola.info>
Date: Wed, 9 Jun 2021 17:54:27 -0500
Subject: initial

---
 hypervideo_dl/extractor/googlesearch.py | 59 +++++++++++++++++++++++++++++++++
 1 file changed, 59 insertions(+)
 create mode 100644 hypervideo_dl/extractor/googlesearch.py

(limited to 'hypervideo_dl/extractor/googlesearch.py')

diff --git a/hypervideo_dl/extractor/googlesearch.py b/hypervideo_dl/extractor/googlesearch.py
new file mode 100644
index 0000000..5279fa8
--- /dev/null
+++ b/hypervideo_dl/extractor/googlesearch.py
@@ -0,0 +1,59 @@
+from __future__ import unicode_literals
+
+import itertools
+import re
+
+from .common import SearchInfoExtractor
+
+
+class GoogleSearchIE(SearchInfoExtractor):
+    IE_DESC = 'Google Video search'
+    _MAX_RESULTS = 1000
+    IE_NAME = 'video.google:search'
+    _SEARCH_KEY = 'gvsearch'
+    _TEST = {
+        'url': 'gvsearch15:python language',
+        'info_dict': {
+            'id': 'python language',
+            'title': 'python language',
+        },
+        'playlist_count': 15,
+    }
+
+    def _get_n_results(self, query, n):
+        """Get a specified number of results for a query"""
+
+        entries = []
+        res = {
+            '_type': 'playlist',
+            'id': query,
+            'title': query,
+        }
+
+        for pagenum in itertools.count():
+            webpage = self._download_webpage(
+                'http://www.google.com/search',
+                'gvsearch:' + query,
+                note='Downloading result page %s' % (pagenum + 1),
+                query={
+                    'tbm': 'vid',
+                    'q': query,
+                    'start': pagenum * 10,
+                    'hl': 'en',
+                })
+
+            for hit_idx, mobj in enumerate(re.finditer(
+                    r'<h3 class="r"><a href="([^"]+)"', webpage)):
+
+                # Skip playlists
+                if not re.search(r'id="vidthumb%d"' % (hit_idx + 1), webpage):
+                    continue
+
+                entries.append({
+                    '_type': 'url',
+                    'url': mobj.group(1)
+                })
+
+            if (len(entries) >= n) or not re.search(r'id="pnnext"', webpage):
+                res['entries'] = entries[:n]
+                return res
-- 
cgit v1.2.3