google_videos.html 21 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267
  1. <!DOCTYPE html>
  2. <html lang="en" data-content_root="../../../">
  3. <head>
  4. <meta charset="utf-8" />
  5. <meta name="viewport" content="width=device-width, initial-scale=1.0" />
  6. <meta name="viewport" content="width=device-width, initial-scale=1">
  7. <title>searx.engines.google_videos &#8212; SearXNG Documentation (2025.4.30+fd33559cf)</title>
  8. <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=6625fa76" />
  9. <link rel="stylesheet" type="text/css" href="../../../_static/searxng.css?v=52e4ff28" />
  10. <script src="../../../_static/documentation_options.js?v=b12b0e12"></script>
  11. <script src="../../../_static/doctools.js?v=9a2dae69"></script>
  12. <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
  13. <script data-project="searxng" data-version="2025.4.30+fd33559cf" src="../../../_static/describe_version.js?v=fa7f30d0"></script>
  14. <link rel="index" title="Index" href="../../../genindex.html" />
  15. <link rel="search" title="Search" href="../../../search.html" />
  16. </head><body>
  17. <div class="related" role="navigation" aria-label="Related">
  18. <h3>Navigation</h3>
  19. <ul>
  20. <li class="right" style="margin-right: 10px">
  21. <a href="../../../genindex.html" title="General Index"
  22. accesskey="I">index</a></li>
  23. <li class="right" >
  24. <a href="../../../py-modindex.html" title="Python Module Index"
  25. >modules</a> |</li>
  26. <li class="nav-item nav-item-0"><a href="../../../index.html">SearXNG Documentation (2025.4.30+fd33559cf)</a> &#187;</li>
  27. <li class="nav-item nav-item-1"><a href="../../index.html" >Module code</a> &#187;</li>
  28. <li class="nav-item nav-item-2"><a href="../engines.html" accesskey="U">searx.engines</a> &#187;</li>
  29. <li class="nav-item nav-item-this"><a href="">searx.engines.google_videos</a></li>
  30. </ul>
  31. </div>
  32. <div class="document">
  33. <div class="documentwrapper">
  34. <div class="bodywrapper">
  35. <div class="body" role="main">
  36. <h1>Source code for searx.engines.google_videos</h1><div class="highlight"><pre>
  37. <span></span><span class="c1"># SPDX-License-Identifier: AGPL-3.0-or-later</span>
  38. <span class="sd">&quot;&quot;&quot;This is the implementation of the Google Videos engine.</span>
  39. <span class="sd">.. admonition:: Content-Security-Policy (CSP)</span>
  40. <span class="sd"> This engine needs to allow images from the `data URLs`_ (prefixed with the</span>
  41. <span class="sd"> ``data:`` scheme)::</span>
  42. <span class="sd"> Header set Content-Security-Policy &quot;img-src &#39;self&#39; data: ;&quot;</span>
  43. <span class="sd">.. _data URLs:</span>
  44. <span class="sd"> https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URIs</span>
  45. <span class="sd">&quot;&quot;&quot;</span>
  46. <span class="kn">from</span><span class="w"> </span><span class="nn">__future__</span><span class="w"> </span><span class="kn">import</span> <span class="n">annotations</span>
  47. <span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">TYPE_CHECKING</span>
  48. <span class="kn">from</span><span class="w"> </span><span class="nn">urllib.parse</span><span class="w"> </span><span class="kn">import</span> <span class="n">urlencode</span>
  49. <span class="kn">from</span><span class="w"> </span><span class="nn">lxml</span><span class="w"> </span><span class="kn">import</span> <span class="n">html</span>
  50. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
  51. <span class="n">eval_xpath</span><span class="p">,</span>
  52. <span class="n">eval_xpath_list</span><span class="p">,</span>
  53. <span class="n">eval_xpath_getindex</span><span class="p">,</span>
  54. <span class="n">extract_text</span><span class="p">,</span>
  55. <span class="p">)</span>
  56. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.engines.google</span><span class="w"> </span><span class="kn">import</span> <span class="n">fetch_traits</span> <span class="c1"># pylint: disable=unused-import</span>
  57. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.engines.google</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
  58. <span class="n">get_google_info</span><span class="p">,</span>
  59. <span class="n">time_range_dict</span><span class="p">,</span>
  60. <span class="n">filter_mapping</span><span class="p">,</span>
  61. <span class="n">suggestion_xpath</span><span class="p">,</span>
  62. <span class="n">detect_google_sorry</span><span class="p">,</span>
  63. <span class="n">ui_async</span><span class="p">,</span>
  64. <span class="n">parse_data_images</span><span class="p">,</span>
  65. <span class="p">)</span>
  66. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.enginelib.traits</span><span class="w"> </span><span class="kn">import</span> <span class="n">EngineTraits</span>
  67. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">get_embeded_stream_url</span>
  68. <span class="k">if</span> <span class="n">TYPE_CHECKING</span><span class="p">:</span>
  69. <span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
  70. <span class="n">logger</span><span class="p">:</span> <span class="n">logging</span><span class="o">.</span><span class="n">Logger</span>
  71. <span class="n">traits</span><span class="p">:</span> <span class="n">EngineTraits</span>
  72. <span class="c1"># about</span>
  73. <span class="n">about</span> <span class="o">=</span> <span class="p">{</span>
  74. <span class="s2">&quot;website&quot;</span><span class="p">:</span> <span class="s1">&#39;https://www.google.com&#39;</span><span class="p">,</span>
  75. <span class="s2">&quot;wikidata_id&quot;</span><span class="p">:</span> <span class="s1">&#39;Q219885&#39;</span><span class="p">,</span>
  76. <span class="s2">&quot;official_api_documentation&quot;</span><span class="p">:</span> <span class="s1">&#39;https://developers.google.com/custom-search&#39;</span><span class="p">,</span>
  77. <span class="s2">&quot;use_official_api&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
  78. <span class="s2">&quot;require_api_key&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
  79. <span class="s2">&quot;results&quot;</span><span class="p">:</span> <span class="s1">&#39;HTML&#39;</span><span class="p">,</span>
  80. <span class="p">}</span>
  81. <span class="c1"># engine dependent config</span>
  82. <span class="n">categories</span> <span class="o">=</span> <span class="p">[</span><span class="s1">&#39;videos&#39;</span><span class="p">,</span> <span class="s1">&#39;web&#39;</span><span class="p">]</span>
  83. <span class="n">paging</span> <span class="o">=</span> <span class="kc">True</span>
  84. <span class="n">max_page</span> <span class="o">=</span> <span class="mi">50</span>
  85. <span class="sd">&quot;&quot;&quot;`Google: max 50 pages`</span>
  86. <span class="sd">.. _Google: max 50 pages: https://github.com/searxng/searxng/issues/2982</span>
  87. <span class="sd">&quot;&quot;&quot;</span>
  88. <span class="n">language_support</span> <span class="o">=</span> <span class="kc">True</span>
  89. <span class="n">time_range_support</span> <span class="o">=</span> <span class="kc">True</span>
  90. <span class="n">safesearch</span> <span class="o">=</span> <span class="kc">True</span>
  91. <div class="viewcode-block" id="request">
  92. <a class="viewcode-back" href="../../../dev/engines/online/google.html#searx.engines.google_videos.request">[docs]</a>
  93. <span class="k">def</span><span class="w"> </span><span class="nf">request</span><span class="p">(</span><span class="n">query</span><span class="p">,</span> <span class="n">params</span><span class="p">):</span>
  94. <span class="w"> </span><span class="sd">&quot;&quot;&quot;Google-Video search request&quot;&quot;&quot;</span>
  95. <span class="n">google_info</span> <span class="o">=</span> <span class="n">get_google_info</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">traits</span><span class="p">)</span>
  96. <span class="n">start</span> <span class="o">=</span> <span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s1">&#39;pageno&#39;</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="mi">10</span>
  97. <span class="n">query_url</span> <span class="o">=</span> <span class="p">(</span>
  98. <span class="s1">&#39;https://&#39;</span>
  99. <span class="o">+</span> <span class="n">google_info</span><span class="p">[</span><span class="s1">&#39;subdomain&#39;</span><span class="p">]</span>
  100. <span class="o">+</span> <span class="s1">&#39;/search&#39;</span>
  101. <span class="o">+</span> <span class="s2">&quot;?&quot;</span>
  102. <span class="o">+</span> <span class="n">urlencode</span><span class="p">(</span>
  103. <span class="p">{</span>
  104. <span class="s1">&#39;q&#39;</span><span class="p">:</span> <span class="n">query</span><span class="p">,</span>
  105. <span class="s1">&#39;tbm&#39;</span><span class="p">:</span> <span class="s2">&quot;vid&quot;</span><span class="p">,</span>
  106. <span class="s1">&#39;start&#39;</span><span class="p">:</span> <span class="mi">10</span> <span class="o">*</span> <span class="n">params</span><span class="p">[</span><span class="s1">&#39;pageno&#39;</span><span class="p">],</span>
  107. <span class="o">**</span><span class="n">google_info</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">],</span>
  108. <span class="s1">&#39;asearch&#39;</span><span class="p">:</span> <span class="s1">&#39;arc&#39;</span><span class="p">,</span>
  109. <span class="s1">&#39;async&#39;</span><span class="p">:</span> <span class="n">ui_async</span><span class="p">(</span><span class="n">start</span><span class="p">),</span>
  110. <span class="p">}</span>
  111. <span class="p">)</span>
  112. <span class="p">)</span>
  113. <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s1">&#39;time_range&#39;</span><span class="p">]</span> <span class="ow">in</span> <span class="n">time_range_dict</span><span class="p">:</span>
  114. <span class="n">query_url</span> <span class="o">+=</span> <span class="s1">&#39;&amp;&#39;</span> <span class="o">+</span> <span class="n">urlencode</span><span class="p">({</span><span class="s1">&#39;tbs&#39;</span><span class="p">:</span> <span class="s1">&#39;qdr:&#39;</span> <span class="o">+</span> <span class="n">time_range_dict</span><span class="p">[</span><span class="n">params</span><span class="p">[</span><span class="s1">&#39;time_range&#39;</span><span class="p">]]})</span>
  115. <span class="k">if</span> <span class="s1">&#39;safesearch&#39;</span> <span class="ow">in</span> <span class="n">params</span><span class="p">:</span>
  116. <span class="n">query_url</span> <span class="o">+=</span> <span class="s1">&#39;&amp;&#39;</span> <span class="o">+</span> <span class="n">urlencode</span><span class="p">({</span><span class="s1">&#39;safe&#39;</span><span class="p">:</span> <span class="n">filter_mapping</span><span class="p">[</span><span class="n">params</span><span class="p">[</span><span class="s1">&#39;safesearch&#39;</span><span class="p">]]})</span>
  117. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;url&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">query_url</span>
  118. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">google_info</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">]</span>
  119. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">google_info</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">])</span>
  120. <span class="k">return</span> <span class="n">params</span></div>
  121. <div class="viewcode-block" id="response">
  122. <a class="viewcode-back" href="../../../dev/engines/online/google.html#searx.engines.google_videos.response">[docs]</a>
  123. <span class="k">def</span><span class="w"> </span><span class="nf">response</span><span class="p">(</span><span class="n">resp</span><span class="p">):</span>
  124. <span class="w"> </span><span class="sd">&quot;&quot;&quot;Get response from google&#39;s search request&quot;&quot;&quot;</span>
  125. <span class="n">results</span> <span class="o">=</span> <span class="p">[]</span>
  126. <span class="n">detect_google_sorry</span><span class="p">(</span><span class="n">resp</span><span class="p">)</span>
  127. <span class="n">data_image_map</span> <span class="o">=</span> <span class="n">parse_data_images</span><span class="p">(</span><span class="n">resp</span><span class="o">.</span><span class="n">text</span><span class="p">)</span>
  128. <span class="c1"># convert the text to dom</span>
  129. <span class="n">dom</span> <span class="o">=</span> <span class="n">html</span><span class="o">.</span><span class="n">fromstring</span><span class="p">(</span><span class="n">resp</span><span class="o">.</span><span class="n">text</span><span class="p">)</span>
  130. <span class="c1"># parse results</span>
  131. <span class="k">for</span> <span class="n">result</span> <span class="ow">in</span> <span class="n">eval_xpath_list</span><span class="p">(</span><span class="n">dom</span><span class="p">,</span> <span class="s1">&#39;//div[contains(@class, &quot;g &quot;)]&#39;</span><span class="p">):</span>
  132. <span class="n">thumbnail</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//img/@src&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
  133. <span class="k">if</span> <span class="n">thumbnail</span><span class="p">:</span>
  134. <span class="k">if</span> <span class="n">thumbnail</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s1">&#39;data:image&#39;</span><span class="p">):</span>
  135. <span class="n">img_id</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//img/@id&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
  136. <span class="k">if</span> <span class="n">img_id</span><span class="p">:</span>
  137. <span class="n">thumbnail</span> <span class="o">=</span> <span class="n">data_image_map</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">img_id</span><span class="p">)</span>
  138. <span class="k">else</span><span class="p">:</span>
  139. <span class="n">thumbnail</span> <span class="o">=</span> <span class="kc">None</span>
  140. <span class="n">title</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//a/h3[1]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
  141. <span class="n">url</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//a/h3[1]/../@href&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
  142. <span class="n">c_node</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//div[contains(@class, &quot;ITZIwc&quot;)]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
  143. <span class="n">content</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">c_node</span><span class="p">)</span>
  144. <span class="n">pub_info</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">eval_xpath</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//div[contains(@class, &quot;gqF9jc&quot;)]&#39;</span><span class="p">))</span>
  145. <span class="n">results</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
  146. <span class="p">{</span>
  147. <span class="s1">&#39;url&#39;</span><span class="p">:</span> <span class="n">url</span><span class="p">,</span>
  148. <span class="s1">&#39;title&#39;</span><span class="p">:</span> <span class="n">title</span><span class="p">,</span>
  149. <span class="s1">&#39;content&#39;</span><span class="p">:</span> <span class="n">content</span><span class="p">,</span>
  150. <span class="s1">&#39;author&#39;</span><span class="p">:</span> <span class="n">pub_info</span><span class="p">,</span>
  151. <span class="s1">&#39;thumbnail&#39;</span><span class="p">:</span> <span class="n">thumbnail</span><span class="p">,</span>
  152. <span class="s1">&#39;iframe_src&#39;</span><span class="p">:</span> <span class="n">get_embeded_stream_url</span><span class="p">(</span><span class="n">url</span><span class="p">),</span>
  153. <span class="s1">&#39;template&#39;</span><span class="p">:</span> <span class="s1">&#39;videos.html&#39;</span><span class="p">,</span>
  154. <span class="p">}</span>
  155. <span class="p">)</span>
  156. <span class="c1"># parse suggestion</span>
  157. <span class="k">for</span> <span class="n">suggestion</span> <span class="ow">in</span> <span class="n">eval_xpath_list</span><span class="p">(</span><span class="n">dom</span><span class="p">,</span> <span class="n">suggestion_xpath</span><span class="p">):</span>
  158. <span class="c1"># append suggestion</span>
  159. <span class="n">results</span><span class="o">.</span><span class="n">append</span><span class="p">({</span><span class="s1">&#39;suggestion&#39;</span><span class="p">:</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">suggestion</span><span class="p">)})</span>
  160. <span class="k">return</span> <span class="n">results</span></div>
  161. </pre></div>
  162. <div class="clearer"></div>
  163. </div>
  164. </div>
  165. </div>
  166. <span id="sidebar-top"></span>
  167. <div class="sphinxsidebar" role="navigation" aria-label="Main">
  168. <div class="sphinxsidebarwrapper">
  169. <p class="logo"><a href="../../../index.html">
  170. <img class="logo" src="../../../_static/searxng-wordmark.svg" alt="Logo of SearXNG"/>
  171. </a></p>
  172. <h3><a href="../../../index.html">Table of Contents</a></h3>
  173. <ul>
  174. <li class="toctree-l1"><a class="reference internal" href="../../../user/index.html">User information</a></li>
  175. <li class="toctree-l1"><a class="reference internal" href="../../../own-instance.html">Why use a private instance?</a></li>
  176. <li class="toctree-l1"><a class="reference internal" href="../../../admin/index.html">Administrator documentation</a></li>
  177. <li class="toctree-l1"><a class="reference internal" href="../../../dev/index.html">Developer documentation</a></li>
  178. <li class="toctree-l1"><a class="reference internal" href="../../../utils/index.html">DevOps tooling box</a></li>
  179. <li class="toctree-l1"><a class="reference internal" href="../../../src/index.html">Source-Code</a></li>
  180. </ul>
  181. <h3>Project Links</h3>
  182. <ul>
  183. <li><a href="https://github.com/searxng/searxng/tree/master">Source</a>
  184. <li><a href="https://github.com/searxng/searxng/wiki">Wiki</a>
  185. <li><a href="https://searx.space">Public instances</a>
  186. <li><a href="https://github.com/searxng/searxng/issues">Issue Tracker</a>
  187. </ul><h3>Navigation</h3>
  188. <ul>
  189. <li><a href="../../../index.html">Overview</a>
  190. <ul>
  191. <li><a href="../../index.html">Module code</a>
  192. <ul>
  193. <li><a href="../engines.html">searx.engines</a>
  194. </ul>
  195. </li></ul>
  196. </li>
  197. </ul>
  198. </li>
  199. </ul>
  200. <search id="searchbox" style="display: none" role="search">
  201. <h3 id="searchlabel">Quick search</h3>
  202. <div class="searchformwrapper">
  203. <form class="search" action="../../../search.html" method="get">
  204. <input type="text" name="q" aria-labelledby="searchlabel" autocomplete="off" autocorrect="off" autocapitalize="off" spellcheck="false"/>
  205. <input type="submit" value="Go" />
  206. </form>
  207. </div>
  208. </search>
  209. <script>document.getElementById('searchbox').style.display = "block"</script>
  210. </div>
  211. </div>
  212. <div class="clearer"></div>
  213. </div>
  214. <div class="footer" role="contentinfo">
  215. &#169; Copyright SearXNG team.
  216. </div>
  217. </body>
  218. </html>