brave.html 62 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616
  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.brave &#8212; SearXNG Documentation (2025.5.1+1a1628149)</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=cc9fc81d"></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="2025.5.1+1a1628149" 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 (2025.5.1+1a1628149)</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.brave</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.brave</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;Brave supports the categories listed in :py:obj:`brave_category` (General,</span>
  39. <span class="sd">news, videos, images). The support of :py:obj:`paging` and :py:obj:`time range</span>
  40. <span class="sd">&lt;time_range_support&gt;` is limited (see remarks).</span>
  41. <span class="sd">Configured ``brave`` engines:</span>
  42. <span class="sd">.. code:: yaml</span>
  43. <span class="sd"> - name: brave</span>
  44. <span class="sd"> engine: brave</span>
  45. <span class="sd"> ...</span>
  46. <span class="sd"> brave_category: search</span>
  47. <span class="sd"> time_range_support: true</span>
  48. <span class="sd"> paging: true</span>
  49. <span class="sd"> - name: brave.images</span>
  50. <span class="sd"> engine: brave</span>
  51. <span class="sd"> ...</span>
  52. <span class="sd"> brave_category: images</span>
  53. <span class="sd"> - name: brave.videos</span>
  54. <span class="sd"> engine: brave</span>
  55. <span class="sd"> ...</span>
  56. <span class="sd"> brave_category: videos</span>
  57. <span class="sd"> - name: brave.news</span>
  58. <span class="sd"> engine: brave</span>
  59. <span class="sd"> ...</span>
  60. <span class="sd"> brave_category: news</span>
  61. <span class="sd"> - name: brave.goggles</span>
  62. <span class="sd"> time_range_support: true</span>
  63. <span class="sd"> paging: true</span>
  64. <span class="sd"> ...</span>
  65. <span class="sd"> brave_category: goggles</span>
  66. <span class="sd">.. _brave regions:</span>
  67. <span class="sd">Brave regions</span>
  68. <span class="sd">=============</span>
  69. <span class="sd">Brave uses two-digit tags for the regions like ``ca`` while SearXNG deals with</span>
  70. <span class="sd">locales. To get a mapping, all *officiat de-facto* languages of the Brave</span>
  71. <span class="sd">region are mapped to regions in SearXNG (see :py:obj:`babel</span>
  72. <span class="sd">&lt;babel.languages.get_official_languages&gt;`):</span>
  73. <span class="sd">.. code:: python</span>
  74. <span class="sd"> &quot;regions&quot;: {</span>
  75. <span class="sd"> ..</span>
  76. <span class="sd"> &quot;en-CA&quot;: &quot;ca&quot;,</span>
  77. <span class="sd"> &quot;fr-CA&quot;: &quot;ca&quot;,</span>
  78. <span class="sd"> ..</span>
  79. <span class="sd"> }</span>
  80. <span class="sd">.. note::</span>
  81. <span class="sd"> The language (aka region) support of Brave&#39;s index is limited to very basic</span>
  82. <span class="sd"> languages. The search results for languages like Chinese or Arabic are of</span>
  83. <span class="sd"> low quality.</span>
  84. <span class="sd">.. _brave googles:</span>
  85. <span class="sd">Brave Goggles</span>
  86. <span class="sd">=============</span>
  87. <span class="sd">.. _list of Goggles: https://search.brave.com/goggles/discover</span>
  88. <span class="sd">.. _Goggles Whitepaper: https://brave.com/static-assets/files/goggles.pdf</span>
  89. <span class="sd">.. _Goggles Quickstart: https://github.com/brave/goggles-quickstart</span>
  90. <span class="sd">Goggles allow you to choose, alter, or extend the ranking of Brave Search</span>
  91. <span class="sd">results (`Goggles Whitepaper`_). Goggles are openly developed by the community</span>
  92. <span class="sd">of Brave Search users.</span>
  93. <span class="sd">Select from the `list of Goggles`_ people have published, or create your own</span>
  94. <span class="sd">(`Goggles Quickstart`_).</span>
  95. <span class="sd">.. _brave languages:</span>
  96. <span class="sd">Brave languages</span>
  97. <span class="sd">===============</span>
  98. <span class="sd">Brave&#39;s language support is limited to the UI (menus, area local notations,</span>
  99. <span class="sd">etc). Brave&#39;s index only seems to support a locale, but it does not seem to</span>
  100. <span class="sd">support any languages in its index. The choice of available languages is very</span>
  101. <span class="sd">small (and its not clear to me where the difference in UI is when switching</span>
  102. <span class="sd">from en-us to en-ca or en-gb).</span>
  103. <span class="sd">In the :py:obj:`EngineTraits object &lt;searx.enginelib.traits.EngineTraits&gt;` the</span>
  104. <span class="sd">UI languages are stored in a custom field named ``ui_lang``:</span>
  105. <span class="sd">.. code:: python</span>
  106. <span class="sd"> &quot;custom&quot;: {</span>
  107. <span class="sd"> &quot;ui_lang&quot;: {</span>
  108. <span class="sd"> &quot;ca&quot;: &quot;ca&quot;,</span>
  109. <span class="sd"> &quot;de-DE&quot;: &quot;de-de&quot;,</span>
  110. <span class="sd"> &quot;en-CA&quot;: &quot;en-ca&quot;,</span>
  111. <span class="sd"> &quot;en-GB&quot;: &quot;en-gb&quot;,</span>
  112. <span class="sd"> &quot;en-US&quot;: &quot;en-us&quot;,</span>
  113. <span class="sd"> &quot;es&quot;: &quot;es&quot;,</span>
  114. <span class="sd"> &quot;fr-CA&quot;: &quot;fr-ca&quot;,</span>
  115. <span class="sd"> &quot;fr-FR&quot;: &quot;fr-fr&quot;,</span>
  116. <span class="sd"> &quot;ja-JP&quot;: &quot;ja-jp&quot;,</span>
  117. <span class="sd"> &quot;pt-BR&quot;: &quot;pt-br&quot;,</span>
  118. <span class="sd"> &quot;sq-AL&quot;: &quot;sq-al&quot;</span>
  119. <span class="sd"> }</span>
  120. <span class="sd"> },</span>
  121. <span class="sd">Implementations</span>
  122. <span class="sd">===============</span>
  123. <span class="sd">&quot;&quot;&quot;</span>
  124. <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">Any</span><span class="p">,</span> <span class="n">TYPE_CHECKING</span>
  125. <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="p">(</span>
  126. <span class="n">urlencode</span><span class="p">,</span>
  127. <span class="n">urlparse</span><span class="p">,</span>
  128. <span class="p">)</span>
  129. <span class="kn">from</span><span class="w"> </span><span class="nn">dateutil</span><span class="w"> </span><span class="kn">import</span> <span class="n">parser</span>
  130. <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>
  131. <span class="kn">from</span><span class="w"> </span><span class="nn">searx</span><span class="w"> </span><span class="kn">import</span> <span class="n">locales</span>
  132. <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>
  133. <span class="n">extr</span><span class="p">,</span>
  134. <span class="n">extract_text</span><span class="p">,</span>
  135. <span class="n">eval_xpath</span><span class="p">,</span>
  136. <span class="n">eval_xpath_list</span><span class="p">,</span>
  137. <span class="n">eval_xpath_getindex</span><span class="p">,</span>
  138. <span class="n">js_variable_to_python</span><span class="p">,</span>
  139. <span class="n">get_embeded_stream_url</span><span class="p">,</span>
  140. <span class="p">)</span>
  141. <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>
  142. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.result_types</span><span class="w"> </span><span class="kn">import</span> <span class="n">EngineResults</span>
  143. <span class="k">if</span> <span class="n">TYPE_CHECKING</span><span class="p">:</span>
  144. <span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
  145. <span class="n">logger</span><span class="p">:</span> <span class="n">logging</span><span class="o">.</span><span class="n">Logger</span>
  146. <span class="n">traits</span><span class="p">:</span> <span class="n">EngineTraits</span>
  147. <span class="n">about</span> <span class="o">=</span> <span class="p">{</span>
  148. <span class="s2">&quot;website&quot;</span><span class="p">:</span> <span class="s1">&#39;https://search.brave.com/&#39;</span><span class="p">,</span>
  149. <span class="s2">&quot;wikidata_id&quot;</span><span class="p">:</span> <span class="s1">&#39;Q22906900&#39;</span><span class="p">,</span>
  150. <span class="s2">&quot;official_api_documentation&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
  151. <span class="s2">&quot;use_official_api&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
  152. <span class="s2">&quot;require_api_key&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
  153. <span class="s2">&quot;results&quot;</span><span class="p">:</span> <span class="s1">&#39;HTML&#39;</span><span class="p">,</span>
  154. <span class="p">}</span>
  155. <span class="n">base_url</span> <span class="o">=</span> <span class="s2">&quot;https://search.brave.com/&quot;</span>
  156. <span class="n">categories</span> <span class="o">=</span> <span class="p">[]</span>
  157. <span class="n">brave_category</span> <span class="o">=</span> <span class="s1">&#39;search&#39;</span>
  158. <span class="n">Goggles</span> <span class="o">=</span> <span class="n">Any</span>
  159. <span class="sd">&quot;&quot;&quot;Brave supports common web-search, videos, images, news, and goggles search.</span>
  160. <span class="sd">- ``search``: Common WEB search</span>
  161. <span class="sd">- ``videos``: search for videos</span>
  162. <span class="sd">- ``images``: search for images</span>
  163. <span class="sd">- ``news``: search for news</span>
  164. <span class="sd">- ``goggles``: Common WEB search with custom rules</span>
  165. <span class="sd">&quot;&quot;&quot;</span>
  166. <span class="n">brave_spellcheck</span> <span class="o">=</span> <span class="kc">False</span>
  167. <span class="sd">&quot;&quot;&quot;Brave supports some kind of spell checking. When activated, Brave tries to</span>
  168. <span class="sd">fix typos, e.g. it searches for ``food`` when the user queries for ``fooh``. In</span>
  169. <span class="sd">the UI of Brave the user gets warned about this, since we can not warn the user</span>
  170. <span class="sd">in SearXNG, the spellchecking is disabled by default.</span>
  171. <span class="sd">&quot;&quot;&quot;</span>
  172. <span class="n">send_accept_language_header</span> <span class="o">=</span> <span class="kc">True</span>
  173. <span class="n">paging</span> <span class="o">=</span> <span class="kc">False</span>
  174. <span class="sd">&quot;&quot;&quot;Brave only supports paging in :py:obj:`brave_category` ``search`` (UI</span>
  175. <span class="sd">category All) and in the goggles category.&quot;&quot;&quot;</span>
  176. <span class="n">max_page</span> <span class="o">=</span> <span class="mi">10</span>
  177. <span class="sd">&quot;&quot;&quot;Tested 9 pages maximum (``&amp;offset=8``), to be save max is set to 10. Trying</span>
  178. <span class="sd">to do more won&#39;t return any result and you will most likely be flagged as a bot.</span>
  179. <span class="sd">&quot;&quot;&quot;</span>
  180. <span class="n">safesearch</span> <span class="o">=</span> <span class="kc">True</span>
  181. <span class="n">safesearch_map</span> <span class="o">=</span> <span class="p">{</span><span class="mi">2</span><span class="p">:</span> <span class="s1">&#39;strict&#39;</span><span class="p">,</span> <span class="mi">1</span><span class="p">:</span> <span class="s1">&#39;moderate&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">:</span> <span class="s1">&#39;off&#39;</span><span class="p">}</span> <span class="c1"># cookie: safesearch=off</span>
  182. <span class="n">time_range_support</span> <span class="o">=</span> <span class="kc">False</span>
  183. <span class="sd">&quot;&quot;&quot;Brave only supports time-range in :py:obj:`brave_category` ``search`` (UI</span>
  184. <span class="sd">category All) and in the goggles category.&quot;&quot;&quot;</span>
  185. <span class="n">time_range_map</span> <span class="o">=</span> <span class="p">{</span>
  186. <span class="s1">&#39;day&#39;</span><span class="p">:</span> <span class="s1">&#39;pd&#39;</span><span class="p">,</span>
  187. <span class="s1">&#39;week&#39;</span><span class="p">:</span> <span class="s1">&#39;pw&#39;</span><span class="p">,</span>
  188. <span class="s1">&#39;month&#39;</span><span class="p">:</span> <span class="s1">&#39;pm&#39;</span><span class="p">,</span>
  189. <span class="s1">&#39;year&#39;</span><span class="p">:</span> <span class="s1">&#39;py&#39;</span><span class="p">,</span>
  190. <span class="p">}</span>
  191. <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>
  192. <span class="c1"># Don&#39;t accept br encoding / see https://github.com/searxng/searxng/pull/1787</span>
  193. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">][</span><span class="s1">&#39;Accept-Encoding&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;gzip, deflate&#39;</span>
  194. <span class="n">args</span> <span class="o">=</span> <span class="p">{</span>
  195. <span class="s1">&#39;q&#39;</span><span class="p">:</span> <span class="n">query</span><span class="p">,</span>
  196. <span class="s1">&#39;source&#39;</span><span class="p">:</span> <span class="s1">&#39;web&#39;</span><span class="p">,</span>
  197. <span class="p">}</span>
  198. <span class="k">if</span> <span class="n">brave_spellcheck</span><span class="p">:</span>
  199. <span class="n">args</span><span class="p">[</span><span class="s1">&#39;spellcheck&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;1&#39;</span>
  200. <span class="k">if</span> <span class="n">brave_category</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">&#39;search&#39;</span><span class="p">,</span> <span class="s1">&#39;goggles&#39;</span><span class="p">):</span>
  201. <span class="k">if</span> <span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;pageno&#39;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
  202. <span class="n">args</span><span class="p">[</span><span class="s1">&#39;offset&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">params</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;pageno&#39;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span>
  203. <span class="k">if</span> <span class="n">time_range_map</span><span class="o">.</span><span class="n">get</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>
  204. <span class="n">args</span><span class="p">[</span><span class="s1">&#39;tf&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">time_range_map</span><span class="o">.</span><span class="n">get</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>
  205. <span class="k">if</span> <span class="n">brave_category</span> <span class="o">==</span> <span class="s1">&#39;goggles&#39;</span><span class="p">:</span>
  206. <span class="n">args</span><span class="p">[</span><span class="s1">&#39;goggles_id&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Goggles</span>
  207. <span class="n">params</span><span class="p">[</span><span class="s2">&quot;url&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">base_url</span><span class="si">}{</span><span class="n">brave_category</span><span class="si">}</span><span class="s2">?</span><span class="si">{</span><span class="n">urlencode</span><span class="p">(</span><span class="n">args</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
  208. <span class="c1"># set properties in the cookies</span>
  209. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">][</span><span class="s1">&#39;safesearch&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">safesearch_map</span><span class="o">.</span><span class="n">get</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> <span class="s1">&#39;off&#39;</span><span class="p">)</span>
  210. <span class="c1"># the useLocation is IP based, we use cookie &#39;country&#39; for the region</span>
  211. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">][</span><span class="s1">&#39;useLocation&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;0&#39;</span>
  212. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">][</span><span class="s1">&#39;summarizer&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;0&#39;</span>
  213. <span class="n">engine_region</span> <span class="o">=</span> <span class="n">traits</span><span class="o">.</span><span class="n">get_region</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s1">&#39;searxng_locale&#39;</span><span class="p">],</span> <span class="s1">&#39;all&#39;</span><span class="p">)</span>
  214. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">][</span><span class="s1">&#39;country&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">engine_region</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;-&#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="n">lower</span><span class="p">()</span> <span class="c1"># type: ignore</span>
  215. <span class="n">ui_lang</span> <span class="o">=</span> <span class="n">locales</span><span class="o">.</span><span class="n">get_engine_locale</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s1">&#39;searxng_locale&#39;</span><span class="p">],</span> <span class="n">traits</span><span class="o">.</span><span class="n">custom</span><span class="p">[</span><span class="s2">&quot;ui_lang&quot;</span><span class="p">],</span> <span class="s1">&#39;en-us&#39;</span><span class="p">)</span>
  216. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">][</span><span class="s1">&#39;ui_lang&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">ui_lang</span>
  217. <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="s2">&quot;cookies </span><span class="si">%s</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">params</span><span class="p">[</span><span class="s1">&#39;cookies&#39;</span><span class="p">])</span>
  218. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">][</span><span class="s1">&#39;Sec-Fetch-Dest&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;document&quot;</span>
  219. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">][</span><span class="s1">&#39;Sec-Fetch-Mode&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;navigate&quot;</span>
  220. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">][</span><span class="s1">&#39;Sec-Fetch-Site&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;same-origin&quot;</span>
  221. <span class="n">params</span><span class="p">[</span><span class="s1">&#39;headers&#39;</span><span class="p">][</span><span class="s1">&#39;Sec-Fetch-User&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;?1&quot;</span>
  222. <span class="k">def</span><span class="w"> </span><span class="nf">_extract_published_date</span><span class="p">(</span><span class="n">published_date_raw</span><span class="p">):</span>
  223. <span class="k">if</span> <span class="n">published_date_raw</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
  224. <span class="k">return</span> <span class="kc">None</span>
  225. <span class="k">try</span><span class="p">:</span>
  226. <span class="k">return</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">published_date_raw</span><span class="p">)</span>
  227. <span class="k">except</span> <span class="n">parser</span><span class="o">.</span><span class="n">ParserError</span><span class="p">:</span>
  228. <span class="k">return</span> <span class="kc">None</span>
  229. <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> <span class="o">-&gt;</span> <span class="n">EngineResults</span><span class="p">:</span>
  230. <span class="k">if</span> <span class="n">brave_category</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">&#39;search&#39;</span><span class="p">,</span> <span class="s1">&#39;goggles&#39;</span><span class="p">):</span>
  231. <span class="k">return</span> <span class="n">_parse_search</span><span class="p">(</span><span class="n">resp</span><span class="p">)</span>
  232. <span class="k">if</span> <span class="n">brave_category</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">&#39;news&#39;</span><span class="p">):</span>
  233. <span class="k">return</span> <span class="n">_parse_news</span><span class="p">(</span><span class="n">resp</span><span class="p">)</span>
  234. <span class="c1"># Example script source containing the data:</span>
  235. <span class="c1">#</span>
  236. <span class="c1"># kit.start(app, element, {</span>
  237. <span class="c1"># node_ids: [0, 19],</span>
  238. <span class="c1"># data: [{type:&quot;data&quot;,data: .... [&quot;q&quot;,&quot;goggles_id&quot;],route:1,url:1}}]</span>
  239. <span class="c1"># ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^</span>
  240. <span class="n">js_object</span> <span class="o">=</span> <span class="s2">&quot;[{&quot;</span> <span class="o">+</span> <span class="n">extr</span><span class="p">(</span><span class="n">resp</span><span class="o">.</span><span class="n">text</span><span class="p">,</span> <span class="s2">&quot;data: [{&quot;</span><span class="p">,</span> <span class="s2">&quot;}}],&quot;</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;}}]&quot;</span>
  241. <span class="n">json_data</span> <span class="o">=</span> <span class="n">js_variable_to_python</span><span class="p">(</span><span class="n">js_object</span><span class="p">)</span>
  242. <span class="c1"># json_data is a list and at the second position (0,1) in this list we find the &quot;response&quot; data we need ..</span>
  243. <span class="n">json_resp</span> <span class="o">=</span> <span class="n">json_data</span><span class="p">[</span><span class="mi">1</span><span class="p">][</span><span class="s1">&#39;data&#39;</span><span class="p">][</span><span class="s1">&#39;body&#39;</span><span class="p">][</span><span class="s1">&#39;response&#39;</span><span class="p">]</span>
  244. <span class="k">if</span> <span class="n">brave_category</span> <span class="o">==</span> <span class="s1">&#39;images&#39;</span><span class="p">:</span>
  245. <span class="k">return</span> <span class="n">_parse_images</span><span class="p">(</span><span class="n">json_resp</span><span class="p">)</span>
  246. <span class="k">if</span> <span class="n">brave_category</span> <span class="o">==</span> <span class="s1">&#39;videos&#39;</span><span class="p">:</span>
  247. <span class="k">return</span> <span class="n">_parse_videos</span><span class="p">(</span><span class="n">json_resp</span><span class="p">)</span>
  248. <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported brave category: </span><span class="si">{</span><span class="n">brave_category</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
  249. <span class="k">def</span><span class="w"> </span><span class="nf">_parse_search</span><span class="p">(</span><span class="n">resp</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EngineResults</span><span class="p">:</span>
  250. <span class="n">result_list</span> <span class="o">=</span> <span class="n">EngineResults</span><span class="p">()</span>
  251. <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>
  252. <span class="c1"># I doubt that Brave is still providing the &quot;answer&quot; class / I haven&#39;t seen</span>
  253. <span class="c1"># answers in brave for a long time.</span>
  254. <span class="n">answer_tag</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">dom</span><span class="p">,</span> <span class="s1">&#39;//div[@class=&quot;answer&quot;]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
  255. <span class="k">if</span> <span class="n">answer_tag</span><span class="p">:</span>
  256. <span class="n">url</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">dom</span><span class="p">,</span> <span class="s1">&#39;//div[@id=&quot;featured_snippet&quot;]/a[@class=&quot;result-header&quot;]/@href&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
  257. <span class="n">answer</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">answer_tag</span><span class="p">)</span>
  258. <span class="k">if</span> <span class="n">answer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
  259. <span class="n">result_list</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">result_list</span><span class="o">.</span><span class="n">types</span><span class="o">.</span><span class="n">Answer</span><span class="p">(</span><span class="n">answer</span><span class="o">=</span><span class="n">answer</span><span class="p">,</span> <span class="n">url</span><span class="o">=</span><span class="n">url</span><span class="p">))</span>
  260. <span class="c1"># xpath_results = &#39;//div[contains(@class, &quot;snippet fdb&quot;) and @data-type=&quot;web&quot;]&#39;</span>
  261. <span class="n">xpath_results</span> <span class="o">=</span> <span class="s1">&#39;//div[contains(@class, &quot;snippet &quot;)]&#39;</span>
  262. <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="n">xpath_results</span><span class="p">):</span>
  263. <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[contains(@class, &quot;h&quot;)]/@href&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
  264. <span class="n">title_tag</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span>
  265. <span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//a[contains(@class, &quot;h&quot;)]//div[contains(@class, &quot;title&quot;)]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span>
  266. <span class="p">)</span>
  267. <span class="k">if</span> <span class="n">url</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">title_tag</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">urlparse</span><span class="p">(</span><span class="n">url</span><span class="p">)</span><span class="o">.</span><span class="n">netloc</span><span class="p">:</span> <span class="c1"># partial url likely means it&#39;s an ad</span>
  268. <span class="k">continue</span>
  269. <span class="n">content</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span>
  270. <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;snippet-description&quot;)]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span>
  271. <span class="p">)</span> <span class="c1"># type: ignore</span>
  272. <span class="n">pub_date_raw</span> <span class="o">=</span> <span class="n">eval_xpath</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;substring-before(.//div[contains(@class, &quot;snippet-description&quot;)], &quot;-&quot;)&#39;</span><span class="p">)</span>
  273. <span class="n">pub_date</span> <span class="o">=</span> <span class="n">_extract_published_date</span><span class="p">(</span><span class="n">pub_date_raw</span><span class="p">)</span>
  274. <span class="k">if</span> <span class="n">pub_date</span> <span class="ow">and</span> <span class="n">content</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="n">pub_date_raw</span><span class="p">):</span>
  275. <span class="n">content</span> <span class="o">=</span> <span class="n">content</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="n">pub_date_raw</span><span class="p">)</span><span class="o">.</span><span class="n">strip</span><span class="p">(</span><span class="s2">&quot;- </span><span class="se">\n\t</span><span class="s2">&quot;</span><span class="p">)</span>
  276. <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[contains(@class, &quot;thumb&quot;)]/@src&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span>
  277. <span class="n">item</span> <span class="o">=</span> <span class="p">{</span>
  278. <span class="s1">&#39;url&#39;</span><span class="p">:</span> <span class="n">url</span><span class="p">,</span>
  279. <span class="s1">&#39;title&#39;</span><span class="p">:</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">title_tag</span><span class="p">),</span>
  280. <span class="s1">&#39;content&#39;</span><span class="p">:</span> <span class="n">content</span><span class="p">,</span>
  281. <span class="s1">&#39;publishedDate&#39;</span><span class="p">:</span> <span class="n">pub_date</span><span class="p">,</span>
  282. <span class="s1">&#39;thumbnail&#39;</span><span class="p">:</span> <span class="n">thumbnail</span><span class="p">,</span>
  283. <span class="p">}</span>
  284. <span class="n">video_tag</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span>
  285. <span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//div[contains(@class, &quot;video-snippet&quot;) and @data-macro=&quot;video&quot;]&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span>
  286. <span class="p">)</span>
  287. <span class="k">if</span> <span class="n">video_tag</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
  288. <span class="c1"># In my tests a video tag in the WEB search was most often not a</span>
  289. <span class="c1"># video, except the ones from youtube ..</span>
  290. <span class="n">iframe_src</span> <span class="o">=</span> <span class="n">get_embeded_stream_url</span><span class="p">(</span><span class="n">url</span><span class="p">)</span>
  291. <span class="k">if</span> <span class="n">iframe_src</span><span class="p">:</span>
  292. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;iframe_src&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">iframe_src</span>
  293. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;template&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;videos.html&#39;</span>
  294. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">video_tag</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="n">default</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span>
  295. <span class="n">pub_date_raw</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span>
  296. <span class="n">eval_xpath</span><span class="p">(</span><span class="n">video_tag</span><span class="p">,</span> <span class="s1">&#39;.//div[contains(@class, &quot;snippet-attributes&quot;)]/div/text()&#39;</span><span class="p">)</span>
  297. <span class="p">)</span>
  298. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;publishedDate&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">_extract_published_date</span><span class="p">(</span><span class="n">pub_date_raw</span><span class="p">)</span>
  299. <span class="k">else</span><span class="p">:</span>
  300. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">eval_xpath_getindex</span><span class="p">(</span><span class="n">video_tag</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="n">default</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span>
  301. <span class="n">result_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">item</span><span class="p">)</span>
  302. <span class="k">return</span> <span class="n">result_list</span>
  303. <span class="k">def</span><span class="w"> </span><span class="nf">_parse_news</span><span class="p">(</span><span class="n">resp</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EngineResults</span><span class="p">:</span>
  304. <span class="n">result_list</span> <span class="o">=</span> <span class="n">EngineResults</span><span class="p">()</span>
  305. <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>
  306. <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;results&quot;)]//div[@data-type=&quot;news&quot;]&#39;</span><span class="p">):</span>
  307. <span class="c1"># import pdb</span>
  308. <span class="c1"># pdb.set_trace()</span>
  309. <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[contains(@class, &quot;result-header&quot;)]/@href&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
  310. <span class="k">if</span> <span class="n">url</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
  311. <span class="k">continue</span>
  312. <span class="n">title</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">eval_xpath_list</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//span[contains(@class, &quot;snippet-title&quot;)]&#39;</span><span class="p">))</span>
  313. <span class="n">content</span> <span class="o">=</span> <span class="n">extract_text</span><span class="p">(</span><span class="n">eval_xpath_list</span><span class="p">(</span><span class="n">result</span><span class="p">,</span> <span class="s1">&#39;.//p[contains(@class, &quot;desc&quot;)]&#39;</span><span class="p">))</span>
  314. <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;.//div[contains(@class, &quot;image-wrapper&quot;)]//img/@src&#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span>
  315. <span class="n">item</span> <span class="o">=</span> <span class="p">{</span>
  316. <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="n">url</span><span class="p">,</span>
  317. <span class="s2">&quot;title&quot;</span><span class="p">:</span> <span class="n">title</span><span class="p">,</span>
  318. <span class="s2">&quot;content&quot;</span><span class="p">:</span> <span class="n">content</span><span class="p">,</span>
  319. <span class="s2">&quot;thumbnail&quot;</span><span class="p">:</span> <span class="n">thumbnail</span><span class="p">,</span>
  320. <span class="p">}</span>
  321. <span class="n">result_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">item</span><span class="p">)</span>
  322. <span class="k">return</span> <span class="n">result_list</span>
  323. <span class="k">def</span><span class="w"> </span><span class="nf">_parse_images</span><span class="p">(</span><span class="n">json_resp</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EngineResults</span><span class="p">:</span>
  324. <span class="n">result_list</span> <span class="o">=</span> <span class="n">EngineResults</span><span class="p">()</span>
  325. <span class="k">for</span> <span class="n">result</span> <span class="ow">in</span> <span class="n">json_resp</span><span class="p">[</span><span class="s2">&quot;results&quot;</span><span class="p">]:</span>
  326. <span class="n">item</span> <span class="o">=</span> <span class="p">{</span>
  327. <span class="s1">&#39;url&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;url&#39;</span><span class="p">],</span>
  328. <span class="s1">&#39;title&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;title&#39;</span><span class="p">],</span>
  329. <span class="s1">&#39;content&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;description&#39;</span><span class="p">],</span>
  330. <span class="s1">&#39;template&#39;</span><span class="p">:</span> <span class="s1">&#39;images.html&#39;</span><span class="p">,</span>
  331. <span class="s1">&#39;resolution&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;properties&#39;</span><span class="p">][</span><span class="s1">&#39;format&#39;</span><span class="p">],</span>
  332. <span class="s1">&#39;source&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;source&#39;</span><span class="p">],</span>
  333. <span class="s1">&#39;img_src&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;properties&#39;</span><span class="p">][</span><span class="s1">&#39;url&#39;</span><span class="p">],</span>
  334. <span class="s1">&#39;thumbnail_src&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">][</span><span class="s1">&#39;src&#39;</span><span class="p">],</span>
  335. <span class="p">}</span>
  336. <span class="n">result_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">item</span><span class="p">)</span>
  337. <span class="k">return</span> <span class="n">result_list</span>
  338. <span class="k">def</span><span class="w"> </span><span class="nf">_parse_videos</span><span class="p">(</span><span class="n">json_resp</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EngineResults</span><span class="p">:</span>
  339. <span class="n">result_list</span> <span class="o">=</span> <span class="n">EngineResults</span><span class="p">()</span>
  340. <span class="k">for</span> <span class="n">result</span> <span class="ow">in</span> <span class="n">json_resp</span><span class="p">[</span><span class="s2">&quot;results&quot;</span><span class="p">]:</span>
  341. <span class="n">url</span> <span class="o">=</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;url&#39;</span><span class="p">]</span>
  342. <span class="n">item</span> <span class="o">=</span> <span class="p">{</span>
  343. <span class="s1">&#39;url&#39;</span><span class="p">:</span> <span class="n">url</span><span class="p">,</span>
  344. <span class="s1">&#39;title&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;title&#39;</span><span class="p">],</span>
  345. <span class="s1">&#39;content&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;description&#39;</span><span class="p">],</span>
  346. <span class="s1">&#39;template&#39;</span><span class="p">:</span> <span class="s1">&#39;videos.html&#39;</span><span class="p">,</span>
  347. <span class="s1">&#39;length&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;video&#39;</span><span class="p">][</span><span class="s1">&#39;duration&#39;</span><span class="p">],</span>
  348. <span class="s1">&#39;duration&#39;</span><span class="p">:</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;video&#39;</span><span class="p">][</span><span class="s1">&#39;duration&#39;</span><span class="p">],</span>
  349. <span class="s1">&#39;publishedDate&#39;</span><span class="p">:</span> <span class="n">_extract_published_date</span><span class="p">(</span><span class="n">result</span><span class="p">[</span><span class="s1">&#39;age&#39;</span><span class="p">]),</span>
  350. <span class="p">}</span>
  351. <span class="k">if</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
  352. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">result</span><span class="p">[</span><span class="s1">&#39;thumbnail&#39;</span><span class="p">][</span><span class="s1">&#39;src&#39;</span><span class="p">]</span>
  353. <span class="n">iframe_src</span> <span class="o">=</span> <span class="n">get_embeded_stream_url</span><span class="p">(</span><span class="n">url</span><span class="p">)</span>
  354. <span class="k">if</span> <span class="n">iframe_src</span><span class="p">:</span>
  355. <span class="n">item</span><span class="p">[</span><span class="s1">&#39;iframe_src&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">iframe_src</span>
  356. <span class="n">result_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">item</span><span class="p">)</span>
  357. <span class="k">return</span> <span class="n">result_list</span>
  358. <div class="viewcode-block" id="fetch_traits">
  359. <a class="viewcode-back" href="../../../dev/engines/online/brave.html#searx.engines.brave.fetch_traits">[docs]</a>
  360. <span class="k">def</span><span class="w"> </span><span class="nf">fetch_traits</span><span class="p">(</span><span class="n">engine_traits</span><span class="p">:</span> <span class="n">EngineTraits</span><span class="p">):</span>
  361. <span class="w"> </span><span class="sd">&quot;&quot;&quot;Fetch :ref:`languages &lt;brave languages&gt;` and :ref:`regions &lt;brave</span>
  362. <span class="sd"> regions&gt;` from Brave.&quot;&quot;&quot;</span>
  363. <span class="c1"># pylint: disable=import-outside-toplevel, too-many-branches</span>
  364. <span class="kn">import</span><span class="w"> </span><span class="nn">babel.languages</span>
  365. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.locales</span><span class="w"> </span><span class="kn">import</span> <span class="n">region_tag</span><span class="p">,</span> <span class="n">language_tag</span>
  366. <span class="kn">from</span><span class="w"> </span><span class="nn">searx.network</span><span class="w"> </span><span class="kn">import</span> <span class="n">get</span> <span class="c1"># see https://github.com/searxng/searxng/issues/762</span>
  367. <span class="n">engine_traits</span><span class="o">.</span><span class="n">custom</span><span class="p">[</span><span class="s2">&quot;ui_lang&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">{}</span>
  368. <span class="n">headers</span> <span class="o">=</span> <span class="p">{</span>
  369. <span class="s1">&#39;Accept-Encoding&#39;</span><span class="p">:</span> <span class="s1">&#39;gzip, deflate&#39;</span><span class="p">,</span>
  370. <span class="p">}</span>
  371. <span class="n">lang_map</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;no&#39;</span><span class="p">:</span> <span class="s1">&#39;nb&#39;</span><span class="p">}</span> <span class="c1"># norway</span>
  372. <span class="c1"># languages (UI)</span>
  373. <span class="n">resp</span> <span class="o">=</span> <span class="n">get</span><span class="p">(</span><span class="s1">&#39;https://search.brave.com/settings&#39;</span><span class="p">,</span> <span class="n">headers</span><span class="o">=</span><span class="n">headers</span><span class="p">)</span>
  374. <span class="k">if</span> <span class="ow">not</span> <span class="n">resp</span><span class="o">.</span><span class="n">ok</span><span class="p">:</span> <span class="c1"># type: ignore</span>
  375. <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;ERROR: response from Brave is not OK.&quot;</span><span class="p">)</span>
  376. <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> <span class="c1"># type: ignore</span>
  377. <span class="k">for</span> <span class="n">option</span> <span class="ow">in</span> <span class="n">dom</span><span class="o">.</span><span class="n">xpath</span><span class="p">(</span><span class="s1">&#39;//section//option[@value=&quot;en-us&quot;]/../option&#39;</span><span class="p">):</span>
  378. <span class="n">ui_lang</span> <span class="o">=</span> <span class="n">option</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;value&#39;</span><span class="p">)</span>
  379. <span class="k">try</span><span class="p">:</span>
  380. <span class="n">l</span> <span class="o">=</span> <span class="n">babel</span><span class="o">.</span><span class="n">Locale</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">ui_lang</span><span class="p">,</span> <span class="n">sep</span><span class="o">=</span><span class="s1">&#39;-&#39;</span><span class="p">)</span>
  381. <span class="k">if</span> <span class="n">l</span><span class="o">.</span><span class="n">territory</span><span class="p">:</span>
  382. <span class="n">sxng_tag</span> <span class="o">=</span> <span class="n">region_tag</span><span class="p">(</span><span class="n">babel</span><span class="o">.</span><span class="n">Locale</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">ui_lang</span><span class="p">,</span> <span class="n">sep</span><span class="o">=</span><span class="s1">&#39;-&#39;</span><span class="p">))</span>
  383. <span class="k">else</span><span class="p">:</span>
  384. <span class="n">sxng_tag</span> <span class="o">=</span> <span class="n">language_tag</span><span class="p">(</span><span class="n">babel</span><span class="o">.</span><span class="n">Locale</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">ui_lang</span><span class="p">,</span> <span class="n">sep</span><span class="o">=</span><span class="s1">&#39;-&#39;</span><span class="p">))</span>
  385. <span class="k">except</span> <span class="n">babel</span><span class="o">.</span><span class="n">UnknownLocaleError</span><span class="p">:</span>
  386. <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;ERROR: can&#39;t determine babel locale of Brave&#39;s (UI) language </span><span class="si">%s</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">ui_lang</span><span class="p">)</span>
  387. <span class="k">continue</span>
  388. <span class="n">conflict</span> <span class="o">=</span> <span class="n">engine_traits</span><span class="o">.</span><span class="n">custom</span><span class="p">[</span><span class="s2">&quot;ui_lang&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">sxng_tag</span><span class="p">)</span>
  389. <span class="k">if</span> <span class="n">conflict</span><span class="p">:</span>
  390. <span class="k">if</span> <span class="n">conflict</span> <span class="o">!=</span> <span class="n">ui_lang</span><span class="p">:</span>
  391. <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;CONFLICT: babel </span><span class="si">%s</span><span class="s2"> --&gt; </span><span class="si">%s</span><span class="s2">, </span><span class="si">%s</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">sxng_tag</span><span class="p">,</span> <span class="n">conflict</span><span class="p">,</span> <span class="n">ui_lang</span><span class="p">))</span>
  392. <span class="k">continue</span>
  393. <span class="n">engine_traits</span><span class="o">.</span><span class="n">custom</span><span class="p">[</span><span class="s2">&quot;ui_lang&quot;</span><span class="p">][</span><span class="n">sxng_tag</span><span class="p">]</span> <span class="o">=</span> <span class="n">ui_lang</span>
  394. <span class="c1"># search regions of brave</span>
  395. <span class="n">resp</span> <span class="o">=</span> <span class="n">get</span><span class="p">(</span><span class="s1">&#39;https://cdn.search.brave.com/serp/v2/_app/immutable/chunks/parameters.734c106a.js&#39;</span><span class="p">,</span> <span class="n">headers</span><span class="o">=</span><span class="n">headers</span><span class="p">)</span>
  396. <span class="k">if</span> <span class="ow">not</span> <span class="n">resp</span><span class="o">.</span><span class="n">ok</span><span class="p">:</span> <span class="c1"># type: ignore</span>
  397. <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;ERROR: response from Brave is not OK.&quot;</span><span class="p">)</span>
  398. <span class="n">country_js</span> <span class="o">=</span> <span class="n">resp</span><span class="o">.</span><span class="n">text</span><span class="p">[</span><span class="n">resp</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">index</span><span class="p">(</span><span class="s2">&quot;options:{all&quot;</span><span class="p">)</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="s1">&#39;options:&#39;</span><span class="p">)</span> <span class="p">:]</span> <span class="c1"># type: ignore</span>
  399. <span class="n">country_js</span> <span class="o">=</span> <span class="n">country_js</span><span class="p">[:</span> <span class="n">country_js</span><span class="o">.</span><span class="n">index</span><span class="p">(</span><span class="s2">&quot;},k={default&quot;</span><span class="p">)]</span>
  400. <span class="n">country_tags</span> <span class="o">=</span> <span class="n">js_variable_to_python</span><span class="p">(</span><span class="n">country_js</span><span class="p">)</span>
  401. <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">country_tags</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
  402. <span class="k">if</span> <span class="n">k</span> <span class="o">==</span> <span class="s1">&#39;all&#39;</span><span class="p">:</span>
  403. <span class="n">engine_traits</span><span class="o">.</span><span class="n">all_locale</span> <span class="o">=</span> <span class="s1">&#39;all&#39;</span>
  404. <span class="k">continue</span>
  405. <span class="n">country_tag</span> <span class="o">=</span> <span class="n">v</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span>
  406. <span class="c1"># add official languages of the country ..</span>
  407. <span class="k">for</span> <span class="n">lang_tag</span> <span class="ow">in</span> <span class="n">babel</span><span class="o">.</span><span class="n">languages</span><span class="o">.</span><span class="n">get_official_languages</span><span class="p">(</span><span class="n">country_tag</span><span class="p">,</span> <span class="n">de_facto</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
  408. <span class="n">lang_tag</span> <span class="o">=</span> <span class="n">lang_map</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">lang_tag</span><span class="p">,</span> <span class="n">lang_tag</span><span class="p">)</span>
  409. <span class="n">sxng_tag</span> <span class="o">=</span> <span class="n">region_tag</span><span class="p">(</span><span class="n">babel</span><span class="o">.</span><span class="n">Locale</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s1">&#39;</span><span class="si">%s</span><span class="s1">_</span><span class="si">%s</span><span class="s1">&#39;</span> <span class="o">%</span> <span class="p">(</span><span class="n">lang_tag</span><span class="p">,</span> <span class="n">country_tag</span><span class="o">.</span><span class="n">upper</span><span class="p">())))</span>
  410. <span class="c1"># print(&quot;%-20s: %s &lt;-- %s&quot; % (v[&#39;label&#39;], country_tag, sxng_tag))</span>
  411. <span class="n">conflict</span> <span class="o">=</span> <span class="n">engine_traits</span><span class="o">.</span><span class="n">regions</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">sxng_tag</span><span class="p">)</span>
  412. <span class="k">if</span> <span class="n">conflict</span><span class="p">:</span>
  413. <span class="k">if</span> <span class="n">conflict</span> <span class="o">!=</span> <span class="n">country_tag</span><span class="p">:</span>
  414. <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;CONFLICT: babel </span><span class="si">%s</span><span class="s2"> --&gt; </span><span class="si">%s</span><span class="s2">, </span><span class="si">%s</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">sxng_tag</span><span class="p">,</span> <span class="n">conflict</span><span class="p">,</span> <span class="n">country_tag</span><span class="p">))</span>
  415. <span class="k">continue</span>
  416. <span class="n">engine_traits</span><span class="o">.</span><span class="n">regions</span><span class="p">[</span><span class="n">sxng_tag</span><span class="p">]</span> <span class="o">=</span> <span class="n">country_tag</span></div>
  417. </pre></div>
  418. <div class="clearer"></div>
  419. </div>
  420. </div>
  421. </div>
  422. <span id="sidebar-top"></span>
  423. <div class="sphinxsidebar" role="navigation" aria-label="Main">
  424. <div class="sphinxsidebarwrapper">
  425. <p class="logo"><a href="../../../index.html">
  426. <img class="logo" src="../../../_static/searxng-wordmark.svg" alt="Logo of SearXNG"/>
  427. </a></p>
  428. <h3><a href="../../../index.html">Table of Contents</a></h3>
  429. <ul>
  430. <li class="toctree-l1"><a class="reference internal" href="../../../user/index.html">User information</a></li>
  431. <li class="toctree-l1"><a class="reference internal" href="../../../own-instance.html">Why use a private instance?</a></li>
  432. <li class="toctree-l1"><a class="reference internal" href="../../../admin/index.html">Administrator documentation</a></li>
  433. <li class="toctree-l1"><a class="reference internal" href="../../../dev/index.html">Developer documentation</a></li>
  434. <li class="toctree-l1"><a class="reference internal" href="../../../utils/index.html">DevOps tooling box</a></li>
  435. <li class="toctree-l1"><a class="reference internal" href="../../../src/index.html">Source-Code</a></li>
  436. </ul>
  437. <h3>Project Links</h3>
  438. <ul>
  439. <li><a href="https://github.com/searxng/searxng/tree/master">Source</a>
  440. <li><a href="https://github.com/searxng/searxng/wiki">Wiki</a>
  441. <li><a href="https://searx.space">Public instances</a>
  442. <li><a href="https://github.com/searxng/searxng/issues">Issue Tracker</a>
  443. </ul><h3>Navigation</h3>
  444. <ul>
  445. <li><a href="../../../index.html">Overview</a>
  446. <ul>
  447. <li><a href="../../index.html">Module code</a>
  448. <ul>
  449. <li><a href="../engines.html">searx.engines</a>
  450. </ul>
  451. </li></ul>
  452. </li>
  453. </ul>
  454. </li>
  455. </ul>
  456. <search id="searchbox" style="display: none" role="search">
  457. <h3 id="searchlabel">Quick search</h3>
  458. <div class="searchformwrapper">
  459. <form class="search" action="../../../search.html" method="get">
  460. <input type="text" name="q" aria-labelledby="searchlabel" autocomplete="off" autocorrect="off" autocapitalize="off" spellcheck="false"/>
  461. <input type="submit" value="Go" />
  462. </form>
  463. </div>
  464. </search>
  465. <script>document.getElementById('searchbox').style.display = "block"</script>
  466. </div>
  467. </div>
  468. <div class="clearer"></div>
  469. </div>
  470. <div class="footer" role="contentinfo">
  471. &#169; Copyright SearXNG team.
  472. </div>
  473. </body>
  474. </html>