SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled.
Python bingbravedegoogleduckduckgogooglemetasearchprivacypythonqwantsearchsearch-enginesearxsearxngstartpageyahoo
[fix] engines: update brave images/videos parser and news xpath
vojkovic8b01679e8fc0
diff --git a/searx/engines/brave.py b/searx/engines/brave.pyindex de0a108b0..8d0994227 100644--- a/searx/engines/brave.py+++ b/searx/engines/brave.py@@ -248,13 +248,13 @@ def extract_json_data(text: str) -> dict[str, t.Any]:# node_ids: [0, 19],# data: [{type:"data",data: .... ["q","goggles_id"],route:1,url:1}}]# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^- text = text[text.index("<script") : text.index("</script")]- if not text:- raise ValueError("can't find JS/JSON data in the given text")+ # form: null,+ # error: null+ # });start = text.index("data: [{")- end = text.rindex("}}]")- js_obj_str = text[start:end]- js_obj_str = "{" + js_obj_str + "}}]}"+ newline = text.index("\n", start)+ end = text.rindex("}}]", start, newline)+ js_obj_str = "{" + text[start:end] + "}}]}"# js_obj_str = js_obj_str.replace("\xa0", "") # remove ASCII for # js_obj_str = js_obj_str.replace(r"\u003C", "<").replace(r"\u003c", "<") # fix broken HTML tags in stringsjson_str = js_obj_str_to_json_str(js_obj_str)@@ -354,14 +354,14 @@ def _parse_news(resp: SXNG_Response) -> EngineResults:res = EngineResults()dom = html.fromstring(resp.text)- for result in eval_xpath_list(dom, "//div[contains(@class, 'results')]//div[@data-type='news']"):- url = eval_xpath_getindex(result, ".//a[contains(@class, 'result-header')]/@href", 0, default=None)+ for result in eval_xpath_list(dom, "//div[@data-type='news']"):+ url = eval_xpath_getindex(result, ".//a/@href", 0, default=None)if url is None:continue- title = eval_xpath_list(result, ".//span[contains(@class, 'snippet-title')]")- content = eval_xpath_list(result, ".//p[contains(@class, 'desc')]")- thumbnail = eval_xpath_getindex(result, ".//div[contains(@class, 'image-wrapper')]//img/@src", 0, default="")+ title = eval_xpath_list(result, ".//div[contains(@class, 'title')]")+ content = eval_xpath_list(result, ".//div[contains(@class, 'description')]")+ thumbnail = eval_xpath_getindex(result, ".//a[contains(@class, 'thumbnail')]//img/@src", 0, default="")item = res.types.LegacyResult(template="default.html",diff --git a/searx/utils.py b/searx/utils.pyindex 9c70c8cc4..b03f8f057 100644--- a/searx/utils.py+++ b/searx/utils.py@@ -735,13 +735,13 @@ def js_obj_str_to_json_str(js_obj_str: str) -> str:if in_string == "'":p = p.replace('"', r'\"')parts[i] = p- # deal with the sequence blackslash then quote- # since js_obj_str splits on quote, we detect this case:- # * the previous part ends with a black slash- # * the current part is a single quote- # when detected the blackslash is removed on the previous part+ # drop a trailing \ that was escaping the quote+ # leave it if it has been escaped twice as a literal i.e. two \ and ' in a rowif blackslash_just_before and p[:1] == "'":- parts[i - 1] = parts[i - 1][:-1]+ prev = parts[i - 1]+ num_backslashes = len(prev) - len(prev.rstrip("\\"))+ if num_backslashes % 2 == 1:+ parts[i - 1] = prev[:-1]elif in_string is None and p in ('"', "'", "`"):# we are not in string but p is string delimiter