The previous implementation could not distinguish a CAPTCHA response from an
ordinary result list. In the previous implementation a CAPTCHA was taken as a
result list where no items are in.
DDG does not block IPs. Instead, a CAPTCHA wall is placed in front of request
on a dubious request.
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This patch adds an additional *isinstance* check within the ast parser to check
for float along with int, fixing the underlying issue.
Co-Authored: Markus Heiser <markus.heiser@darmarit.de>
Improve region and language detection / all locale
Testing has shown the following behaviour for the different
default and empty values of Mojeeks parameters:
| param | idx | value | behaviour |
| -------- | --- | ------ | ------------------------- |
| region | 0 | '' | detect region based on IP |
| region | 1 | 'none' | all regions |
| language | 0 | '' | all languages |
Without this patch the Gitea Search Engine is only partially compatible with
modern gitea or forgejo:
- Fixing some JSON Fields
- Using Repository Avatar when Available
To Verify My results you can look at the Modern API doc and results, its
available on all Gitea and Forgejo instance by Default. Heres an Search API
result of Mine:
- https://git.euph.dev/api/v1/repos/search?q=ccna
All favicons implementations have been documented and moved to the Python
package:
searx.favicons
There is a configuration (based on Pydantic) for the favicons and all its
components:
searx.favicons.config
A solution for caching favicons has been implemented:
searx.favicon.cache
If the favicon is already in the cache, the returned URL is a data URL [1]
(something like `data:image/png;base64,...`). By generating a data url from
the FaviconCache, additional HTTP roundtripps via the favicon_proxy are saved:
favicons.proxy.favicon_url
The favicon proxy service now sets a HTTP header "Cache-Control: max-age=...":
favicons.proxy.favicon_proxy
The resolvers now also provide the mime type (data, mime):
searx.favicon.resolvers
[1] https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URLs
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
- for tests which perform the same arrange/act/assert pattern but with different
data, the data portion has been moved to the ``paramaterized.expand`` fields
- for monolithic tests which performed multiple arrange/act/asserts,
they have been broken up into different unit tests.
- when possible, change generic assert statements to more concise
asserts (i.e. ``assertIsNone``)
This work ultimately is focused on creating smaller and more concise tests.
While paramaterized may make adding new configurations for existing tests
easier, that is just a beneficial side effect. The main benefit is that smaller
tests are easier to reason about, meaning they are easier to debug when they
start failing. This improves the developer experience in debugging what went
wrong when refactoring the project.
Total number of tests went from 192 -> 259; or, broke apart larger tests into 69
more concise ones.
add Cloudflare AI Gateway engine
add settings for Cloudflare AI Gateway engine
set utf8 encode for data, fix non english char cause 500 error
format json data
fixed indentation and config format error
fix line-length limitation in CI
reformatted code for CI
reformatted code for CI
limit system prompts to less 120 chars
cleanup unused variable & format code
In its previous implementation, the macro ``checkbox_onoff_reversed`` always
created an ``aria-labelledby`` attribute, even if there was no descriptive tag
with the generated ID (used as the value of the ``aria-labelledby``).
Before this patch, the Nu-HTML-Checker [1] reported 255 issues of this type::
The aria-labelledby attribute must point to an element in the same document. (255)
[1] https://validator.w3.org/nu/
Signed-off-by: Markus <markus@venom.fritz.box>
The ``aria-labelledby`` [1] attribute identifies the element that labels the
element it is applied to. The templates ``infinite_scroll.html`` and
``search_on_category_select.html`` define a ``aria-labelledby`` at the <input>
tag but miss the id in the <div> with the description.
[1] https://developer.mozilla.org/en-US/docs/Web/Accessibility/ARIA/Attributes/aria-labelledby
Signed-off-by: Markus <markus@venom.fritz.box>
So far a CAPTCHA was not recognized in the response of the qwant engine and a
SearxEngineAPIException was raised by mistake. With this patch a CAPTCHA
redirect is recognized and the correct SearxEngineCaptchaException is raised.
Closes: https://github.com/searxng/searxng/issues/3806
Signed-off-by: Markus <markus@venom.fritz.box>
This patch fixes a bug reported by CI "Fetch traits" [1] (brave) and improves
other fetch traits functions (google, annas_archive & radio_browser).
brave:
File "/home/runner/work/searxng/searxng/searx/engines/brave.py", line 434, in fetch_traits
sxng_tag = region_tag(babel.Locale.parse(ui_lang, sep='-'))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/searxng/searxng/searx/locales.py", line 155, in region_tag
Error: raise ValueError('%s missed a territory')
google:
change ERROR message about unknow UI language to INFO message
radio_browser:
country_list contains duplicates that differ only in upper/lower case
annas_archive:
for better diff; sort the persistence of the traits
[1] https://github.com/searxng/searxng/actions/runs/10606312371/job/29433352518#step:6:41
Signed-off-by: Markus <markus@venom.fritz.box>
* [feat] engine: add right dao
* [enh] right dao engine: allow additional classes
Allow additional classes while parsing to prevent the engine from breaking in the future if additional classes are added to the elements.
Co-authored-by: Bnyro <bnyro@tutanota.com>
---------
Co-authored-by: Bnyro <bnyro@tutanota.com>
The user_agent attribute of the Flask request object is an instance of
the werkzeug.user_agent.UserAgent class.
This will fix the following error of the self_info plugin:
> ERROR:searx.plugins.self_info: Exception while calling post_search
> Traceback (most recent call last):
> File "searx/plugins/__init__.py", line 203, in call
> ret = getattr(plugin, plugin_type)(*args, **kwargs)
> File "searx/plugins/self_info.py", line 31, in post_search
> search.result_container.answers['user-agent'] = {'answer': gettext('Your user-agent is: ') + ua}
> TypeError: can only concatenate str (not "UserAgent") to str
We do not want to show the user-agent information from the duckduckgo
zero click info. This is the user-agent used by searxng and not the
user-agent used by the user.
This was already done for the IP address in:
0fb3f0e4ae
Changes made to tineye engine:
1. Importing logging if TYPE_CHECKING is enabled
2. Remove unecessary try-catch around json parsing the response, as this
masked the original error and had no immediate benefit
3. Improve error handling explicitely for status code 422 and 400
upfront, deferring json_parsing only for these status codes and
successful status codes
4. Unit test all new applicable changes to ensure compatability
You have to copy and paste the query without using the keyboard to replicate the
issue. As soon as you press the keyboard the cross appears. [1]
- [1] https://github.com/searxng/searxng/issues/3725#issuecomment-2282655272
Reported-by: @Immortality-IMT in [1]
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
The successor is “hostname plugin” from PR:
- https://github.com/searxng/searxng/pull/3463
---
Revert "[refactor] hostnames plugin: add fallback for old hostname_replace plugin"
This reverts commit f5eb56b63f.
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
Google underlines words inside of answers that can be clicked to show
additional definitions. These definitions inside the answer were not
correctly handled and ended up in the middle of the answer text. With
this fix, the extra definitions are stripped from the answer shown by
the frontend.
