jandieman/searxng
2
0
Fork 0
mirror of https://github.com/searxng/searxng.git synced 2026-07-29 19:31:35 +00:00
Code Issues Projects Releases Wiki Activity

Compare commits

base: jandieman:ci-data-update_ahmia_blacklist.py
Branches Tags
jandieman:master jandieman:ci-data-update_engine_traits.py jandieman:ci-data-update_ahmia_blacklist.py jandieman:ci-data-update_firefox_version.py jandieman:gh-pages jandieman:translations_update jandieman:translations jandieman:update_data_update_engine_descriptions.py jandieman:update_data_update_engine_traits.py
..
compare: jandieman:d8dd88add91583f0f1252fbc8b8a38e5725214e0
Branches Tags
jandieman:ci-data-update_engine_traits.py jandieman:ci-data-update_ahmia_blacklist.py jandieman:ci-data-update_firefox_version.py jandieman:gh-pages jandieman:master jandieman:translations_update jandieman:translations jandieman:update_data_update_engine_descriptions.py jandieman:update_data_update_engine_traits.py

1 Commits
ci-data-up ... d8dd88add9

Author SHA1 Message Date
dependabot[bot]
d8dd88add9 [upd] pypi: Update docutils requirement from >=0.21.2 to >=0.23 
Updates the requirements on [docutils](https://github.com/rtfd/recommonmark) to permit the latest version.
- [Changelog](https://github.com/readthedocs/recommonmark/blob/master/CHANGELOG.md)
- [Commits](https://github.com/rtfd/recommonmark/commits)

---
updated-dependencies:
- dependency-name: docutils
  dependency-version: '0.23'
  dependency-type: direct:development
...

Signed-off-by: dependabot[bot] <support@github.com>
 2026-05-30 09:35:01 +00:00
332 changed files with 11557 additions and 15142 deletions
Expand all files Collapse all files

163
.dir-locals-template.el Normal file
View File

@@ -0,0 +1,163 @@
;;; .dir-locals.el
;;
;; Per-Directory Local Variables:
;; https://www.gnu.org/software/emacs/manual/html_node/emacs/Directory-Variables.html
;;
;; For full fledge developer tools install emacs packages:
;;
;; M-x package-install ...
;;
;; magit gitconfig
;; nvm lsp-mode lsp-pyright lsp-eslint
;; pyvenv pylint pip-requirements
;; jinja2-mode
;; json-mode
;; company company-jedi company-quickhelp company-shell
;; realgud
;; sphinx-doc markdown-mode graphviz-dot-mode
;; apache-mode nginx-mode
;;
;; To setup a developer environment, build target::
;;
;; $ make node.env.dev pyenv.install
;;
;; Some buffer locals are referencing the project environment:
;;
;; - prj-root --> <repo>/
;; - nvm-dir --> <repo>/.nvm
;; - python-environment-directory --> <repo>/local
;; - python-environment-default-root-name --> py3
;; - python-shell-virtualenv-root --> <repo>/local/py3
;; When this variable is set with the path of the virtualenv to use,
;; `process-environment' and `exec-path' get proper values in order to run
;; shells inside the specified virtualenv, example::
;; (setq python-shell-virtualenv-root "/path/to/env/")
;; - python-shell-interpreter --> <repo>/local/py3/bin/python
;;
;; Python development:
;;
;; Jedi, flycheck & other python stuff should use the 'python-shell-interpreter'
;; from the local py3 environment.
;;
((nil
. ((fill-column . 80)
(indent-tabs-mode . nil)
(eval . (progn
(add-to-list 'auto-mode-alist '("\\.html\\'" . jinja2-mode))
;; project root folder is where the `.dir-locals.el' is located
(setq-local prj-root
(locate-dominating-file default-directory ".dir-locals.el"))
(setq-local python-environment-directory
(expand-file-name "./local" prj-root))
;; to get in use of NVM environment, install https://github.com/rejeep/nvm.el
(setq-local nvm-dir (expand-file-name "./.nvm" prj-root))
;; use nodejs from the (local) NVM environment (see nvm-dir)
(nvm-use-for-buffer)
(ignore-errors (require 'lsp))
(setq-local lsp-server-install-dir (car (cdr nvm-current-version)))
(setq-local lsp-enable-file-watchers nil)
;; use 'py3' environment as default
(setq-local python-environment-default-root-name
"py3")
(setq-local python-shell-virtualenv-root
(expand-file-name
python-environment-default-root-name python-environment-directory))
(setq-local python-shell-interpreter
(expand-file-name
"bin/python" python-shell-virtualenv-root))))))
(makefile-gmake-mode
. ((indent-tabs-mode . t)))
(yaml-mode
. ((eval . (progn
;; flycheck should use the local py3 environment
(setq-local flycheck-yaml-yamllint-executable
(expand-file-name "bin/yamllint" python-shell-virtualenv-root))
(setq-local flycheck-yamllintrc
(expand-file-name ".yamllint.yml" prj-root))
(flycheck-checker . yaml-yamllint)))))
(json-mode
. ((eval . (progn
(setq-local js-indent-level 4)
(flycheck-checker . json-python-json)))))
(js-mode
. ((eval . (progn
(ignore-errors (require 'lsp-eslint))
(setq-local js-indent-level 2)
;; flycheck should use the eslint checker from developer tools
(setq-local flycheck-javascript-eslint-executable
(expand-file-name "node_modules/.bin/eslint" prj-root))
;; (flycheck-mode)
(if (featurep 'lsp-eslint)
(lsp))
))))
(python-mode
. ((eval . (progn
(ignore-errors (require 'jedi-core))
(ignore-errors (require 'lsp-pyright))
(ignore-errors (sphinx-doc-mode))
(setq-local python-environment-virtualenv
(list (expand-file-name "bin/virtualenv" python-shell-virtualenv-root)
;;"--system-site-packages"
"--quiet"))
(setq-local pylint-command
(expand-file-name "bin/pylint" python-shell-virtualenv-root))
(if (featurep 'lsp-pyright)
(lsp))
;; pylint will find the '.pylintrc' file next to the CWD
;; https://pylint.readthedocs.io/en/latest/user_guide/run.html#command-line-options
(setq-local flycheck-pylintrc
".pylintrc")
;; flycheck & other python stuff should use the local py3 environment
(setq-local flycheck-python-pylint-executable
python-shell-interpreter)
;; use 'M-x jedi:show-setup-info' and 'M-x epc:controller' to inspect jedi server
;; https://tkf.github.io/emacs-jedi/latest/#jedi:environment-root -- You
;; can specify a full path instead of a name (relative path). In that case,
;; python-environment-directory is ignored and Python virtual environment
;; is created at the specified path.
(setq-local jedi:environment-root
python-shell-virtualenv-root)
;; https://tkf.github.io/emacs-jedi/latest/#jedi:server-command
(setq-local jedi:server-command
(list python-shell-interpreter
jedi:server-script))
;; jedi:environment-virtualenv --> see above 'python-environment-virtualenv'
;; is set buffer local! No need to setup jedi:environment-virtualenv:
;;
;; Virtualenv command to use. A list of string. If it is nil,
;; python-environment-virtualenv is used instead. You must set non-nil
;; value to jedi:environment-root in order to make this setting work.
;;
;; https://tkf.github.io/emacs-jedi/latest/#jedi:environment-virtualenv
;;
;; (setq-local jedi:environment-virtualenv
;; (list (expand-file-name "bin/virtualenv" python-shell-virtualenv-root)
;; "--python"
;; "/usr/bin/python3.4"
;; ))
))))
)

1
.dockerignore
View File

@@ -1,6 +1,5 @@
*
!container/*.template.*
!container/entrypoint.sh
!searx/**
!requirements*.txt

154
.github/workflows/container.yml vendored
View File

@@ -25,21 +25,25 @@ env:
jobs:
build:
if: |
github.event_name == 'workflow_dispatch'
|| (github.repository_owner == 'searxng' && github.event.workflow_run.conclusion == 'success')
if: github.repository_owner == 'searxng' || github.event_name == 'workflow_dispatch'
name: Build (${{ matrix.arch }})
runs-on: ${{ matrix.runner }}
runs-on: ${{ matrix.os }}
strategy:
fail-fast: false
matrix:
include:
- runner: ubuntu-26.04
arch: amd64
- runner: ubuntu-26.04-arm
arch: arm64
- runner: ubuntu-26.04-arm
arch: armv7
- arch: amd64
march: amd64
os: ubuntu-24.04
emulation: false
- arch: arm64
march: arm64
os: ubuntu-24.04-arm
emulation: false
- arch: armv7
march: arm64
os: ubuntu-24.04-arm
emulation: true
permissions:
packages: write
@@ -49,82 +53,109 @@ jobs:
git_url: ${{ steps.build.outputs.git_url }}
steps:
- name: Login to GHCR
uses: docker/login-action@06fb636fac595d6fb4b28a5dfcb21a6f5091859c # v4.5.0
with:
registry: "ghcr.io"
username: "${{ github.repository_owner }}"
password: "${{ secrets.GITHUB_TOKEN }}"
# yamllint disable rule:line-length
- name: Setup podman
env:
PODMAN_VERSION: "v5.7.1"
run: |
sudo apt-get purge -y podman runc crun conmon
curl -fsSLO "https://github.com/mgoltzsche/podman-static/releases/download/${{ env.PODMAN_VERSION }}/podman-linux-${{ matrix.march }}.tar.gz"
curl -fsSLO "https://github.com/mgoltzsche/podman-static/releases/download/${{ env.PODMAN_VERSION }}/podman-linux-${{ matrix.march }}.tar.gz.asc"
gpg --keyserver hkps://keyserver.ubuntu.com --recv-keys 0CCF102C4F95D89E583FF1D4F8B5AF50344BB503
gpg --batch --verify "podman-linux-${{ matrix.march }}.tar.gz.asc" "podman-linux-${{ matrix.march }}.tar.gz"
tar -xzf "podman-linux-${{ matrix.march }}.tar.gz"
sudo cp -rfv ./podman-linux-${{ matrix.march }}/etc/. /etc/
sudo cp -rfv ./podman-linux-${{ matrix.march }}/usr/. /usr/
sudo sysctl -w kernel.apparmor_restrict_unprivileged_userns=0
# yamllint enable rule:line-length
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Setup QEMU
uses: docker/setup-qemu-action@96fe6ef7f33517b61c61be40b68a1882f3264fb8 # v4.2.0
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
ref: "${{ github.event.workflow_run.head_sha || github.sha }}"
persist-credentials: "false"
fetch-depth: "0"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-
path: "./local/"
- name: Get date
id: date
run: echo "date=$(date +'%Y%m%d')" >>$GITHUB_OUTPUT
- name: Setup cache container
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "container-${{ matrix.arch }}-${{ hashFiles('./requirements*.txt') }}"
key: "container-${{ matrix.arch }}-${{ steps.date.outputs.date }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
container-${{ matrix.arch }}-${{ steps.date.outputs.date }}-
container-${{ matrix.arch }}-
path: "/var/tmp/buildah-cache-*/*"
- name: Build
id: build
env:
OVERRIDE_ARCH: "${{ matrix.arch }}"
run: make container.build
- if: ${{ matrix.emulation }}
name: Setup QEMU
uses: docker/setup-qemu-action@06116385d9baf250c9f4dcb4858b16962ea869c3 # v4.1.0
test:
name: Test (${{ matrix.arch }})
runs-on: ${{ matrix.runner }}
needs: build
strategy:
fail-fast: false
matrix:
include:
- runner: ubuntu-26.04
arch: amd64
- runner: ubuntu-26.04-arm
arch: arm64
- runner: ubuntu-26.04-arm
arch: armv7
steps:
- name: Login to GHCR
uses: docker/login-action@06fb636fac595d6fb4b28a5dfcb21a6f5091859c # v4.5.0
uses: docker/login-action@650006c6eb7dba73a995cc03b0b2d7f5ca915bee # v4.2.0
with:
registry: "ghcr.io"
username: "${{ github.repository_owner }}"
password: "${{ secrets.GITHUB_TOKEN }}"
- name: Setup QEMU
uses: docker/setup-qemu-action@96fe6ef7f33517b61c61be40b68a1882f3264fb8 # v4.2.0
- name: Build
id: build
env:
OVERRIDE_ARCH: "${{ matrix.arch }}"
run: make podman.build
test:
name: Test (${{ matrix.arch }})
runs-on: ${{ matrix.os }}
needs: build
strategy:
fail-fast: false
matrix:
include:
- arch: amd64
os: ubuntu-24.04
emulation: false
- arch: arm64
os: ubuntu-24.04-arm
emulation: false
- arch: armv7
os: ubuntu-24.04-arm
emulation: true
steps:
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
ref: "${{ github.event.workflow_run.head_sha || github.sha }}"
persist-credentials: "false"
- if: ${{ matrix.emulation }}
name: Setup QEMU
uses: docker/setup-qemu-action@06116385d9baf250c9f4dcb4858b16962ea869c3 # v4.1.0
- name: Login to GHCR
uses: docker/login-action@650006c6eb7dba73a995cc03b0b2d7f5ca915bee # v4.2.0
with:
registry: "ghcr.io"
username: "${{ github.repository_owner }}"
password: "${{ secrets.GITHUB_TOKEN }}"
- name: Test
env:
OVERRIDE_ARCH: "${{ matrix.arch }}"
@@ -134,7 +165,7 @@ jobs:
release:
if: github.repository_owner == 'searxng' && github.ref_name == 'master'
name: Release
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
needs:
- build
- test
@@ -143,25 +174,24 @@ jobs:
packages: write
steps:
- name: Login to Docker Hub
uses: docker/login-action@06fb636fac595d6fb4b28a5dfcb21a6f5091859c # v4.5.0
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
registry: "docker.io"
username: "${{ secrets.DOCKER_USER }}"
password: "${{ secrets.DOCKER_TOKEN }}"
persist-credentials: "false"
- name: Login to GHCR
uses: docker/login-action@06fb636fac595d6fb4b28a5dfcb21a6f5091859c # v4.5.0
uses: docker/login-action@650006c6eb7dba73a995cc03b0b2d7f5ca915bee # v4.2.0
with:
registry: "ghcr.io"
username: "${{ github.repository_owner }}"
password: "${{ secrets.GITHUB_TOKEN }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
- name: Login to Docker Hub
uses: docker/login-action@650006c6eb7dba73a995cc03b0b2d7f5ca915bee # v4.2.0
with:
ref: "${{ github.event.workflow_run.head_sha || github.sha }}"
persist-credentials: "false"
registry: "docker.io"
username: "${{ secrets.DOCKER_USER }}"
password: "${{ secrets.DOCKER_TOKEN }}"
- name: Release
env:

23
.github/workflows/data-update.yml vendored
View File

@@ -21,7 +21,7 @@ jobs:
data:
if: github.repository_owner == 'searxng'
name: ${{ matrix.fetch }}
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
strategy:
fail-fast: false
matrix:
@@ -33,6 +33,7 @@ jobs:
- update_engine_traits.py
- update_wikidata_units.py
- update_engine_descriptions.py
- update_gsa_useragents.py
permissions:
contents: write
@@ -40,17 +41,17 @@ jobs:
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: "false"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
@@ -64,17 +65,23 @@ jobs:
run: V=1 ./manage pyenv.cmd python "./searxng_extra/update/${{ matrix.fetch }}"
- name: Create PR
id: cpr
uses: peter-evans/create-pull-request@5f6978faf089d4d20b00c7766989d076bb2fc7f1 # v8.1.1
with:
author: "searxng-bot <searxng-bot@users.noreply.github.com>"
committer: "searxng-bot <searxng-bot@users.noreply.github.com>"
title: "[mod] data: update searx.data - ${{ matrix.fetch }}"
commit-message: "[mod] data: update searx.data - ${{ matrix.fetch }}"
branch: "ci-data-${{ matrix.fetch }}"
title: "[data] update searx.data - ${{ matrix.fetch }}"
commit-message: "[data] update searx.data - ${{ matrix.fetch }}"
branch: "update_data_${{ matrix.fetch }}"
delete-branch: "true"
draft: "false"
signoff: "false"
body: |
Update searx.data - ${{ matrix.fetch }}
[data] update searx.data - ${{ matrix.fetch }}
labels: |
data
- name: Display information
run: |
echo "Pull Request Number - ${{ steps.cpr.outputs.pull-request-number }}"
echo "Pull Request URL - ${{ steps.cpr.outputs.pull-request-url }}"

15
.github/workflows/documentation.yml vendored
View File

@@ -25,39 +25,36 @@ jobs:
release:
if: github.repository_owner == 'searxng' || github.event_name == 'workflow_dispatch'
name: Release
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
permissions:
# for JamesIves/github-pages-deploy-action to push
contents: write
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: "false"
fetch-depth: "0"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-
path: "./local/"
- name: Setup dependencies
run: sudo ./utils/searxng.sh install buildhost
- name: Setup venv
run: make V=1 install
- name: Build documentation
run: make V=1 docs.html
run: make V=1 docs.clean docs.html
- if: github.ref_name == 'master'
name: Release
@@ -65,7 +62,7 @@ jobs:
with:
folder: "dist/docs"
branch: "gh-pages"
commit-message: "[mod] docs: build from commit ${{ github.sha }}"
commit-message: "[doc] build from commit ${{ github.sha }}"
# Automatically remove deleted files from the deploy branch
clean: "true"
single-commit: "true"

41
.github/workflows/integration.yml vendored
View File

@@ -23,7 +23,7 @@ env:
jobs:
test:
name: Python ${{ matrix.python-version }}
runs-on: ubuntu-26.04
runs-on: ubuntu-24.04
strategy:
matrix:
python-version:
@@ -34,17 +34,17 @@ jobs:
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ matrix.python-version }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: "false"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ matrix.python-version }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
@@ -59,40 +59,37 @@ jobs:
theme:
name: Theme
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Setup Node.js
uses: actions/setup-node@820762786026740c76f36085b0efc47a31fe5020 # v7.0.0
with:
node-version: "26"
check-latest: "true"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: "false"
- name: Setup Node.js
uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
with:
node-version-file: "./.nvmrc"
- name: Setup cache Node.js
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "nodejs-${{ runner.arch }}-${{ hashFiles('./.nvmrc', './package.json') }}"
path: "./client/simple/node_modules/"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-
path: "./local/"
- name: Setup cache Node.js
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
with:
key: "nodejs-${{ runner.arch }}-${{ hashFiles('**/package-lock.json') }}"
restore-keys: |
nodejs-${{ runner.arch }}-
path: "./client/simple/node_modules/"
- name: Setup venv
run: make V=1 install

34
.github/workflows/l10n.yml vendored
View File

@@ -26,27 +26,27 @@ env:
jobs:
update:
if: github.event.workflow_run.conclusion == 'success' && github.repository_owner == 'searxng'
if: github.repository_owner == 'searxng' && github.event.workflow_run.conclusion == 'success'
name: Update
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
permissions:
# For "make V=1 weblate.push.translations"
contents: write
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
token: "${{ secrets.WEBLATE_GITHUB_TOKEN }}"
fetch-depth: "0"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
@@ -59,7 +59,7 @@ jobs:
- name: Setup Weblate
run: |
mkdir -p ~/.config
echo "${{ secrets.WEBLATE_CONFIG }}" >~/.config/weblate
echo "${{ secrets.WEBLATE_CONFIG }}" > ~/.config/weblate
- name: Setup Git
run: |
@@ -74,7 +74,7 @@ jobs:
github.repository_owner == 'searxng'
&& (github.event_name == 'workflow_dispatch' || github.event_name == 'schedule')
name: Pull Request
runs-on: ubuntu-26.04-arm
runs-on: ubuntu-24.04-arm
permissions:
# For "make V=1 weblate.translations.commit"
contents: write
@@ -83,18 +83,18 @@ jobs:
steps:
- name: Setup Python
uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: "${{ env.PYTHON_VERSION }}"
- name: Checkout
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
token: "${{ secrets.WEBLATE_GITHUB_TOKEN }}"
fetch-depth: "0"
- name: Setup cache Python
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
restore-keys: |
@@ -107,7 +107,7 @@ jobs:
- name: Setup Weblate
run: |
mkdir -p ~/.config
echo "${{ secrets.WEBLATE_CONFIG }}" >~/.config/weblate
echo "${{ secrets.WEBLATE_CONFIG }}" > ~/.config/weblate
- name: Setup Git
run: |
@@ -118,17 +118,23 @@ jobs:
run: make V=1 weblate.translations.commit
- name: Create PR
id: cpr
uses: peter-evans/create-pull-request@5f6978faf089d4d20b00c7766989d076bb2fc7f1 # v8.1.1
with:
author: "searxng-bot <searxng-bot@users.noreply.github.com>"
committer: "searxng-bot <searxng-bot@users.noreply.github.com>"
title: "[mod] i18n: update translations from Weblate"
commit-message: "[mod] i18n: update translations from Weblate"
title: "[l10n] update translations from Weblate"
commit-message: "[l10n] update translations from Weblate"
branch: "translations_update"
delete-branch: "true"
draft: "false"
signoff: "false"
body: |
Update translations from Weblate
[l10n] update translations from Weblate
labels: |
area:i18n
- name: Display information
run: |
echo "Pull Request Number - ${{ steps.cpr.outputs.pull-request-number }}"
echo "Pull Request URL - ${{ steps.cpr.outputs.pull-request-url }}"

46
.github/workflows/security.yml vendored Normal file
View File

@@ -0,0 +1,46 @@
---
name: Security
# yamllint disable-line rule:truthy
on:
workflow_dispatch:
schedule:
- cron: "42 05 * * *"
concurrency:
group: ${{ github.workflow }}
cancel-in-progress: false
permissions:
contents: read
jobs:
container:
if: github.repository_owner == 'searxng'
name: Container
runs-on: ubuntu-24.04-arm
permissions:
security-events: write
steps:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
persist-credentials: "false"
- name: Sync GHCS from Docker Scout
uses: docker/scout-action@cd72f264beff1cd72735de31148b9d3244a0234a # v1.21.0
with:
organization: "searxng"
dockerhub-user: "${{ secrets.DOCKER_USER }}"
dockerhub-password: "${{ secrets.DOCKER_TOKEN }}"
image: "registry://ghcr.io/searxng/searxng:latest"
command: "cves"
sarif-file: "./scout.sarif"
exit-code: "false"
write-comment: "false"
- name: Upload SARIFs
uses: github/codeql-action/upload-sarif@7211b7c8077ea37d8641b6271f6a365a22a5fbfa # v4.36.0
with:
sarif_file: "./scout.sarif"

11
.helix/languages.toml
View File

@@ -1,11 +0,0 @@
[[language]]
name = "python"
language-servers = ["basedpyright", "pylsp"]
auto-format = true
[language-server.pylsp.config.pylsp]
plugins.pylint.enabled = true
plugins.isort.enabled = true
plugins.black.enabled = true
plugins.black.skip_string_normalization = true
plugins.black.line_length = 120

2
Makefile
View File

@@ -63,7 +63,7 @@ format: format.python format.shell
# wrap ./manage script
MANAGE += weblate.translations.commit weblate.push.translations
MANAGE += data.all data.traits data.useragents data.locales data.currencies
MANAGE += data.all data.traits data.useragents data.gsa_useragents data.locales data.currencies
MANAGE += docs.html docs.live docs.gh-pages docs.prebuild docs.clean
MANAGE += podman.build
MANAGE += docker.build docker.buildx

44
client/simple/biome.json
View File

@@ -2,12 +2,12 @@
"$schema": "./node_modules/@biomejs/biome/configuration_schema.json",
"files": {
"ignoreUnknown": true,
"includes": ["**", "!node_modules", "!src/brand", "!src/svg"]
"includes": ["**", "!node_modules"]
},
"assist": {
"enabled": true,
"actions": {
"preset": "recommended",
"recommended": true,
"source": {
"useSortedAttributes": "on",
"useSortedProperties": "on"
@@ -27,14 +27,12 @@
"linter": {
"enabled": true,
"rules": {
"preset": "recommended",
"recommended": true,
"complexity": {
"noForEach": "error",
"noImplicitCoercions": "error",
"noRedundantDefaultExport": "error",
"noUselessCatchBinding": "error",
"noUselessUndefined": "error",
"useArrayFind": "error",
"useSimplifiedLogicExpression": "error"
},
"correctness": {
@@ -44,11 +42,25 @@
"useSingleJsDocAsterisk": "error"
},
"nursery": {
"noContinue": "warn",
"noEqualsToNull": "warn",
"noFloatingPromises": "warn",
"noForIn": "warn",
"noIncrementDecrement": "warn",
"noMisusedPromises": "warn",
"noMultiAssign": "warn",
"noMultiStr": "warn",
"noNestedPromises": "warn",
"noParametersOnlyUsedInRecursion": "warn",
"noRedundantDefaultExport": "warn",
"noReturnAssign": "warn",
"noUselessReturn": "off",
"useAwaitThenable": "off",
"useConsistentEnumValueType": "warn",
"useDestructuring": "warn",
"useExhaustiveSwitchCases": "warn",
"useExplicitType": "off",
"useFind": "warn",
"useRegexpExec": "warn"
},
"performance": {
@@ -63,15 +75,23 @@
"noCommonJs": "error",
"noEnum": "error",
"noImplicitBoolean": "error",
"noIncrementDecrement": "error",
"noInferrableTypes": "error",
"noMultiAssign": "error",
"noMultilineString": "error",
"noNamespace": "error",
"noNegationElse": "error",
"noNestedTernary": "error",
"noParameterAssign": "error",
"noParameterProperties": "error",
"noRestrictedTypes": {
"level": "error",
"options": {
"types": {
"Element": {
"message": "Element is too generic",
"use": "HTMLElement"
}
}
}
},
"noSubstr": "error",
"noUnusedTemplateLiteral": "error",
"noUselessElse": "error",
@@ -87,7 +107,6 @@
}
},
"useConsistentBuiltinInstantiation": "error",
"useConsistentEnumValueType": "error",
"useConsistentMemberAccessibility": {
"level": "error",
"options": {
@@ -107,7 +126,6 @@
}
},
"useDefaultSwitchClause": "error",
"useDestructuring": "error",
"useExplicitLengthCheck": "error",
"useForOf": "error",
"useGroupedAccessorPairs": "error",
@@ -124,17 +142,13 @@
"useUnifiedTypeSignatures": "error"
},
"suspicious": {
"noAlert": "error",
"noBitwiseOperators": "error",
"noConstantBinaryExpressions": "error",
"noDeprecatedImports": "error",
"noEmptyBlockStatements": "error",
"noEqualsToNull": "error",
"noEvolvingTypes": "error",
"noForIn": "error",
"noImportCycles": "error",
"noNestedPromises": "error",
"noParametersOnlyUsedInRecursion": "error",
"noReturnAssign": "error",
"noUnassignedVariables": "error",
"noVar": "error",
"useNumberToFixedDigitsArgument": "error",

1550
client/simple/package-lock.json generated
View File

File diff suppressed because it is too large Load Diff

24
client/simple/package.json
View File

@@ -29,21 +29,21 @@
"swiped-events": "1.2.0"
},
"devDependencies": {
"@biomejs/biome": "2.5.5",
"@types/node": "^26.1.1",
"browserslist": "^4.28.7",
"@biomejs/biome": "2.4.15",
"@types/node": "^25.8.0",
"browserslist": "^4.28.2",
"browserslist-to-esbuild": "^2.1.1",
"edge.js": "^6.5.1",
"less": "^4.8.0",
"edge.js": "^6.5.0",
"less": "^4.6.4",
"mathjs": "^15.2.0",
"sharp": "~0.35.3",
"sort-package-json": "^4.0.0",
"stylelint": "^17.14.1",
"sharp": "~0.34.5",
"sort-package-json": "^3.6.1",
"stylelint": "^17.11.1",
"stylelint-config-standard-less": "^4.1.0",
"stylelint-prettier": "^5.0.3",
"svgo": "^4.0.2",
"typescript": "~7.0.2",
"vite": "^8.1.5",
"vite-bundle-analyzer": "^1.3.9"
"svgo": "^4.0.1",
"typescript": "~6.0.3",
"vite": "^8.0.13",
"vite-bundle-analyzer": "^1.3.8"
}
}

5
client/simple/src/js/main/results.ts
View File

@@ -35,9 +35,8 @@ const imageLoader = (resultElement: HTMLElement): void => {
}, 1000) as unknown as number;
};
const imageThumbnails: NodeListOf<HTMLImageElement> = document.querySelectorAll<HTMLImageElement>(
"#urls img.image_thumbnail, img.thumbnail"
);
const imageThumbnails: NodeListOf<HTMLImageElement> =
document.querySelectorAll<HTMLImageElement>("#urls img.image_thumbnail");
for (const thumbnail of imageThumbnails) {
if (thumbnail.complete && thumbnail.naturalWidth === 0) {
thumbnail.src = `${settings.theme_static_path}/img/img_load_error.svg`;

2
client/simple/src/js/plugin/Calculator.ts
View File

@@ -77,9 +77,9 @@ export default class Calculator extends Plugin {
protected async run(): Promise<string | undefined> {
const searchInput = getElement<HTMLInputElement>("q");
const node = Calculator.math.parse(searchInput.value);
try {
const node = Calculator.math.parse(searchInput.value);
return `${node.toString()} = ${node.evaluate()}`;
} catch {
// not a compatible math expression

1
client/simple/src/js/toolkit.ts
View File

@@ -5,6 +5,7 @@ import type { KeyBindingLayout } from "./main/keyboard.ts";
// synced with searx/webapp.py get_client_settings
type Settings = {
plugins?: string[];
advanced_search?: boolean;
autocomplete?: string;
autocomplete_min?: number;
doi_resolver?: string;

7
container/builder.dockerfile
View File

@@ -1,4 +1,4 @@
FROM docker.io/searxng/base:searxng-builder AS builder
FROM ghcr.io/searxng/base:searxng-builder AS builder
COPY ./requirements.txt ./requirements-server.txt ./
@@ -21,6 +21,8 @@ RUN --mount=type=cache,id=uv,target=/root/.cache/uv set -eux -o pipefail; \
COPY --exclude=./searx/version_frozen.py ./searx/ ./searx/
ARG TIMESTAMP_SETTINGS="0"
RUN set -eux -o pipefail; \
python -m compileall -q -f -j 0 --invalidation-mode=unchecked-hash ./searx/; \
find ./searx/static/ -type f \
@@ -28,4 +30,5 @@ RUN set -eux -o pipefail; \
-exec gzip -9 -k {} + \
-exec brotli -9 -k {} + \
-exec gzip --test {}.gz + \
-exec brotli --test {}.br +
-exec brotli --test {}.br +; \
touch -c --date="@$TIMESTAMP_SETTINGS" ./searx/settings.yml

2
container/dist.dockerfile
View File

@@ -2,7 +2,7 @@ ARG CONTAINER_IMAGE_ORGANIZATION="searxng"
ARG CONTAINER_IMAGE_NAME="searxng"
FROM localhost/$CONTAINER_IMAGE_ORGANIZATION/$CONTAINER_IMAGE_NAME:builder AS builder
FROM docker.io/searxng/base:searxng AS dist
FROM ghcr.io/searxng/base:searxng AS dist
COPY --chown=977:977 --from=builder /usr/local/searxng/.venv/ ./.venv/
COPY --chown=977:977 --from=builder /usr/local/searxng/searx/ ./searx/

50
container/entrypoint.sh
View File

@@ -77,23 +77,43 @@ volume_handler() {
setup_ownership "$target" "directory"
}
setup() {
local template_settings="/usr/local/searxng/settings.template.yml"
local target_settings="$__SEARXNG_CONFIG_PATH/settings.yml"
# Handle configuration file updates
config_handler() {
local target="$1"
local template="$2"
local new_template_target="$target.new"
if [ ! -f "$target_settings" ]; then
# Create/Update the configuration file
if [ -f "$target" ]; then
setup_ownership "$target" "file"
if [ "$template" -nt "$target" ]; then
cp -pfT "$template" "$new_template_target"
cat <<EOF
...
... INFORMATION
... Update available for "$target"
... It is recommended to update the configuration file to ensure proper functionality
...
... New version placed at "$new_template_target"
... Please review and merge changes
...
EOF
fi
else
cat <<EOF
...
... INFORMATION
... "$target_settings" does not exist, creating from template...
... "$target" does not exist, creating from template...
...
EOF
cp -pfT "$template_settings" "$target_settings"
cp -pfT "$template" "$target"
sed -i "s/ultrasecretkey/$(head -c 24 /dev/urandom | base64 | tr -dc 'a-zA-Z0-9')/g" "$target_settings"
sed -i "s/ultrasecretkey/$(head -c 24 /dev/urandom | base64 | tr -dc 'a-zA-Z0-9')/g" "$target"
fi
check_file "$target_settings"
check_file "$target"
}
cat <<EOF
@@ -104,7 +124,8 @@ EOF
volume_handler "$__SEARXNG_CONFIG_PATH"
volume_handler "$__SEARXNG_DATA_PATH"
setup
# Check for files
config_handler "$__SEARXNG_SETTINGS_PATH" "/usr/local/searxng/searx/settings.yml"
# root only features
if [ "$(id -u)" -eq 0 ]; then
@@ -112,15 +133,6 @@ if [ "$(id -u)" -eq 0 ]; then
fi
# ENVs aliases
# https://github.com/searxng/searxng/issues/5934
case "${SEARXNG_PORT:-}" in
'') ;;
*[!0-9]*)
unset SEARXNG_PORT
;;
*)
export GRANIAN_PORT="$SEARXNG_PORT"
;;
esac
export GRANIAN_PORT="${SEARXNG_PORT:-$GRANIAN_PORT}"
exec /usr/local/searxng/.venv/bin/granian searx.webapp:app

8
container/settings.template.yml
View File

@@ -1,8 +0,0 @@
# Read the documentation before extending the defaults:
# https://docs.searxng.org/admin/settings/
use_default_settings: true
server:
secret_key: "ultrasecretkey"
image_proxy: true

12
docs/admin/installation-docker.rst
View File

@@ -283,12 +283,12 @@ container images are not officially supported):
$ make container
$ docker images
REPOSITORY TAG IMAGE ID SIZE
localhost/searxng/searxng 2026.6.19-93f66bfb4 ... 265 MB
localhost/searxng/searxng latest ... 265 MB
localhost/searxng/searxng builder ... 687 MB
docker.io/searxng/base searxng-builder ... 565 MB
docker.io/searxng/base searxng ... 143 MB
REPOSITORY TAG IMAGE ID CREATED SIZE
localhost/searxng/searxng 2025.8.1-3d96414 ... About a minute ago 183 MB
localhost/searxng/searxng latest ... About a minute ago 183 MB
localhost/searxng/searxng builder ... About a minute ago 524 MB
ghcr.io/searxng/base searxng-builder ... 2 days ago 378 MB
ghcr.io/searxng/base searxng ... 2 days ago 42.2 MB
Migrate from ``searxng-docker``
===============================

2
docs/admin/searx.favicons.rst
View File

@@ -32,7 +32,7 @@ By default and without any extensions, SearXNG serves these resolvers:
- ``yandex``
With the above setting favicons are displayed, the user has the option to
deactivate this feature in their settings. If the user is to have the option of
deactivate this feature in his settings. If the user is to have the option of
selecting from several *resolvers*, a further setting is required / but this
setting will be discussed :ref:`later <register resolvers>` in this article,
first we have to setup the favicons cache.

1
docs/admin/settings/index.rst
View File

@@ -19,7 +19,6 @@ Settings
settings_search
settings_server
settings_ui
settings_preferences
settings_redis
settings_valkey
settings_outgoing

8
docs/admin/settings/settings_preferences.rst
View File

@@ -1,8 +0,0 @@
.. _settings preferences:
================
``preferences:``
================
.. autoclass:: searx._settings.SettingsPref
:members:

1
docs/admin/settings/settings_search.rst
View File

@@ -43,7 +43,6 @@
- ``google``
- ``mwmbl``
- ``naver``
- ``privacywall``
- ``quark``
- ``qwant``
- ``seznam``

1
docs/admin/settings/settings_server.rst
View File

@@ -47,7 +47,6 @@
activated:
- :py:obj:`searx.botdetection.link_token` in the :ref:`limiter`
- :ref:`image_proxy`
.. _image_proxy:

8
docs/dev/engines/online/aol.rst Normal file
View File

@@ -0,0 +1,8 @@
.. _aol engine:
===
AOL
===
.. automodule:: searx.engines.aol
:members:

8
docs/dev/engines/online/exaapi.rst
View File

@@ -1,8 +0,0 @@
.. _exaapi engine:
==============
Exa API Engine
==============
.. automodule:: searx.engines.exaapi
:members:

9
docs/dev/engines/online/kagi.rst
View File

@@ -1,9 +0,0 @@
.. _kagi engines:
============
Kagi Engines
============
.. automodule:: searx.engines.kagi
:members:

8
docs/dev/engines/online/karmasearch.rst Normal file
View File

@@ -0,0 +1,8 @@
.. _karmasearch engine:
===========
Karmasearch
===========
.. automodule:: searx.engines.karmasearch
:members:

98
docs/dev/search_api.rst
View File

@@ -4,32 +4,31 @@
Search API
==========
SearXNG supports querying via a simple HTTP API. Two endpoints, ``/`` and
``/search``, are supported for both GET and POST methods. The ``GET`` method
expects parameters as URL query parameters, while the POST method expects
parameters as form data (``application/x-www-form-urlencoded``).
SearXNG supports querying via a simple HTTP API.
Two endpoints, ``/`` and ``/search``, are supported for both GET and POST methods.
The GET method expects parameters as URL query parameters, while the POST method expects parameters as form data.
If you want to consume the results as JSON, CSV, or RSS, you need to set the
``format`` parameter accordingly. Supported formats are defined in
``settings.yml``, under the :ref:`settings search` section. Requesting an
unset format will return a 403 Forbidden error. Be aware that many public
instances have these formats disabled.
``format`` parameter accordingly. Supported formats are defined in ``settings.yml``, under the ``search`` section.
Requesting an unset format will return a 403 Forbidden error. Be aware that many public instances have these formats disabled.
Endpoints:
.. code::
``GET /``
``GET /search``
GET /
GET /search
POST /
POST /search
``POST /``
``POST /search``
example cURL calls:
.. code:: bash
.. code-block:: bash
curl 'https://searx.example.org/search?q=searxng&format=json'
curl -X POST 'https://searx.example.org/search' -d 'q=searxng&format=csv'
curl -L -X POST -d 'q=searxng&format=json' 'https://searx.example.org/'
Parameters
@@ -54,27 +53,90 @@ Parameters
Comma separated list, specifies the active search categories (see
:ref:`configured engines`)
``engines`` : optional
Comma separated list, specifies the active search engines (see
:ref:`configured engines`).
``language`` : default from :ref:`settings search`
Code of the language.
``pageno`` : default ``1``
Search page number.
``time_range`` : optional : [ ``day``, ``month``, ``year`` ]
``time_range`` : optional
[ ``day``, ``month``, ``year`` ]
Time range of search for engines which support it. See if an engine supports
time range search in the preferences page of an instance.
``format`` : optional : [ ``json``, ``csv``, ``rss`` ]
``format`` : optional
[ ``json``, ``csv``, ``rss`` ]
Output format of results. Format needs to be activated in :ref:`settings
search`.
``safesearch`` : default from :ref:`settings search` : [ ``0``, ``1``, ``2`` ]
``results_on_new_tab`` : default ``0``
[ ``0``, ``1`` ]
Open search results on new tab.
``image_proxy`` : default from :ref:`settings server`
[ ``True``, ``False`` ]
Proxy image results through SearXNG.
``autocomplete`` : default from :ref:`settings search`
[ ``google``, ``dbpedia``, ``duckduckgo``, ``mwmbl``, ``startpage``,
``wikipedia``, ``swisscows``, ``qwant`` ]
Service which completes words as you type.
``safesearch`` : default from :ref:`settings search`
[ ``0``, ``1``, ``2`` ]
Filter search results of engines which support safe search. See if an engine
supports safe search in the preferences page of an instance.
``theme`` : default ``simple`` : [ ``simple`` ]
``theme`` : default ``simple``
[ ``simple`` ]
Theme of instance.
Please note, available themes depend on an instance. It is possible that an
instance administrator deleted, created or renamed themes on their instance.
See the available options in the preferences page of the instance.
``enabled_plugins`` : optional
List of enabled plugins.
:default:
``Hash_plugin``, ``Self_Information``,
``Tracker_URL_remover``, ``Ahmia_blacklist``
:values:
.. enabled by default
``Hash_plugin``, ``Self_Information``,
``Tracker_URL_remover``, ``Ahmia_blacklist``,
.. disabled by default
``Hostnames_plugin``, ``Open_Access_DOI_rewrite``,
``Vim-like_hotkeys``, ``Tor_check_plugin``
``disabled_plugins``: optional
List of disabled plugins.
:default:
``Hostnames_plugin``, ``Open_Access_DOI_rewrite``,
``Vim-like_hotkeys``, ``Tor_check_plugin``
:values:
see values from ``enabled_plugins``
``enabled_engines`` : optional : *all* :origin:`engines <searx/engines>`
List of enabled engines.
``disabled_engines`` : optional : *all* :origin:`engines <searx/engines>`
List of disabled engines.

4
docs/user/configured_engines.rst
View File

@@ -58,8 +58,8 @@ Configured Engines
{% for mod in engines %}
* - `{{mod.name}} <{{mod.about and mod.about.website}}>`_
{%- if mod.language %}
({{mod.language | upper}})
{%- if mod.about and mod.about.language %}
({{mod.about.language | upper}})
{%- endif %}
- ``!{{mod.shortcut}}``
- {%- if 'searx.engines.' + mod.__name__ in documented_modules %}

2
manage
View File

@@ -48,7 +48,7 @@ PATH="${PY_ENV}/bin:${REPO_ROOT}/node_modules/.bin:${GOROOT}/bin:${GOPATH}/bin:$
PYOBJECTS="searx"
PY_SETUP_EXTRAS='[test]'
GECKODRIVER_VERSION="v0.37.0"
GECKODRIVER_VERSION="v0.36.0"
# SPHINXOPTS=
BLACK_OPTIONS=("--target-version" "py311" "--line-length" "120" "--skip-string-normalization")
BLACK_TARGETS=("--exclude" "(searx/static|searx/languages.py)" "--include" 'searxng.msg|\.pyi?$' "searx" "searxng_extra" "tests")

14
requirements-dev.txt
View File

@@ -2,27 +2,27 @@ mock==5.2.0
nose2[coverage_plugin]==0.16.0
cov-core==1.15.0
black==25.9.0
pylint==4.0.6
pylint==4.0.5
splinter==0.21.0
selenium==4.46.0
selenium==4.44.0
Sphinx==8.2.3;python_version <= "3.11"
Sphinx==9.1.0; python_version > "3.11"
sphinx-issues==6.0.0
sphinx-jinja==2.0.2
sphinx-tabs==3.5.0
furo==2025.12.19
sphinxcontrib-programoutput==0.20
sphinxcontrib-programoutput==0.19
sphinx-autobuild==2025.8.25
sphinx-notfound-page==1.1.0
myst-parser==5.0.0
linuxdoc==20260504
aiounittest==1.5.0
yamllint==1.38.0
wlc==2.1.0
wlc==2.0.0
coloredlogs==15.0.1
docutils>=0.21.2;python_version <= "3.11"
docutils>=0.23;python_version <= "3.11"
docutils>=0.22.4; python_version > "3.11"
parameterized==0.9.0
granian[reload]==2.7.9
basedpyright==1.39.9
granian[reload]==2.7.5
basedpyright==1.39.6
types-lxml==2026.2.16

4
requirements-server.txt
View File

@@ -1,2 +1,2 @@
granian==2.7.9
granian[pname]==2.7.9
granian==2.7.5
granian[pname]==2.7.5

6
requirements.txt
View File

@@ -1,4 +1,4 @@
certifi==2026.7.22
certifi==2026.5.20
babel==2.18.0
flask-babel==4.0.0
flask==3.1.3
@@ -13,7 +13,7 @@ sniffio==1.3.1
valkey==6.1.1
markdown-it-py==4.2.0
msgspec==0.21.1
typer==0.27.0
typer==0.26.3
isodate==0.7.2
whitenoise==6.12.0
typing-extensions==4.16.0
typing-extensions==4.15.0

9
searx/__init__.py
View File

@@ -10,7 +10,6 @@ from os.path import dirname, abspath
import logging
import msgspec
from ._settings import SettingsPref
# Debug
LOG_FORMAT_DEBUG: str = '%(levelname)-7s %(name)-30.30s: %(message)s'
@@ -48,12 +47,6 @@ def init_settings():
settings.clear()
settings.update(cfg)
if get_setting("server.public_instance"):
# enable image proxy for public instances #6125
settings["server"]["image_proxy"] = True
pref: SettingsPref = get_setting("preferences")
pref.lock.add("image_proxy")
sxng_debug = get_setting("general.debug")
if sxng_debug:
_logging_config_debug()
@@ -73,7 +66,7 @@ def init_settings():
if settings['server']['public_instance']:
logger.warning(
"Be aware you have activated features intended only for public instances. "
"This force the usage of the limiter, link_token and image proxy / "
"This force the usage of the limiter and link_token / "
"see https://docs.searxng.org/admin/searx.limiter.html"
)

42
searx/_settings.py
View File

@@ -1,42 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Implementation of the :py:obj:`preference <searx.preference>` settings."""
# pylint: disable = too-few-public-methods
import typing as t
import msgspec
class SettingsPref(msgspec.Struct, kw_only=True, forbid_unknown_fields=True):
"""Options for configuring the preferences
.. code:: yaml
preferences:
lock:
- favicon_resolver
- image_proxy
- method
# ...
"""
lock: set[
t.Literal[
"categories",
"language",
"locale",
"autocomplete",
"favicon_resolver",
"image_proxy",
"method",
"safesearch",
"theme",
"results_on_new_tab",
"doi_resolver",
"simple_style",
"center_alignment",
"query_in_title",
"search_on_category_select",
]
] = set()
"""Lock arbitrary settings on the preferences page."""

24
searx/autocomplete.py
View File

@@ -21,8 +21,6 @@ from searx.engines import (
from searx.network import get as http_get, post as http_post
from searx.exceptions import SearxEngineResponseException
from searx.utils import extr, gen_useragent
from searx.data import ENGINE_TRAITS
from searx.enginelib.traits import EngineTraits
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
@@ -135,9 +133,7 @@ def google_complete(query: str, sxng_locale: str) -> list[str]:
"""
data = ENGINE_TRAITS.get("google") or {}
traits = EngineTraits(**data)
google_info: dict[str, t.Any] = google.get_google_info({'searxng_locale': sxng_locale}, traits)
google_info: dict[str, t.Any] = google.get_google_info({'searxng_locale': sxng_locale}, engines['google'].traits)
url = 'https://{subdomain}/complete/search?{args}'
args = urlencode(
{
@@ -183,23 +179,6 @@ def naver(query: str, _sxng_locale: str) -> list[str]:
return results
def privacywall(query: str, sxng_locale: str) -> list[str]:
# Privacywall search autocompleter
country = None
if "-" in sxng_locale:
country = sxng_locale.split("-")[1]
args = {'q': query, 'cc': country}
url = f"https://www.privacywall.org/search/secure/suggestions.php?{urlencode(args)}"
response = get(url)
if not response.ok:
return []
data: list[list[str]] = response.json()
return data[1]
def qihu360search(query: str, _sxng_locale: str) -> list[str]:
# 360Search search autocompleter
url = f"https://sug.so.360.cn/suggest?{urlencode({'format': 'json', 'word': query})}"
@@ -382,7 +361,6 @@ backends: dict[str, t.Callable[[str, str], list[str]]] = {
'google': google_complete,
'mwmbl': mwmbl,
'naver': naver,
'privacywall': privacywall,
'quark': quark,
'qwant': qwant,
'seznam': seznam,

29
searx/cache.py
View File

@@ -48,7 +48,7 @@ class ExpireCacheCfg(msgspec.Struct): # pylint: disable=too-few-public-methods
MAXHOLD_TIME: int = 60 * 60 * 24 * 7 # 7 days
"""Hold time (default in sec.), after which a value is removed from the cache."""
MAINTENANCE_PERIOD: int = 60 * 60 # 1h
MAINTENANCE_PERIOD: int = 60 * 60 # 2h
"""Maintenance period in seconds / when :py:obj:`MAINTENANCE_MODE` is set to
``auto``."""
@@ -444,10 +444,12 @@ class ExpireCacheSQLite(sqlitedb.SQLiteAppl, ExpireCache):
def get(self, key: str, default: typing.Any = None, ctx: str | None = None) -> typing.Any:
"""Get value of ``key`` from table given by argument ``ctx``. If
``ctx`` argument is ``None`` (the default), a table name is generated
from the :py:obj:`ExpireCacheCfg.name`. If ``key`` not exists in
the table or the table not exists, the ``default`` value is returned.
from the :py:obj:`ExpireCacheCfg.name`. If ``key`` not exists (in
table), the ``default`` value is returned.
"""
table = ctx
self.maintenance()
if not table:
table = self.normalize_name(self.cfg.name)
@@ -455,39 +457,24 @@ class ExpireCacheSQLite(sqlitedb.SQLiteAppl, ExpireCache):
if table not in self.table_names:
return default
# Before values are taken from the table, a maintenance interval may
# need to be carried out.
self.maintenance()
sql = f"SELECT value, expire FROM {table} WHERE key = ?"
sql = f"SELECT value FROM {table} WHERE key = ?"
row = self.DB.execute(sql, (key,)).fetchone()
if row is None:
return default
# Check if value is expired. It's possible that it's expired but has not
# yet been automatically deleted by the periodic maintenance
(value, expire) = row
now = time.time()
if expire < now:
# The record is deleted during the maintenance interval. Deleting
# the record at this point offers no advantage, as a SELECT
# statement must be executed for every cache.get request anyways.
return default
return self.deserialize(value)
return self.deserialize(row[0])
def pairs(self, ctx: str) -> Iterator[tuple[str, typing.Any]]:
"""Iterate over key/value pairs from table given by argument ``ctx``.
If ``ctx`` argument is ``None`` (the default), a table name is
generated from the :py:obj:`ExpireCacheCfg.name`."""
table = ctx
self.maintenance()
if not table:
table = self.normalize_name(self.cfg.name)
if table in self.table_names:
# Before values are taken from the table, a maintenance interval may
# need to be carried out.
self.maintenance()
for row in self.DB.execute(f"SELECT key, value FROM {table}"):
yield row[0], self.deserialize(row[1])

24
searx/data/__init__.py
View File

@@ -6,7 +6,7 @@ make data.all
"""
# pylint: disable=invalid-name
__all__ = ["ahmia_blacklist_loader", "data_dir", "get_cache"]
__all__ = ["ahmia_blacklist_loader", "gsa_useragents_loader", "data_dir", "get_cache"]
import json
import typing as t
@@ -63,6 +63,7 @@ lazy_globals = {
"ENGINE_TRAITS": None,
"LOCALES": None,
"TRACKER_PATTERNS": TrackerPatternsDB(),
"GSA_USER_AGENTS": None,
}
data_json_files = {
@@ -105,3 +106,24 @@ def ahmia_blacklist_loader() -> list[str]:
"""
with open(data_dir / 'ahmia_blacklist.txt', encoding='utf-8') as f:
return f.read().split()
def gsa_useragents_loader() -> list[str]:
"""Load data from `gsa_useragents.txt` and return a list of user agents
suitable for Google. The user agents are fetched by::
searxng_extra/update/update_gsa_useragents.py
This function is used by :py:mod:`searx.engines.google`.
"""
data = lazy_globals["GSA_USER_AGENTS"]
if data is not None:
return data
log.debug("init searx.data.%s", "GSA_USER_AGENTS")
with open(data_dir / 'gsa_useragents.txt', encoding='utf-8') as f:
lazy_globals["GSA_USER_AGENTS"] = f.read().splitlines()
return lazy_globals["GSA_USER_AGENTS"]

1224
searx/data/ahmia_blacklist.txt
View File

File diff suppressed because it is too large Load Diff

40
searx/data/currencies.json
View File

@@ -773,7 +773,7 @@
"nl": "Boliviaanse boliviano",
"oc": "Boliviano",
"pa": "ਬੋਲੀਵੀਆਨੋ",
"pl": "boliwiano",
"pl": "boliviano",
"pt": "Boliviano",
"ro": "boliviano",
"ru": "боливиано",
@@ -1107,7 +1107,7 @@
"fi": "Kongon frangi",
"fr": "franc congolais",
"ga": "franc an Chongó",
"gl": "franco congolés",
"gl": "Franco congolés",
"he": "פרנק קונגולזי",
"hr": "Kongoanski franak",
"hu": "kongói frank",
@@ -1636,7 +1636,7 @@
"fi": "Algerian dinaari",
"fr": "dinar algérien",
"ga": "dinar na hAilgéire",
"gl": "dinar alxeriano",
"gl": "Dinar alxeriano",
"he": "דינר אלג'ירי",
"hr": "Alžirski dinar",
"hu": "algériai dinár",
@@ -1651,7 +1651,7 @@
"pap": "dinar argelino",
"pl": "dinar algierski",
"pt": "dinar argelino",
"ro": "dinar algerian",
"ro": "Dinar algerian",
"ru": "алжирский динар",
"sk": "Alžírský dinár",
"sl": "alžirski dinar",
@@ -1786,7 +1786,7 @@
"cy": "Ewro",
"da": "Euro",
"de": "Euro",
"en": "euro",
"en": "Euro",
"eo": "eŭro",
"es": "Euro",
"et": "Euro",
@@ -1803,7 +1803,7 @@
"it": "Euro",
"ja": "ユーロ",
"ko": "유로",
"lt": "euras",
"lt": "Euras",
"lv": "eiro",
"ml": "യൂറോ",
"ms": "Euro",
@@ -2759,7 +2759,7 @@
"pa": "ਜਪਾਨੀ ਯੈੱਨ",
"pl": "jen",
"pt": "iene",
"ro": "yen",
"ro": "yeni",
"ru": "японская иена",
"sk": "jen",
"sl": "japonski jen",
@@ -3337,7 +3337,6 @@
"fi": "Libyan dinaari",
"fr": "dinar libyen",
"ga": "dinar na Libia",
"gl": "dinar libio",
"he": "דינר לובי ",
"hr": "Libijski dinar",
"hu": "líbiai dinár",
@@ -3540,7 +3539,6 @@
"ja": "チャット",
"ko": "미얀마 짯",
"lt": "Kijatas",
"lv": "Kjats",
"ml": "ബർമ്മീസ് ക്യാറ്റ്",
"nl": "Myanmarese kyat",
"oc": "Kyat",
@@ -4312,7 +4310,7 @@
"ar": "بيسو فلبيني",
"bg": "Филипинско песо",
"ca": "peso filipí",
"cs": "filipínské peso",
"cs": "Filipínské peso",
"de": "philippinischer Peso",
"en": "Philippine peso",
"eo": "filipina peso",
@@ -4616,7 +4614,7 @@
"fi": "Serbian dinaari",
"fr": "dinar serbe",
"ga": "Dinar na Seirbia",
"gl": "dinar serbio",
"gl": "Dinar serbio",
"he": "דינר סרבי",
"hr": "srpski dinar",
"hu": "szerb dinár",
@@ -5333,7 +5331,7 @@
"pa": "ਤਾਜਿਕਿਸਤਾਨੀ ਸੋਮੋਨੀ",
"pl": "Somoni",
"pt": "Somoni",
"ro": "somoni tadjic",
"ro": "Somoni tadjic",
"ru": "таджикский сомони",
"sk": "tadžický som",
"sl": "tadžikistanski somoni",
@@ -5396,7 +5394,6 @@
"fi": "Tunisian dinaari",
"fr": "dinar tunisien",
"ga": "dinar na Túinéise",
"gl": "dinar tunisiano",
"he": "דינר תוניסאי",
"hr": "tuniski dinar",
"hu": "tunéziai dinár",
@@ -5769,7 +5766,7 @@
"fi": "Uruguayn peso",
"fr": "peso uruguayen",
"ga": "peso Uragua",
"gl": "peso uruguaio",
"gl": "Peso uruguaio",
"he": "פסו של אורוגוואי",
"hr": "Urugvajski pezo",
"hu": "uruguayi peso",
@@ -6150,7 +6147,7 @@
"oc": "Drechs de tiratge Especials",
"pl": "specjalne prawa ciągnienia",
"pt": "direitos especiais de saque",
"ro": "drepturi speciale de tragere",
"ro": "Drepturi speciale de tragere",
"ru": "специальные права заимствования",
"sk": "Zvláštne práva čerpania",
"sl": "posebne pravice črpanja",
@@ -6228,7 +6225,6 @@
"ja": "CFPフラン",
"ko": "CFP 프랑",
"lt": "CFP frankas",
"lv": "Klusā okeāna franks",
"ms": "Franc CFP",
"nl": "CFP-frank",
"oc": "Franc CFP",
@@ -7059,7 +7055,6 @@
"bolivjano": "BOB",
"bolivya bolivianosu": "BOB",
"bolivya bolivyanosu": "BOB",
"boliwiano": "BOB",
"bolívar digital": "VED",
"bolívar soberano": "VES",
"bolívar sobirà": "VES",
@@ -8430,7 +8425,6 @@
"drame arménio": "AMD",
"dramm": "AMD",
"drechs de tiratge especials": "XDR",
"drept special de tragere": "XDR",
"drepturi speciale de tragere": "XDR",
"drets especials de gir": "XDR",
"droits de tirage speciaux": "XDR",
@@ -9524,8 +9518,6 @@
"kíp lào": "LAK",
"kīp": "LAK",
"kjat": "MMK",
"kjats": "MMK",
"klusā okeāna franks": "XPF",
"km": "BAM",
"kmf": "KMF",
"koeweitse dinar": "KWD",
@@ -10031,7 +10023,6 @@
"lire sterline": "GBP",
"lire turque": "TRY",
"lisente": "LSL",
"list of syrian coins": "SYP",
"liura de gibartar": "GIP",
"liura egipciana": "EGP",
"liura esterlina": "GBP",
@@ -11322,6 +11313,7 @@
"riel camboxano": "KHR",
"riel camboyano": "KHR",
"riel campuchia": "KHR",
"riel kambodżański": "KHR",
"riel kamboja": "KHR",
"riel na cambóide": "KHR",
"rietumāfrikas franks": "XOF",
@@ -12265,7 +12257,6 @@
"thaise baht": "THB",
"thajský baht": "THB",
"thb": "THB",
"the australian dollar": "AUD",
"thebe": "BWP",
"third belarusian ruble": "BYN",
"tical": "THB",
@@ -12749,7 +12740,6 @@
"yen": "JPY",
"yen giapponese": "JPY",
"yen japones": "JPY",
"yen japonez": "JPY",
"yen japonés": "JPY",
"yeni": "JPY",
"yeni i̇srail şekeli": "ILS",
@@ -12816,6 +12806,7 @@
"zimbabwe zig": "ZWG",
"zimbabwean dollar": "ZWL",
"zimbabwean gold": "ZWG",
"zimbabwean zig": "ZWG",
"zimbabwen kulta": "ZWG",
"zimbabwiansky zlatý": "ZWG",
"zimbabwský dolar": "ZWL",
@@ -12974,7 +12965,6 @@
"FKP",
"EGP"
],
"£S": "SYP",
"£e": "EGP",
"£s": "SYP",
"¥": [
@@ -14969,8 +14959,6 @@
"فورنت مجري": "HUF",
"فورينت مجري": "HUF",
"فِرَنْكٌ رُوَنْدِيٌّ": "RWF",
"قائمة النقود المعدنية السورية": "SYP",
"قائمة عملات سوريا المعدنية": "SYP",
"ك": "KWD",
"كتزال غواتيمالي": "GTQ",
"كرونة آيسلندية": "ISK",

5421
searx/data/engine_descriptions.json
View File

File diff suppressed because it is too large Load Diff

1926
searx/data/engine_traits.json
View File

File diff suppressed because it is too large Load Diff

4502
searx/data/gsa_useragents.txt Normal file
View File

File diff suppressed because it is too large Load Diff

4
searx/data/useragents.json
View File

@@ -5,7 +5,7 @@
],
"ua": "Mozilla/5.0 ({os}; rv:{version}) Gecko/20100101 Firefox/{version}",
"versions": [
"152.0",
"151.0"
"151.0",
"150.0"
]
}

2
searx/data/wikidata_units.json
View File

@@ -3272,7 +3272,7 @@
"Q128822": {
"si_name": "Q182429",
"symbol": "kn",
"to_si_factor": 0.514
"to_si_factor": 0.5144444444444445
},
"Q12912288": {
"si_name": null,

232
searx/enginelib/__init__.py
View File

@@ -3,7 +3,6 @@
- :py:obj:`searx.enginelib.EngineCache`
- :py:obj:`searx.enginelib.Engine`
- :py:obj:`searx.enginelib.EngineAbout`
- :py:obj:`searx.enginelib.traits`
There is a command line for developer purposes and for deeper analysis. Here is
@@ -24,7 +23,7 @@ an example in which the command line is called in the development environment::
"""
__all__ = ["EngineCache", "Engine", "EngineAbout", "ENGINES_CACHE"]
__all__ = ["EngineCache", "Engine", "ENGINES_CACHE"]
import typing as t
import abc
@@ -32,7 +31,6 @@ from collections.abc import Callable
import logging
import string
import typer
import msgspec
from ..cache import ExpireCacheSQLite, ExpireCacheCfg
@@ -41,13 +39,13 @@ if t.TYPE_CHECKING:
from searx.enginelib.traits import EngineTraits
from searx.extended_types import SXNG_Response
from searx.result_types import EngineResults
from searx.search.processors import OfflineParamTypes, OnlineParamTypes, ProcessorType
from searx.search.processors import OfflineParamTypes, OnlineParamTypes
ENGINES_CACHE: ExpireCacheSQLite = ExpireCacheSQLite.build_cache(
ExpireCacheCfg(
name="ENGINES_CACHE",
MAXHOLD_TIME=60 * 60 * 24 * 7, # 7 days
MAINTENANCE_PERIOD=60 * 60, # 1h
MAINTENANCE_PERIOD=60 * 60, # 2h
MAX_VALUE_LEN=1024 * 1024 * 1024, # 1MB
)
)
@@ -180,58 +178,11 @@ class EngineCache:
return ENGINES_CACHE.secret_hash(name=name)
class EngineAbout(msgspec.Struct, kw_only=True):
"""Additional fields describing the engine.
.. code:: yaml
about:
website: https://example.com
wikidata_id: Q306656
official_api_documentation: https://example.com/api-doc
use_official_api: true
require_api_key: true
results: HTML
"""
# pylint: disable=too-few-public-methods
website: str = ""
"""Official web-site of the origin."""
wikidata_id: str = ""
"""`Wikidata ID <https://www.wikidata.org/wiki/Wikidata:Identifiers>`_"""
official_api_documentation: str = ""
"""URL of the official API (regardless of whether it is used)"""
use_official_api: bool = False
"""SearXNG engine makes use of the official API or not"""
require_api_key: bool = False
"""API requires a key or not."""
results: str = ""
"""Data format of the source (online-engines: of the response)."""
description: str = ""
"""Brief description of the engine and where it gets its data from.
This value should only be set as long as no description of the data source
is available via a :py:obj:`EngineAbout.wikidata_id`."""
language: str = ""
"""Deprecated! Migrate your setting from `engine.about.language` to
`engine.language`"""
class Engine(abc.ABC): # pylint: disable=too-few-public-methods
"""Class of engine instances build from YAML settings.
Further documentation see :ref:`general engine configuration`.
The defaults are taken from :py:obj:`searx.engines.ENGINE_DEFAULT_ARGS`.
.. hint::
This class is currently never initialized and only used for type hinting.
@@ -239,71 +190,39 @@ class Engine(abc.ABC): # pylint: disable=too-few-public-methods
logger: logging.Logger
# Common options of the engine module
# Common options in the engine module
engine_type: "ProcessorType" = "online"
engine_type: str
"""Type of the engine (:ref:`searx.search.processors`)"""
paging: bool = False
paging: bool
"""Engine supports multiple pages."""
max_page: int = 0
"""If the engine supports paging, then this is the value for the last page
that is still supported. ``0`` means unlimited numbers of pages."""
time_range_support: bool = False
time_range_support: bool
"""Engine supports search time range."""
safesearch: bool = False
safesearch: bool
"""Engine supports SafeSearch"""
language_support: bool = False
language_support: bool
"""Engine supports languages (locales) search."""
fetch_traits: "Callable[[EngineTraits, bool], None]"
"""Function to to fetch engine's traits from origin."""
traits: "traits.EngineTraits"
"""Traits of the engine."""
# settings.yml
name: str
"""Name that will be used across SearXNG to define this engine. In settings, on
the result page .."""
engine: str
"""Name of the python file used to handle requests and responses to and from
this search engine (file name from :origin:`searx/engines` without
``.py``)."""
categories: list[str] = ["general"]
"""Specifies to which :ref:`engine categories` the engine should be added."""
language: str = ""
"""If the engine supports only one language, this language is specified here
(``en``, ``de``, ``"no"`` or ..); otherwise, the value remains empty. For
the YAML configuration: think of the `YAML-Norway problem
<https://ruuda.nl/2023/the-yaml-document-from-hell#the-norway-problem>`_
language: str
"""For an engine, when there is ``language: ...`` in the YAML settings the engine
does support only this one language:
.. code:: yaml
- name: google norway
- name: google french
engine: google
language: "no"
Depending on ``language_support``, this value has similar but also slightly
different meanings.
- When ``language_support`` is **true**, the map of
:py:obj:`traits.EngineTraits.languages` is reduced to the selected
language
- When ``language_support`` is **false**, then the implementation of the
engine only supports this one ``language``
language: fr
"""
region: str = ""
region: str
"""For an engine, when there is ``region: ...`` in the YAML settings the engine
does support only this one region::
@@ -314,6 +233,26 @@ class Engine(abc.ABC): # pylint: disable=too-few-public-methods
region: fr-BE
"""
fetch_traits: "Callable[[EngineTraits, bool], None]"
"""Function to to fetch engine's traits from origin."""
traits: "traits.EngineTraits"
"""Traits of the engine."""
# settings.yml
categories: list[str]
"""Specifies to which :ref:`engine categories` the engine should be added."""
name: str
"""Name that will be used across SearXNG to define this engine. In settings, on
the result page .."""
engine: str
"""Name of the python file used to handle requests and responses to and from
this search engine (file name from :origin:`searx/engines` without
``.py``)."""
enable_http: bool
"""Enable HTTP (by default only HTTPS is enabled)."""
@@ -326,31 +265,6 @@ class Engine(abc.ABC): # pylint: disable=too-few-public-methods
display_error_messages: bool
"""Display error messages on the web UI."""
disabled: bool = False
"""To disable by default the engine, but not deleting it. It will allow the
user to manually activate it in the settings."""
inactive: bool = False
"""Remove the engine from the settings (*disabled & removed*)."""
about: EngineAbout = EngineAbout()
"""Additional fields describing the engine."""
using_tor_proxy: bool = False
"""Using tor proxy (``true``) or not (``false``) for this engine."""
send_accept_language_header: bool = True
"""When this option is activated (default), the language (locale) that is
selected by the user is used to build and send a ``Accept-Language`` header
in the request to the origin search engine."""
tokens: list[str] = []
"""A list of secret tokens to make this engine *private*, more details see
:ref:`private engines`."""
weight: float = 1.0
"""Weighting of the results of this engine (:ref:`weight <settings engines>`)."""
proxies: dict[str, dict[str, str]]
"""Set proxies for a specific engine (YAML):
@@ -361,43 +275,73 @@ class Engine(abc.ABC): # pylint: disable=too-few-public-methods
https: socks5://proxy:port
"""
def setup(self, engine_settings: dict[str, t.Any]) -> bool | None: # pylint: disable=unused-argument
disabled: bool
"""To disable by default the engine, but not deleting it. It will allow the
user to manually activate it in the settings."""
inactive: bool
"""Remove the engine from the settings (*disabled & removed*)."""
about: dict[str, dict[str, str]]
"""Additional fields describing the engine.
.. code:: yaml
about:
website: https://example.com
wikidata_id: Q306656
official_api_documentation: https://example.com/api-doc
use_official_api: true
require_api_key: true
results: HTML
"""
using_tor_proxy: bool
"""Using tor proxy (``true``) or not (``false``) for this engine."""
send_accept_language_header: bool
"""When this option is activated (default), the language (locale) that is
selected by the user is used to build and send a ``Accept-Language`` header
in the request to the origin search engine."""
tokens: list[str]
"""A list of secret tokens to make this engine *private*, more details see
:ref:`private engines`."""
weight: int
"""Weighting of the results of this engine (:ref:`weight <settings engines>`)."""
def setup(self, engine_settings: dict[str, t.Any]) -> bool: # pylint: disable=unused-argument
"""Dynamic setup of the engine settings.
With this method, the engine's setup is carried out. For example, to
check or dynamically adapt the values handed over in the parameter
``engine_settings``.
``engine_settings``. The return value (True/False) indicates whether
the setup was successful and the engine can be built or rejected.
Whether the initialization was successful can be indicated by the return
value ``True`` or even ``False``.
- If no return value (``None`` ) is given from this method , this is
equivalent to ``True``.
- If an exception is thrown as part of the initialization, this is
equivalent to ``False``.
The method is optional and is called **synchronously** as part of the
The method is optional and is called synchronously as part of the
initialization of the service and is therefore only suitable for simple
(local) exams/changes at the engine setting.
The :py:obj:`Engine.init` method must be used for longer tasks in which
values of a remote must be determined, for example.
(local) exams/changes at the engine setting. The :py:obj:`Engine.init`
method must be used for longer tasks in which values of a remote must be
determined, for example.
"""
return True
def init(self, engine_settings: dict[str, t.Any]) -> bool | None: # pylint: disable=unused-argument
"""Initialization of the engine.
The method is optional and called **asynchronous** (in a thread). The
method is comparable to :py:obj:`Engine.setup`, it is suitable, for
caching data that first needs to be requested from a remote.
The method is optional and asynchronous (in a thread). It is suitable,
for example, for setting up a cache (for the engine) or for querying
values (required by the engine) from a remote.
The method is optional and runs **asynchronously** (in a thread), it is
comparable to :py:obj:`Engine.setup`. For instance, it is suitable for
caching data that first needs to be requested from a remote source.
Whether the initialization was successful can be indicated by the return
value ``True`` or even ``False``.
The evaluation of the return value is analogous to :py:obj:`Engine.setup`.
- If no return value is given from this init method (``None``), this is
equivalent to ``True``.
- If an exception is thrown as part of the initialization, this is
equivalent to ``False``.
"""
return True

27
searx/enginelib/traits.py
View File

@@ -142,11 +142,11 @@ class EngineTraits:
"""
if self.data_type == "traits_v1":
self._set_traits_v1(engine) # pyright: ignore[reportArgumentType]
self._set_traits_v1(engine)
else:
raise TypeError("engine traits of type %s is unknown" % self.data_type)
def _set_traits_v1(self, engine: "Engine") -> None:
def _set_traits_v1(self, engine: "Engine | types.ModuleType") -> None:
# For an engine, when there is `language: ...` in the YAML settings the engine
# does support only this one language (region)::
#
@@ -159,25 +159,22 @@ class EngineTraits:
_msg = "settings.yml - engine: '%s' / %s: '%s' not supported"
if engine.language:
if engine.language_support:
if not len(traits.languages) > 1:
raise ValueError(
f"engine {engine.name}: activated language_support with just one or less languages"
)
if engine.language not in traits.languages:
raise ValueError(_msg % (engine.name, "language", engine.language))
traits.languages = {engine.language: traits.languages[engine.language]}
languages = traits.languages
if hasattr(engine, "language"):
if engine.language not in languages:
raise ValueError(_msg % (engine.name, "language", engine.language))
traits.languages = {engine.language: languages[engine.language]}
if engine.region:
if engine.region not in traits.regions:
regions = traits.regions
if hasattr(engine, "region"):
if engine.region not in regions:
raise ValueError(_msg % (engine.name, "region", engine.region))
traits.regions = {engine.region: traits.regions[engine.region]}
traits.regions = {engine.region: regions[engine.region]}
engine.language_support = bool(traits.languages or traits.regions)
# set the copied & modified traits in engine's namespace
engine.traits = traits
engine.traits = traits # pyright: ignore[reportAttributeAccessIssue]
class EngineTraitsMap(dict[str, EngineTraits]):

2
searx/engines/360search.py
View File

@@ -22,8 +22,8 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": "HTML",
"language": "zh",
}
language = "zh"
# Engine Configuration
categories = ["general"]

12
searx/engines/__builtins__.pyi
View File

@@ -5,19 +5,19 @@ intended monkey patching of the engine modules.
.. attention::
Monkey-patching modules is a practice from the past that shouldn't be
expanded upon. In the long run, engines should be instances of
:py:obj:`searx.enginelib.Engine`. However, as long as long as all engine
modules aren't converted to this class, these builtin types will still be
needed.
expanded upon. In the long run, there should be an engine class that can be
inherited. However, as long as this class doesn't exist, and as long as all
engine modules aren't converted to an engine class, these builtin types will
still be needed.
"""
import logging
from searx.enginelib import traits as _traits
logger: logging.Logger
supported_languages: str
language_aliases: str
language_support: bool
language: str
region: str
traits: _traits.EngineTraits
# from searx.engines.ENGINE_DEFAULT_ARGS

75
searx/engines/__init__.py
View File

@@ -12,50 +12,41 @@ import typing as t
import sys
import copy
import os
from os.path import realpath, dirname
import warnings
import types
import inspect
import msgspec
from searx import logger, settings
from searx.utils import load_module
from searx.data import ENGINE_TRAITS
from searx.enginelib import Engine, EngineAbout
if t.TYPE_CHECKING:
from searx.enginelib import Engine
logger = logger.getChild('engines')
ENGINE_DIR = dirname(realpath(__file__))
# Defaults for the namespace of an engine module, see load_engine()
ENGINE_DEFAULT_ARGS: dict[str, t.Any] = {
ENGINE_DEFAULT_ARGS: dict[str, int | str | list[t.Any] | dict[str, t.Any] | bool] = {
# Common options in the engine module
"engine_type": "online",
"paging": False,
"max_page": 0,
"time_range_support": False,
"safesearch": False,
"language_support": False,
# settings.yml
"categories": ["general"],
"language": "",
"region": "",
"enable_http": False,
"shortcut": "-",
"timeout": settings["outgoing"]["request_timeout"],
"display_error_messages": True,
"disabled": False,
"inactive": False,
"about": EngineAbout(),
"about": {},
"using_tor_proxy": False,
"send_accept_language_header": True,
"tokens": [],
"weight": 1.0,
"max_page": 0,
}
"""Default values that are set in an engine of type *module*, please compare
with the class :py:obj:`searx.enginelib.Engine`."""
# set automatically when an engine does not have any tab category
DEFAULT_CATEGORY = 'other'
@@ -185,41 +176,14 @@ def set_loggers(engine: "Engine|types.ModuleType", engine_name: str):
def update_engine_attributes(engine: "Engine | types.ModuleType", engine_data: dict[str, t.Any]):
# pylint: disable=too-many-branches
# set engine attributes from engine_data
kvargs: dict[str, t.Any]
if isinstance(engine.about, EngineAbout):
kvargs = {**msgspec.to_builtins(engine.about), **engine_data.get("about", {})}
else:
kvargs = {**engine.about, **engine_data.get("about", {})}
try:
engine.about = EngineAbout(**kvargs)
except TypeError as exc:
raise TypeError(
f"engine '{engine_data['name']}' ({engine_data['engine']}) - in the about section --> {exc}"
) from exc
# warn about deprecated engine settings
if engine.about.language:
if hasattr(engine, "language") and not engine.language:
engine.language = engine.about.language
warnings.warn(
f"engine '{engine_data['name']}' ({engine_data['engine']})"
f" - migrate engine.about.language to engine.language!",
DeprecationWarning,
2,
)
for param_name, param_value in engine_data.items():
if param_name == "about":
continue
if param_name == 'categories':
if isinstance(param_value, str):
param_value = list(map(str.strip, param_value.split(',')))
engine.categories = param_value # type: ignore
elif hasattr(engine, 'about') and param_name == 'about':
engine.about = {**engine.about, **engine_data['about']} # type: ignore
else:
setattr(engine, param_name, param_value)
@@ -228,9 +192,6 @@ def update_engine_attributes(engine: "Engine | types.ModuleType", engine_data: d
if not hasattr(engine, arg_name):
setattr(engine, arg_name, copy.deepcopy(arg_value))
if ENGINE_TRAITS.get(engine.name, {}).get("languages") and not engine.language_support:
raise ValueError(f"engine '{engine.name}' ({engine_data['engine']}) language_support should be set to True")
def update_attributes_for_tor(engine: "Engine | types.ModuleType"):
if using_tor_proxy(engine) and hasattr(engine, 'onion_url'):
@@ -269,27 +230,21 @@ def is_engine_active(engine: "Engine | types.ModuleType"):
def call_engine_setup(engine: "Engine | types.ModuleType", engine_data: dict[str, t.Any]) -> bool:
setup_ok: bool | None = False
setup_ok = False
setup_func = getattr(engine, "setup", None)
if setup_func is None:
setup_ok = True
elif not callable(setup_func):
logger.error(f"engine's setup method isn't a callable (is of type: {type(setup_func)})")
logger.error("engine's setup method isn't a callable (is of type: %s)", type(setup_func))
else:
try:
setup_ok = engine.setup(engine_data)
except Exception as e: # pylint: disable=broad-except
logger.exception(f"(PID {os.getpid()}) {engine.name}: engine SETUP failed, exception: {e}")
setup_ok = False
# The evaluation of the return value is analogous to Engine.init
if setup_ok is None:
setup_ok = True
logger.exception('exception : {0}'.format(e))
if not setup_ok:
logger.error(f"(PID {os.getpid()}) {engine.name}: engine setup was not successful")
logger.error("%s: Engine setup was not successful, engine is set to inactive.", engine.name)
return setup_ok
@@ -317,16 +272,12 @@ def load_engines(engine_list: list[dict[str, t.Any]]):
for engine_data in engine_list:
if engine_data.get("inactive") is True:
continue
engine = load_engine(engine_data)
if engine:
register_engine(engine)
else:
# if an engine can't be loaded (if for example the engine is missing
# tor or some other requirements) its set to inactive!
logger.error(
f"(PID {os.getpid()}) {engine_data.get('name', '???')}: can't register engine (loading engine failed)"
)
logger.error("loading engine %s failed: set engine to inactive!", engine_data.get("name", "???"))
engine_data["inactive"] = True
return engines

4
searx/engines/acfun.py
View File

@@ -16,12 +16,12 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": "HTML",
"language": "zh",
}
# Engine Configuration
categories = ["videos"]
paging = True
language = "zh"
# Base URL
base_url = "https://www.acfun.cn"
@@ -83,7 +83,7 @@ def extract_video_data(video_block):
published_date = None
if create_time:
try:
published_date = datetime.fromisoformat(create_time.strip())
published_date = datetime.strptime(create_time.strip(), "%Y-%m-%d")
except (ValueError, TypeError):
pass

1
searx/engines/annas_archive.py
View File

@@ -64,7 +64,6 @@ about: dict[str, t.Any] = {
# engine dependent config
categories = ["files", "books"]
paging: bool = True
language_support = True
# search-url
base_url: list[str] | str = []

2
searx/engines/ansa.py
View File

@@ -42,8 +42,8 @@ about = {
'use_official_api': False,
'require_api_key': False,
'results': 'HTML',
'language': 'it',
}
language = "it"
def request(query, params):

210
searx/engines/aol.py Normal file
View File

@@ -0,0 +1,210 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""AOL supports WEB, image, and video search. Internally, it uses the Bing
index.
AOL doesn't seem to support setting the language via request parameters, instead
the results are based on the URL. For example, there is
- `search.aol.com <https://search.aol.com>`_ for English results
- `suche.aol.de <https://suche.aol.de>`_ for German results
However, AOL offers its services only in a few regions:
- en-US: search.aol.com
- de-DE: suche.aol.de
- fr-FR: recherche.aol.fr
- en-GB: search.aol.co.uk
- en-CA: search.aol.ca
In order to still offer sufficient support for language and region, the `search
keywords`_ known from Bing, ``language`` and ``loc`` (region), are added to the
search term (AOL is basically just a proxy for Bing).
.. _search keywords:
https://support.microsoft.com/en-us/topic/advanced-search-keywords-ea595928-5d63-4a0b-9c6b-0b769865e78a
"""
from urllib.parse import urlencode, unquote_plus
import typing as t
from lxml import html
from dateutil import parser
from searx.result_types import EngineResults
from searx.utils import eval_xpath_list, eval_xpath, extract_text
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://www.aol.com",
"wikidata_id": "Q2407",
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "HTML",
}
categories = ["general"]
search_type = "search" # supported: search, image, video
paging = True
safesearch = True
time_range_support = True
results_per_page = 10
base_url = "https://search.aol.com"
time_range_map = {"day": "1d", "week": "1w", "month": "1m", "year": "1y"}
safesearch_map = {0: "p", 1: "r", 2: "i"}
enable_http2 = False
def init(_):
if search_type not in ("search", "image", "video"):
raise ValueError(f"unsupported search type {search_type}")
def request(query: str, params: "OnlineParams") -> None:
language, region = (params["searxng_locale"].split("-") + [None])[:2]
if language and language != "all":
query = f"{query} language:{language}"
if region:
query = f"{query} loc:{region}"
args: dict[str, str | int | None] = {
"q": query,
"b": params["pageno"] * results_per_page + 1, # page is 1-indexed
"pz": results_per_page,
}
if params["time_range"]:
args["fr2"] = "time"
args["age"] = params["time_range"]
else:
args["fr2"] = "sb-top-search"
params["cookies"]["sB"] = f"vm={safesearch_map[params['safesearch']]}"
params["url"] = f"{base_url}/aol/{search_type}?{urlencode(args)}"
logger.debug(params)
def _deobfuscate_url(obfuscated_url: str) -> str | None:
# URL looks like "https://search.aol.com/click/_ylt=AwjFSDjd;_ylu=JfsdjDFd/RV=2/RE=1774058166/RO=10/RU=https%3a%2f%2fen.wikipedia.org%2fwiki%2fTree/RK=0/RS=BP2CqeMLjscg4n8cTmuddlEQA2I-" # pylint: disable=line-too-long
if not obfuscated_url:
return None
for part in obfuscated_url.split("/"):
if part.startswith("RU="):
return unquote_plus(part[3:])
# pattern for de-obfuscating URL not found, fall back to Yahoo's tracking link
return obfuscated_url
def _general_results(doc: html.HtmlElement) -> EngineResults:
res = EngineResults()
for result in eval_xpath_list(doc, "//div[@id='web']//ol/li[not(contains(@class, 'first'))]"):
obfuscated_url = extract_text(eval_xpath(result, ".//h3/a/@href"))
if not obfuscated_url:
continue
url = _deobfuscate_url(obfuscated_url)
if not url:
continue
res.add(
res.types.MainResult(
url=url,
title=extract_text(eval_xpath(result, ".//h3/a")) or "",
content=extract_text(eval_xpath(result, ".//div[contains(@class, 'compText')]")) or "",
thumbnail=extract_text(eval_xpath(result, ".//a[contains(@class, 'thm')]/img/@data-src")) or "",
)
)
return res
def _video_results(doc: html.HtmlElement) -> EngineResults:
res = EngineResults()
for result in eval_xpath_list(doc, "//div[contains(@class, 'results')]//ol/li"):
obfuscated_url = extract_text(eval_xpath(result, ".//a/@href"))
if not obfuscated_url:
continue
url = _deobfuscate_url(obfuscated_url)
if not url:
continue
published_date_raw = extract_text(eval_xpath(result, ".//div[contains(@class, 'v-age')]"))
try:
published_date = parser.parse(published_date_raw or "")
except parser.ParserError:
published_date = None
res.add(
res.types.LegacyResult(
{
"template": "videos.html",
"url": url,
"title": extract_text(eval_xpath(result, ".//h3")),
"content": extract_text(eval_xpath(result, ".//div[contains(@class, 'compText')]")),
"thumbnail": extract_text(eval_xpath(result, ".//img[contains(@class, 'thm')]/@src")),
"length": extract_text(eval_xpath(result, ".//span[contains(@class, 'v-time')]")),
"publishedDate": published_date,
}
)
)
return res
def _image_results(doc: html.HtmlElement) -> EngineResults:
res = EngineResults()
for result in eval_xpath_list(doc, "//section[@id='results']//ul/li"):
obfuscated_url = extract_text(eval_xpath(result, "./a/@href"))
if not obfuscated_url:
continue
url = _deobfuscate_url(obfuscated_url)
if not url:
continue
res.add(
res.types.LegacyResult(
{
"template": "images.html",
# results don't have an extra URL, only the image source
"url": url,
"title": extract_text(eval_xpath(result, ".//a/@aria-label")),
"thumbnail_src": extract_text(eval_xpath(result, ".//img/@src")),
"img_src": url,
}
)
)
return res
def response(resp: "SXNG_Response") -> EngineResults:
doc = html.fromstring(resp.text)
match search_type:
case "search":
results = _general_results(doc)
case "image":
results = _image_results(doc)
case "video":
results = _video_results(doc)
case _:
raise ValueError("unsupported search type")
for suggestion in eval_xpath_list(doc, ".//ol[contains(@class, 'searchRightBottom')]//table//a"):
results.add(results.types.LegacyResult({"suggestion": extract_text(suggestion)}))
return results

1
searx/engines/archlinux.py
View File

@@ -35,7 +35,6 @@ about = {
categories = ["it", "software wikis"]
paging = True
main_wiki = "wiki.archlinux.org"
language_support = True
def request(query, params):

2
searx/engines/arxiv.py
View File

@@ -109,7 +109,7 @@ def response(resp: "SXNG_Response") -> EngineResults:
comments_elements = eval_xpath_getindex(entry, xpath_comment, 0, default=None)
comments: str = "" if comments_elements is None else comments_elements.text
publishedDate = datetime.fromisoformat(eval_xpath_getindex(entry, xpath_published, 0).text.rstrip("Z"))
publishedDate = datetime.strptime(eval_xpath_getindex(entry, xpath_published, 0).text, "%Y-%m-%dT%H:%M:%SZ")
res.add(
res.types.Paper(

2
searx/engines/azure.py
View File

@@ -54,8 +54,8 @@ about = {
"use_official_api": True,
"require_api_key": True,
"results": "JSON",
"language": "en",
}
language = "en"
CACHE: EngineCache
"""Persistent (SQLite) key/value cache that deletes its values after ``expire``

40
searx/engines/baidu.py
View File

@@ -7,17 +7,13 @@
# There exits a https://github.com/ohblue/baidu-serp-api/
# but we don't use it here (may we can learn from).
import typing as t
from urllib.parse import urlencode
from datetime import datetime
from html import unescape
import time
import json
from searx.exceptions import SearxEngineAPIException, SearxEngineCaptchaException, SearxEngineAccessDeniedException
from searx.enginelib import EngineCache
from searx.network import get as http_get
from searx.exceptions import SearxEngineAPIException, SearxEngineCaptchaException
from searx.utils import html_to_text
about = {
@@ -27,8 +23,8 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
"language": "zh",
}
language = "zh"
paging = True
categories = []
@@ -39,31 +35,6 @@ baidu_category = 'general'
time_range_support = True
time_range_dict = {"day": 86400, "week": 604800, "month": 2592000, "year": 31536000}
image_base_url = "https://image.baidu.com/"
COOKIE_CACHE_KEY = "cookie"
COOKIE_CACHE_EXPIRATION_SECONDS = 3600
CACHE: EngineCache
"""Stores cookies from Baidu image search warmup."""
def setup(engine_settings: dict[str, t.Any]) -> bool:
global CACHE # pylint: disable=global-statement
CACHE = EngineCache(engine_settings["name"])
return True
def get_image_cookies(headers: dict[str, str]) -> dict[str, str]:
cookies: dict[str, str] | None = CACHE.get(COOKIE_CACHE_KEY)
if cookies:
return cookies
warmup = http_get(image_base_url, headers=headers, timeout=10)
cookies = dict(warmup.cookies.items())
CACHE.set(key=COOKIE_CACHE_KEY, value=cookies, expire=COOKIE_CACHE_EXPIRATION_SECONDS)
return cookies
def init(_):
if baidu_category not in ('general', 'images', 'it'):
@@ -117,9 +88,6 @@ def request(query, params):
if baidu_category == 'it':
query_params["paramList"] += f",timestamp_range={past}-{now}"
if baidu_category == 'images':
params["cookies"] = get_image_cookies(params["headers"])
params["url"] = f"{query_url}?{urlencode(query_params)}"
params["allow_redirects"] = False
return params
@@ -135,8 +103,6 @@ def response(resp):
# baidu's JSON encoder wrongly quotes / and ' characters by \\ and \'
text = text.replace(r"\/", "/").replace(r"\'", "'")
data = json.loads(text, strict=False)
if data.get("antiFlag") == 1:
raise SearxEngineAccessDeniedException(data.get("message", "Forbid spider access"))
parsers = {'general': parse_general, 'images': parse_images, 'it': parse_it}
return parsers[baidu_category](data)
@@ -186,7 +152,7 @@ def parse_images(data):
img_date = item.get("bdImgnewsDate")
publishedDate = None
if img_date:
publishedDate = datetime.fromisoformat(img_date)
publishedDate = datetime.strptime(img_date, "%Y-%m-%d %H:%M")
results.append(
{
"template": "images.html",

37
searx/engines/bilibili.py
View File

@@ -8,7 +8,6 @@ import random
import string
from urllib.parse import urlencode
from datetime import datetime, timedelta
from zoneinfo import ZoneInfo
from searx import utils
@@ -40,32 +39,6 @@ cookie = {
"home_feed_column": "4",
}
_CN_TZ = ZoneInfo("Asia/Shanghai")
# Calendar-day time filter (Asia/Shanghai); dict values are days to subtract from today.
time_range_support = True
time_range_dict = {"day": 0, "week": 6, "month": 29, "year": 364}
def _pubtime_range(time_range: str) -> tuple[int, int]:
"""Return ``(pubtime_begin_s, pubtime_end_s)`` for Bilibili's search API.
Time ranges follow Bilibili's website semantics: they are counted in
**calendar days** in China Standard Time (``Asia/Shanghai``), not as
sliding 24-hour windows. For example, ``day`` means from 00:00:00 to
23:59:59 of the current local day; ``week`` spans from 00:00:00 on the
calendar day six days ago through the end of today, and so on.
The returned Unix timestamps (seconds) map to Bilibili's
``pubtime_begin_s`` and ``pubtime_end_s`` query parameters.
"""
now = datetime.now(_CN_TZ)
pubtime_end_s = int(now.replace(hour=23, minute=59, second=59, microsecond=0).timestamp())
begin_day = now - timedelta(days=time_range_dict[time_range])
pubtime_begin_s = int(begin_day.replace(hour=0, minute=0, second=0, microsecond=0).timestamp())
return pubtime_begin_s, pubtime_end_s
def request(query, params):
query_params = {
@@ -77,16 +50,12 @@ def request(query, params):
"search_type": "video",
}
if params.get("time_range") in time_range_dict:
pubtime_begin_s, pubtime_end_s = _pubtime_range(params["time_range"])
query_params["pubtime_begin_s"] = pubtime_begin_s
query_params["pubtime_end_s"] = pubtime_end_s
params["url"] = f"{base_url}?{urlencode(query_params)}"
params["headers"]["Referer"] = "https://www.bilibili.com/"
params["headers"]["Accept"] = "application/json, text/javascript, */*; q=0.01"
params["headers"]["Referer"] = "https://www.bilibili.com"
params["cookies"] = cookie
return params
def response(resp):
search_res = resp.json()

2
searx/engines/bitchute.py
View File

@@ -44,7 +44,7 @@ def response(resp):
"url": 'https://www.bitchute.com/video/' + item['video_id'],
"content": html_to_text(item['description']),
"author": item['channel']['channel_name'],
"publishedDate": datetime.fromisoformat(item["date_published"].rstrip("Z")),
"publishedDate": datetime.strptime(item["date_published"], "%Y-%m-%dT%H:%M:%S.%fZ"),
"length": item['duration'],
"views": item['view_count'],
"thumbnail": item['thumbnail_url'],

3
searx/engines/boardreader.py
View File

@@ -34,7 +34,6 @@ about = {
categories = ["general", "social media"]
paging = True
time_range_support = True
language_support = True
base_url = "https://boardreader.com"
time_range_map = {"day": "1", "week": "7", "month": "30", "year": "365"}
@@ -104,7 +103,7 @@ def response(resp: "SXNG_Response") -> EngineResults:
title=_remove_keyword_marker(result["Subject"]),
content=_remove_keyword_marker(result["Text"]),
url=result["Url"],
publishedDate=datetime.fromisoformat(result["Published"]),
publishedDate=datetime.strptime(result["Published"], "%Y-%m-%d %H:%M:%S"),
metadata=gettext.gettext("Posted by {author}").format(author=result["Author"]),
)
)

2
searx/engines/bpb.py
View File

@@ -13,8 +13,8 @@ about = {
'use_official_api': False,
'require_api_key': False,
'results': 'JSON',
'language': 'de',
}
language = "de"
paging = True
categories = ['general']

15
searx/engines/braveapi.py
View File

@@ -31,7 +31,6 @@ from dateutil import parser
from searx.exceptions import SearxEngineAPIException
from searx.result_types import EngineResults
from searx.utils import html_to_text
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
@@ -76,7 +75,6 @@ def request(query: str, params: "OnlineParams") -> None:
"q": query,
"count": results_per_page,
"offset": (params["pageno"] - 1) * results_per_page,
"text_decorations": False,
}
# Apply time filter if specified
@@ -114,19 +112,14 @@ def response(resp: "SXNG_Response") -> EngineResults:
res = EngineResults()
data = resp.json()
for result in (data.get("web") or {}).get("results", []):
thumbnail_obj = result.get("thumbnail")
thumbnail = ""
if thumbnail_obj and not thumbnail_obj.get("logo", False):
thumbnail = thumbnail_obj.get("src") or ""
for result in data.get("web", {}).get("results", []):
res.add(
res.types.MainResult(
url=result["url"],
title=html_to_text(result["title"]),
content=html_to_text(result.get("description", "")),
title=result["title"],
content=result.get("description", ""),
publishedDate=_extract_published_date(result.get("age")),
thumbnail=thumbnail,
thumbnail=result.get("thumbnail", {}).get("src"),
),
)

115
searx/engines/chatnoir.py
View File

@@ -1,115 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Chatnoir is an open source search engine developed by Webis, a network of
researchers from the universities of Weimar, Halle and Leipzig. It supports
different different text corpora as indexes, e.g. CommonCrawl. See its
`announcement`_ for more information.
.. _announcement : https://groups.google.com/g/common-crawl/c/3o2dOHpeRxo/m/H2Osqz9dAAAJ
"""
import typing as t
from searx.exceptions import SearxEngineAPIException
from searx.extended_types import SXNG_Response
from searx.network import get, post
from searx.result_types import EngineResults
from searx.utils import html_to_text
if t.TYPE_CHECKING:
from searx.search.processors import OnlineParams
about = {
"website": "https://www.chatnoir.eu",
"official_api_documentation": "https://www.chatnoir.eu/docs/api-general",
"use_official_api": True,
"require_api_key": False,
"results": "JSON",
}
base_url = "https://www.chatnoir.eu"
categories = ["general"]
paging = True
page_size = 10
api_key = ""
"""You can optionally provide your own API key here. This one will then be used
instead of scraping an API key."""
search_index = "cw22"
"""Search index to browse in. See `the API documentation
<https://www.chatnoir.eu/docs/api-general>`_ for a full list."""
def _obtain_api_key() -> tuple[str, str, str]:
home_resp = get(base_url)
if not home_resp.ok:
raise SearxEngineAPIException("failed to obtain api key")
csrf_token = home_resp.cookies["csrftoken"]
token_resp = post(
"https://www.chatnoir.eu/?init",
headers={
"Referer": f"{base_url}/",
"X-Requested-With": "XMLHttpRequest",
"X-Csrf-Token": csrf_token,
},
cookies=home_resp.cookies,
)
if not token_resp.ok:
raise SearxEngineAPIException("failed to obtain api key")
session_id = token_resp.cookies["sessionid"]
scraped_api_key = token_resp.json()["token"]["token"]
return csrf_token, session_id, scraped_api_key
def request(query: str, params: "OnlineParams"):
if api_key:
# use user-provided API key instead of scraping one
headers = {
"Authorization": f"Bearer {api_key}",
}
params["headers"].update(headers)
else:
csrf_token, session_id, scraped_api_key = _obtain_api_key()
headers = {
"Authorization": f"Bearer {scraped_api_key}",
"X-Csrf-Token": csrf_token,
}
params["headers"].update(headers)
params["cookies"] = {"csrftoken": session_id, "sessionid": session_id}
params["url"] = f"{base_url}/api/v1/_search"
params["method"] = "POST"
json_data = {
"query": query,
"index": [
search_index,
],
"from": (params["pageno"] - 1) * page_size,
"size": page_size,
"_extended_meta": True,
}
params["json"] = json_data
def response(resp: "SXNG_Response") -> EngineResults:
res = EngineResults()
results = resp.json()["results"]
for result in results:
res.add(
res.types.MainResult(
url=result["target_uri"],
title=html_to_text(result["title"]),
content=html_to_text(result["snippet"]),
)
)
return res

4
searx/engines/chefkoch.py
View File

@@ -10,8 +10,8 @@ about = {
'use_official_api': False,
'require_api_key': False,
'results': 'JSON',
'language': 'de',
}
language = "de"
paging = True
categories = []
@@ -43,7 +43,7 @@ def response(resp):
publishedDate = None
if recipe['submissionDate']:
publishedDate = datetime.fromisoformat(result['recipe']['submissionDate'][:19])
publishedDate = datetime.strptime(result['recipe']['submissionDate'][:19], "%Y-%m-%dT%H:%M:%S")
content = [
f"Schwierigkeitsstufe (1-3): {recipe['difficulty']}",

114
searx/engines/chinaso.py
View File

@@ -19,9 +19,11 @@ Configuration
The engine has the following additional settings:
- :py:obj:`chinaso_category` (:py:obj:`ChinasoCategoryType`)
- :py:obj:`chinaso_news_source` (:py:obj:`ChinasoNewsSourceType`)
In the example below, ChinaSO is configured for news search.
In the example below, all three ChinaSO engines are using the :ref:`network
<engine network>` from the ``chinaso news`` engine.
.. code:: yaml
@@ -29,8 +31,23 @@ In the example below, ChinaSO is configured for news search.
engine: chinaso
shortcut: chinaso
categories: [news]
chinaso_category: news
chinaso_news_source: all
- name: chinaso images
engine: chinaso
network: chinaso news
shortcut: chinasoi
categories: [images]
chinaso_category: images
- name: chinaso videos
engine: chinaso
network: chinaso news
shortcut: chinasov
categories: [videos]
chinaso_category: videos
Implementations
===============
@@ -53,13 +70,26 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
"language": "zh",
}
paging = True
time_range_support = True
results_per_page = 10
categories = []
language = "zh"
ChinasoCategoryType = t.Literal['news', 'videos', 'images']
"""ChinaSo supports news, videos, images search.
- ``news``: search for news
- ``videos``: search for videos
- ``images``: search for images
In the category ``news`` you can additionally filter by option
:py:obj:`chinaso_news_source`.
"""
chinaso_category = 'news'
"""Configure ChinaSo category (:py:obj:`ChinasoCategoryType`)."""
ChinasoNewsSourceType = t.Literal['CENTRAL', 'LOCAL', 'BUSINESS', 'EPAPER', 'all']
"""Filtering ChinaSo-News results by source:
@@ -79,24 +109,39 @@ base_url = "https://www.chinaso.com"
def init(_):
if chinaso_news_source not in t.get_args(ChinasoNewsSourceType):
if chinaso_category not in ('news', 'videos', 'images'):
raise ValueError(f"Unsupported category: {chinaso_category}")
if chinaso_category == 'news' and chinaso_news_source not in t.get_args(ChinasoNewsSourceType):
raise ValueError(f"Unsupported news source: {chinaso_news_source}")
def request(query, params):
query_params = {'q': query, 'pn': params["pageno"], 'ps': results_per_page}
query_params = {"q": query}
if time_range_dict.get(params['time_range']):
query_params["stime"] = time_range_dict[params['time_range']]
query_params["etime"] = 'now'
category_config = {
'news': {'endpoint': '/v5/general/v1/web/search', 'params': {'pn': params["pageno"], 'ps': results_per_page}},
'images': {
'endpoint': '/v5/general/v1/search/image',
'params': {'start_index': (params["pageno"] - 1) * results_per_page, 'rn': results_per_page},
},
'videos': {
'endpoint': '/v5/general/v1/search/video',
'params': {'start_index': (params["pageno"] - 1) * results_per_page, 'rn': results_per_page},
},
}
if chinaso_news_source != 'all':
if chinaso_news_source == 'EPAPER':
query_params["type"] = 'EPAPER'
category_config['news']['params']["type"] = 'EPAPER'
else:
query_params["cate"] = chinaso_news_source
category_config['news']['params']["cate"] = chinaso_news_source
params["url"] = f"{base_url}/v5/general/v1/web/search?{urlencode(query_params)}"
query_params.update(category_config[chinaso_category]['params'])
params["url"] = f"{base_url}{category_config[chinaso_category]['endpoint']}?{urlencode(query_params)}"
cookie = {
"uid": base64.b64encode(secrets.token_bytes(16)).decode("utf-8"),
}
@@ -111,13 +156,12 @@ def response(resp):
except Exception as e:
raise SearxEngineAPIException(f"Invalid response: {e}") from e
# Upstream returns {'status': 0, 'msg': 'empty result', 'data': {}} when there
# are no results; this is a valid empty result rather than an API error.
if not isinstance(data, dict) or "data" not in data:
raise SearxEngineAPIException("Invalid response")
if not data["data"]:
return []
parsers = {'news': parse_news, 'images': parse_images, 'videos': parse_videos}
return parsers[chinaso_category](data)
def parse_news(data):
results = []
if not data.get("data", {}).get("data"):
raise SearxEngineAPIException("Invalid response")
@@ -139,3 +183,47 @@ def response(resp):
}
)
return results
def parse_images(data):
results = []
if not data.get("data", {}).get("arrRes"):
raise SearxEngineAPIException("Invalid response")
for entry in data["data"]["arrRes"]:
results.append(
{
'url': entry["web_url"],
'title': html_to_text(entry["title"]),
'content': html_to_text(entry.get("ImageInfo", "")),
'template': 'images.html',
'img_src': entry["url"].replace("http://", "https://"),
'thumbnail_src': entry["largeimage"].replace("http://", "https://"),
}
)
return results
def parse_videos(data):
results = []
if not data.get("data", {}).get("arrRes"):
raise SearxEngineAPIException("Invalid response")
for entry in data["data"]["arrRes"]:
published_date = None
if entry.get("VideoPubDate"):
try:
published_date = datetime.fromtimestamp(int(entry["VideoPubDate"]))
except (ValueError, TypeError):
pass
results.append(
{
'url': entry["url"],
'title': html_to_text(entry["raw_title"]),
'template': 'videos.html',
'publishedDate': published_date,
'thumbnail': entry["image_src"].replace("http://", "https://"),
}
)
return results

1
searx/engines/dailymotion.py
View File

@@ -40,7 +40,6 @@ categories = ["videos"]
paging = True
page_size = 10
language_support = True
time_range_support = True
time_delta_dict = {
"day": timedelta(days=1),

14
searx/engines/demo_offline.py
View File

@@ -24,7 +24,7 @@ import typing as t
import json
from searx.result_types import EngineResults
from searx.enginelib import EngineCache, EngineAbout
from searx.enginelib import EngineCache
if t.TYPE_CHECKING:
from searx.search.processors import RequestParams
@@ -35,11 +35,13 @@ categories = ["general"]
disabled = True
timeout = 2.0
language = "en"
about = EngineAbout(
results="JSON",
description="Demo offline engine Engine with results in the English language.",
)
about = {
"wikidata_id": None,
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
# if there is a need for globals, use a leading underline
_my_offline_engine: str = ""

17
searx/engines/demo_online.py
View File

@@ -25,7 +25,6 @@ import typing as t
from urllib.parse import urlencode
from searx.result_types import EngineResults
from searx.enginelib import EngineAbout
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
@@ -44,14 +43,14 @@ page_size = 20
search_api = "https://api.artic.edu/api/v1/artworks/search"
image_api = "https://www.artic.edu/iiif/2/"
about = EngineAbout(
website="https://www.artic.edu",
wikidata_id="Q239303",
official_api_documentation="http://api.artic.edu/docs/",
use_official_api=True,
require_api_key=False,
results="JSON",
)
about = {
"website": "https://www.artic.edu",
"wikidata_id": "Q239303",
"official_api_documentation": "http://api.artic.edu/docs/",
"use_official_api": True,
"require_api_key": False,
"results": "JSON",
}
# if there is a need for globals, use a leading underline

2
searx/engines/destatis.py
View File

@@ -11,8 +11,8 @@ about = {
'use_official_api': False,
'require_api_key': False,
'results': 'HTML',
'language': 'de',
}
language = "de"
categories = []
paging = True

101
searx/engines/dogpile.py
View File

@@ -1,101 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Dogpile is a metasearch engine by the American advertising company `System1`_.
.. _System1: https://system1.com/
"""
import typing as t
from datetime import datetime, timezone
import html
from searx.utils import format_duration, html_to_text, humanize_number
from searx.result_types import EngineResults
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://www.dogpile.com",
"wikidata_id": "Q3595363",
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
paging = True
safesearch = True
categories = ["general"]
dogpile_categ = "search"
"""Category to search in. Can be either "search", "images", "videos" or "news"."""
base_url = "https://www.dogpile.com"
safe_search_map = {0: "none", 1: "moderate", 2: "heavy"}
def init(_):
if dogpile_categ not in ("search", "images", "videos", "news"):
raise ValueError("invalid search type: %s" % dogpile_categ)
def request(query: str, params: "OnlineParams"):
params["url"] = f"{base_url}/api/{dogpile_categ}"
params["method"] = "POST"
params["json"] = {"q": query, "qadf": safe_search_map[params["safesearch"]], "page": params["pageno"]}
return params
def response(resp: "SXNG_Response"):
res = EngineResults()
json_resp = resp.json()
for result in json_resp["results"]:
if dogpile_categ == "search":
res.add(
res.types.MainResult(
url=result["clickUrl"],
title=html_to_text(result["title"]),
content=html_to_text(result["description"]),
)
)
elif dogpile_categ == "news":
res.add(
res.types.MainResult(
url=result["clickUrl"],
title=html_to_text(html.unescape(result["title"])),
content=html_to_text(html.unescape(result["description"])),
thumbnail=result["thumbnailUrl"],
publishedDate=datetime.fromtimestamp(result["date"], tz=timezone.utc),
)
)
elif dogpile_categ == "videos":
res.add(
res.types.LegacyResult(
template="videos.html",
url=result["clickUrl"],
title=html_to_text(result["title"]),
content=html_to_text(result["description"]),
thumbnail=result["thumbnailUrl"],
publishedDate=datetime.fromisoformat(result["publishDate"]),
length=format_duration(result["duration"]),
views=humanize_number(result["viewCount"]),
)
)
elif dogpile_categ == "images":
res.add(
res.types.Image(
url=result["altClickUrl"],
title=html_to_text(result["title"]),
content=html_to_text(result["description"]),
img_src=result["clickUrl"],
thumbnail_src=result["thumbnailUrl"],
resolution=f"{result['width']}x{result['height']}",
img_format=result["format"],
)
)
return res

1
searx/engines/duckduckgo.py
View File

@@ -203,7 +203,6 @@ about: dict[str, str | bool] = {
categories: list[str] = ["general", "web"]
paging: bool = True
time_range_support: bool = True
language_support = True
safesearch: bool = True
"""DDG-lite: user can't select but the results are filtered."""

3
searx/engines/duckduckgo_extra.py
View File

@@ -28,7 +28,6 @@ about = {
"require_api_key": False,
"results": "JSON (site requires js to get images)",
}
language_support = True
# engine dependent config
categories = []
@@ -42,9 +41,7 @@ safesearch_cookies = {0: "-2", 1: None, 2: "1"}
safesearch_args = {0: "1", 1: None, 2: "1"}
search_path_map = {"images": "i", "videos": "v", "news": "news"}
_HTTP_User_Agent: str = gen_useragent()
send_accept_language_header = False
def init(engine_settings: dict[str, t.Any]):

1
searx/engines/duckduckgo_weather.py
View File

@@ -26,7 +26,6 @@ about = {
"require_api_key": False,
"results": "JSON",
}
language_support = True
# engine dependent config
categories = ["weather"]

156
searx/engines/duckduckgo_web.py
View File

@@ -1,156 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""DuckDuckGo Web (general)
This implementation fetches the link to the first API page
(i.e. ``links.duckduckgo.com/d.js?...``) from duckduckgo.com and uses the ``n``
parameter of the API to fetch all subsequent pages.
This also means that it's not possible to immediately search for the third
page - the first and the second page would need to be loaded first.
The reason why we can't just normally use the `vqd` value is that the API URLs
require an additional parameter `dp` which seems generated at server-side, so we
can't build it ourselves and must scrape it from the HTML pages.
"""
import typing as t
from urllib.parse import quote_plus
from lxml import html
from searx.utils import html_to_text, gen_useragent, extract_text, eval_xpath
from searx.result_types import EngineResults
from searx.enginelib import EngineCache
from searx.network import get
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://duckduckgo.com/",
"wikidata_id": "Q12805",
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
# engine dependent config
categories = ["general"]
paging = True
_HTTP_User_Agent: str = gen_useragent()
base_url = "https://duckduckgo.com"
CACHE: EngineCache
"""Cache to store the API URLs for combinations of (query, page)."""
def setup(engine_settings: dict[str, str]):
global CACHE # pylint:disable=global-statement
CACHE = EngineCache(engine_settings["name"])
return CACHE
def _fetch_first_page_link(
query: str,
headers: dict[str, str],
):
"""Search for a::
<link id="deep_preload_link" rel="preload" as="script"
href="https://links.duckduckgo.com/d.js?q=rust&t=D&l=us-en&s=0&a=h_&ct=DE&vqd=VQD_VALUE&bing_market=en-US&p_ent=&ex=-1&dp=LONG_TOKEN
>
This points to the first page
""" # pylint:disable=line-too-long
cache_key = _cache_key(query, 1)
cached: str | None = CACHE.get(cache_key)
if cached:
return cached
resp = get(
url=f"{base_url}/?q={quote_plus(query)}&t=h_&ia=web",
headers=headers,
timeout=2,
)
if resp.status_code != 200:
logger.error("vqd: got HTTP %s from duckduckgo.com", resp.status_code)
dom = html.fromstring(resp.text)
first_page_link = extract_text(eval_xpath(dom, "//link[@id='deep_preload_link']/@href"))
if not first_page_link:
logger.error("vqd: failed to load first page JS url from ddg response (return empty string)")
return ""
logger.debug("got link to first page from duckduckgo.com request: '%s'", first_page_link)
CACHE.set(cache_key, first_page_link, expire=7200)
return first_page_link
def _cache_key(query: str, pageno: int) -> str:
return f"nextpage_url|{query}|{pageno}"
def request(query: str, params: "OnlineParams") -> None:
if len(query) >= 500:
# DDG does not accept queries with more than 499 chars
params["url"] = None
return
headers = params["headers"]
# The vqd value is generated from the query and the UA header. To be able
# to reuse the vqd value, the UA header must be static.
headers["User-Agent"] = _HTTP_User_Agent
headers["Accept"] = "*/*"
headers["Referer"] = f"{base_url}/"
headers["Host"] = "duckduckgo.com"
# Sec-Fetch headers are required to not get blocked when sending a Firefox user agent
headers["Sec-Fetch-Dest"] = "script"
headers["Sec-Fetch-Mode"] = "no-cors"
headers["Sec-Fetch-Site"] = "same-site"
api_url = ""
if params["pageno"] > 1:
api_url = CACHE.get(_cache_key(query, params["pageno"]))
else:
api_url = _fetch_first_page_link(query, headers)
if not api_url:
params["url"] = None
return
params["url"] = api_url.replace("/d.js?", "/d.js?o=json&")
# TODO: support safesearch, timerange and engine traits # pylint:disable=fixme
def response(resp: "SXNG_Response"):
res = EngineResults()
res_json = resp.json()
for result in res_json["results"]:
if "u" not in result:
continue
res.add(
res.types.MainResult(url=result["u"], title=html_to_text(result["t"]), content=html_to_text(result["a"]))
)
# link to next page
next_page_path = res_json["results"][-1].get("n")
if next_page_path:
CACHE.set(
_cache_key(resp.search_params["query"], resp.search_params["pageno"] + 1),
base_url + next_page_path,
expire=60 * 60,
)
return res

2
searx/engines/duden.py
View File

@@ -14,8 +14,8 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": 'HTML',
"language": 'de',
}
language = "de"
categories = ['dictionaries']
paging = True

2
searx/engines/elasticsearch.py
View File

@@ -55,7 +55,7 @@ about = {
'official_api_documentation': 'https://www.elastic.co/guide/en/elasticsearch/reference/current/search-search.html',
'use_official_api': True,
'require_api_key': False,
"results": "JSON",
'format': 'JSON',
}
base_url = 'http://localhost:9200'

168
searx/engines/exaapi.py
View File

@@ -1,168 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Engine to search using the official `Exa Search API`_. Exa is a search engine for AI agents.
.. _Exa Search API: https://exa.ai/docs/reference/search
Configuration
=============
The engine has the following mandatory setting:
- :py:obj:`api_key`
You can obtain an API key from the `API Key section <https://dashboard.exa.ai/api-keys>`_ in the Exa dashboard.
Optional settings are:
- :py:obj:`results_per_page`
- :py:obj:`search_type`
- :py:obj:`content_mode`
- :py:obj:`content_max_characters`
.. code:: yaml
- name: exaapi
engine: exaapi
shortcut: exa
api_key: "..."
results_per_page: 10
search_type: auto
content_mode: highlights
inactive: false
The API supports SafeSearch and region-aware results.
"""
import typing as t
from dateutil import parser
from searx.exceptions import SearxEngineAPIException
from searx.result_types import EngineResults
from searx.utils import html_to_text
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
SearchType = t.Literal["fast", "auto", "instant", "deep", "deep-lite", "deep-reasoning"]
ContentMode = t.Literal["highlights", "text"]
about = {
"website": "https://exa.ai",
"wikidata_id": None,
"official_api_documentation": "https://exa.ai/docs/reference/search",
"use_official_api": True,
"require_api_key": True,
"results": "JSON",
}
api_key: str = ""
"""API key for Exa Search API (required)."""
categories = ["general", "web"]
safesearch = True
base_url = "https://api.exa.ai/search"
results_per_page: int = 10
"""Maximum number of results per request. Value must be between 1 and 100, default is 10."""
search_type: SearchType = "auto"
"""Search type. Default is auto, see documentation for more information."""
content_mode: ContentMode = "highlights"
"""Content to request from the API: ``highlights`` (excerpts) or ``text`` (page text)."""
content_max_characters: int = 500
"""Maximum characters for the requested content."""
def init(_):
if not api_key:
raise SearxEngineAPIException("No API key provided")
if not 1 <= results_per_page <= 100:
raise ValueError("results_per_page must be between 1 and 100")
if search_type not in t.get_args(SearchType):
raise ValueError(f"Unsupported search type: {search_type}")
if content_mode not in t.get_args(ContentMode):
raise ValueError(f"Unsupported content mode: {content_mode}")
if content_max_characters < 1:
raise ValueError("content_max_characters must be at least 1")
def _contents_payload() -> dict[str, t.Any]:
if content_mode == "text":
return {"text": {"maxCharacters": content_max_characters, "stripLinks": True}}
return {"highlights": {"maxCharacters": content_max_characters}}
def _extract_content(result: dict[str, t.Any]) -> str:
if content_mode == "text":
return html_to_text(result.get("text") or "")
return html_to_text(" ".join(result.get("highlights") or []))
def request(query: str, params: "OnlineParams") -> None:
"""Create the API request."""
body: dict[str, t.Any] = {
"query": query,
"type": search_type,
"numResults": results_per_page,
"contents": _contents_payload(),
}
# Apply SafeSearch if enabled
if params["safesearch"]:
body["moderation"] = True
# Apply region-aware results if specified
locale_parts = params["searxng_locale"].split("-")
region = locale_parts[-1]
if len(locale_parts) > 1:
body["userLocation"] = region.upper()
params["url"] = base_url
params["method"] = "POST"
params["headers"]["x-api-key"] = api_key
params["json"] = body
def _extract_published_date(value: str | None):
"""Extract and parse the published date from the API response.
Args:
value: Raw date string from the API
Returns:
Parsed datetime object or None if parsing fails
"""
if not value:
return None
try:
return parser.parse(value)
except (parser.ParserError, TypeError, OverflowError):
return None
def response(resp: "SXNG_Response") -> EngineResults:
"""Process the API response and return results."""
res = EngineResults()
for result in resp.json().get("results", []):
url = result.get("url")
if not url:
continue
res.add(
res.types.MainResult(
url=url,
title=html_to_text(result.get("title") or url),
content=_extract_content(result),
thumbnail=result.get("image") or "",
publishedDate=_extract_published_date(result.get("publishedDate")),
author=result.get("author") or "",
)
)
return res

118
searx/engines/findfiles.py
View File

@@ -1,118 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""FindFiles.net_ is a Germany-based file search engine.
FindFiles.net_ is a specialized file search engine designed to help you search
files online with precision. Unlike traditional search engines that mainly index
web pages, FindFiles focuses on finding real files on the internet - including
PDFs, documents, archives, videos, datasets, and more.
.. _FindFiles.net: https://findfiles.net
"""
from os.path import basename
from urllib.parse import urlencode
import typing as t
from lxml import html
from searx.result_types import EngineResults
from searx.utils import extract_text, eval_xpath, eval_xpath_list
if t.TYPE_CHECKING:
from extended_types import SXNG_Response
from search.processors import OnlineParams
about = {
"website": "https://findfiles.net",
"wikidata_id": None,
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "HTML",
}
base_url = "https://findfiles.net"
categories = ["files"]
paging = True
safeserach = True
safesearch_map = {
0: "contentguard.off",
1: "contentguard.moderate",
2: "contentguard.strict",
}
FindFilesCategory = t.Literal[
"all",
"document",
"text",
"image",
"audio",
"video",
]
FINDFILES_CATEGORIES = t.get_args(FindFilesCategory)
findfiles_categ: FindFilesCategory = "all"
"""Category to search in."""
def setup(_: dict[str, t.Any]) -> bool:
if findfiles_categ not in FINDFILES_CATEGORIES:
raise ValueError("invalid category: %s" % findfiles_categ)
return True
def request(query: str, params: "OnlineParams") -> None:
args = {
"query": query,
"contentguard": safesearch_map[params["safesearch"]],
"page": params["pageno"],
}
# the language in the path doesn't change anything about the results, it
# only changes the UI
params["url"] = f"{base_url}/en/serp/{findfiles_categ}/?{urlencode(args)}"
def response(resp: "SXNG_Response") -> EngineResults:
res = EngineResults()
dom = html.fromstring(resp.text)
if findfiles_categ == "image":
for result in eval_xpath_list(
dom, "//div[contains(@class, 'image-mosaic')]/div[contains(@class, 'image-item')]"
):
res.add(
res.types.Image(
url=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/a/@href")) or "",
title=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/a")) or "",
thumbnail_src=extract_text(eval_xpath(result, ".//img/@src")) or "",
)
)
elif findfiles_categ == "video":
for result in eval_xpath_list(
dom, "//div[contains(@class, 'video-mosaic')]/div[contains(@class, 'video-item')]"
):
video_src = extract_text(eval_xpath(result, ".//video/@src")) or ""
res.add(
res.types.LegacyResult(
template="videos.html",
url=video_src,
title=extract_text(eval_xpath(result, ".//div[contains(@class, 'caption')]/span")) or "",
iframe_src=video_src or "",
)
)
else:
for result in eval_xpath_list(dom, "//ol/li[contains(@class, 'result-item')]/article"):
filename = basename(extract_text(eval_xpath(result, ".//h3")) or "")
res.add(
res.types.File(
url=extract_text(eval_xpath(result, ".//h3/a/@href")) or "",
title=filename,
content=" ".join(extract_text(el) or "" for el in eval_xpath_list(result, "./div/span")),
filename=filename,
size=extract_text(eval_xpath(result, "(.//span[@id])[1]")) or "",
embedded=extract_text(eval_xpath(result, ".//audio/@src")) or "",
)
)
return res

2
searx/engines/findthatmeme.py
View File

@@ -37,7 +37,7 @@ def response(resp):
for item in search_res:
img = 'https://s3.thehackerblog.com/findthatmeme/' + item['image_path']
thumb = 'https://s3.thehackerblog.com/findthatmeme/thumb/' + item.get('thumbnail', '')
date = datetime.fromisoformat(item["updated_at"].split("T")[0])
date = datetime.strptime(item["updated_at"].split("T")[0], "%Y-%m-%d")
formatted_date = datetime.fromtimestamp(date.timestamp())
results.append(

169
searx/engines/fireball.py
View File

@@ -1,169 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Fireball_ is a Germany-based, privacy-focused search engine.
It likely doesn't have its own index, but it's unclear where its results come
from.
.. _Fireball: https://fireball.com
"""
import typing as t
from datetime import datetime
from urllib.parse import urlencode
from searx.enginelib import EngineCache
from searx.exceptions import SearxEngineAPIException
from searx.extended_types import SXNG_Response
from searx.result_types import EngineResults
from searx.network import post
from searx.utils import html_to_text
if t.TYPE_CHECKING:
from searx.search.processors import OnlineParams
about = {
"website": "https://fireball.com",
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
base_url = "https://fireball.com"
categories = ["general"]
fireball_category = "web" # values: "web", "news", "videos"
paging = False
safesearch = True
safe_search_map = {0: "off", 1: "moderate", 2: "strict"}
CACHE: EngineCache
"""Cache to store the settings cookie (contains e.g. language, safesearch, ...)."""
CACHE_VALID_DURATION = 30 * 24 * 3600 # one month, same as website
"""Duration how long settings cookies are valid."""
def init(engine_settings: dict[str, t.Any]):
global CACHE # pylint: disable=global-statement
CACHE = EngineCache(engine_settings["name"])
if fireball_category not in ("web", "news", "videos"):
raise ValueError(f"Unsupported category: {fireball_category}")
def _cache_key(fireball_settings: dict[str, str]) -> str:
return f"fireball_settings_{fireball_settings['safesearch']}_{fireball_settings['market']}"
def _get_search_settings_cookie(params: 'OnlineParams') -> str:
"""Get a 'fireball' cookie for the given locale and safesearch setting set
in params."""
# the language is set by only specifying the search country on their
# website, they only list DE and US, but in fact it supports much more
# countries
country = "US"
if params["searxng_locale"] != "all":
language_parts = params["searxng_locale"].split("-")
country = language_parts[-1].upper()
fireball_settings = {
"action": "save",
"language": "en", # language is irrelevant, only changes UI language
"market": country,
"adprovider": "automatic",
"target": "_blank",
"tiles": "on",
"safesearch": safe_search_map[params["safesearch"]],
}
cache_key = _cache_key(fireball_settings)
cached_cookie = CACHE.get(cache_key)
if cached_cookie:
return cached_cookie
resp = post("https://fireball.com/settings", data=fireball_settings)
if not resp.ok:
raise SearxEngineAPIException("failed to obtain cookie for settings")
cookie = resp.cookies.get("fireball")
if not cookie:
raise SearxEngineAPIException("failed to obtain cookie for settings")
CACHE.set(cache_key, cookie, expire=CACHE_VALID_DURATION)
return cookie
def request(query: str, params: "OnlineParams"):
# no matter the category, the request is always the same, i.e. we get all
# different categories with one HTTP request
args = {
"f": "web",
"q": query,
}
params["url"] = f"{base_url}/getResults/?{urlencode(args)}"
params["cookies"]["fireball"] = _get_search_settings_cookie(params)
# referer header has to be set, otherwise the requests get blocked
params["headers"]["Referer"] = f"{base_url}/search?{urlencode(args)}"
def response(resp: "SXNG_Response") -> EngineResults:
res = EngineResults()
json_data = resp.json()
for result in json_data.get(fireball_category, {}).get("results", []):
published_date = None
if result.get("page_age"):
published_date = datetime.fromisoformat(result["page_age"])
if fireball_category == "web":
res.add(
res.types.MainResult(
url=result["url"],
title=html_to_text(result["title"]),
content=html_to_text(result["description"]),
publishedDate=published_date,
)
)
elif fireball_category == "news":
thumbnail: str | None = None
if result.get("thumbnail"):
thumbnail = result["thumbnail"]["src"]
res.add(
res.types.MainResult(
url=result["url"],
title=html_to_text(result["title"]),
content=html_to_text(result["description"]),
thumbnail=thumbnail or "",
publishedDate=published_date,
)
)
elif fireball_category == "videos":
length = None
if result.get("video"):
length = result["video"].get("duration")
res.add(
res.types.LegacyResult(
{
"template": "videos.html",
"url": result["url"],
"title": html_to_text(result["title"]),
"content": html_to_text(result["description"]),
"thumbnail": result.get("thumbnail", {}).get("original"),
"length": length,
"publishedDate": published_date,
}
)
)
return res

6
searx/engines/flaticon.py
View File

@@ -53,17 +53,13 @@ def response(resp: "SXNG_Response"):
result: dict[str, str] # TBH: dict[str, t.Any]
for result in resp.json()["items"]:
tags = [
tag_info["tag"] for tag_info in result["tags"] if tag_info["tag"] # pyright: ignore[reportArgumentType]
]
res.add(
res.types.Image(
title=result["name"],
content=", ".join(tags),
content=", ".join([tag["tag"] for tag in result["tags"]]), # pyright: ignore[reportArgumentType]
url=_fix_url(result["slug"]),
thumbnail_src=_fix_url(result["png"]),
img_src=_fix_url(result["png512"]),
img_format="PNG",
author=result["team_name"],
)
)

2
searx/engines/fyyd.py
View File

@@ -47,7 +47,7 @@ def response(resp: "SXNG_Response"):
title=result["title"],
content=result["description"],
thumbnail=result["smallImageURL"],
publishedDate=datetime.fromisoformat(result["status_since"]),
publishedDate=datetime.strptime(result["status_since"], "%Y-%m-%d %H:%M:%S"),
metadata=f"Rank: {result['rank']} || {result['episode_count']} episodes",
)
)

2
searx/engines/geizhals.py
View File

@@ -27,8 +27,8 @@ about = {
'official_api_documentation': None,
'require_api_key': False,
'results': 'HTML',
'language': 'de',
}
language = "de"
paging = True
categories = ['shopping']

127
searx/engines/giphy.py
View File

@@ -1,127 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Giphy (images)"""
import random
from urllib.parse import urlencode
import re
import typing as t
from lxml import html
from searx.enginelib import EngineCache
from searx.exceptions import SearxEngineAPIException
from searx.network import get
from searx.result_types import EngineResults
from searx.result_types.image import ImageRef
from searx.utils import eval_xpath_list, humanize_bytes
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://giphy.com",
"wikidata_id": "Q17054335",
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
base_url = "https://giphy.com"
api_url = "https://api.giphy.com"
categories = ["images"]
paging = True
page_size = 15
GiphyCategs = t.Literal["gifs", "stickers", "clips"]
giphy_categ: GiphyCategs = "gifs"
"""Giphy category to search in."""
CACHE: EngineCache
"""Cache for storing the extracted api key."""
_GIPHY_API_KEY_RE = re.compile(r"[Aa]piKey\s*:\s*\"(\w+)\"")
def setup(engine_settings: dict[str, str]) -> bool:
if giphy_categ not in t.get_args(GiphyCategs):
raise ValueError("invalid category: %s" % giphy_categ)
global CACHE # pylint: disable=global-statement
CACHE = EngineCache(engine_settings["name"])
return True
def _get_api_key() -> str:
"""
Extract the Giphy API key from the JavaScript code. There are different API keys
(e.g. for mobile, desktop, ...), so we just pick a random one of these.
"""
cached = CACHE.get("api_key")
if cached:
return cached
homepage_resp = get(base_url)
homepage_doc = html.fromstring(homepage_resp.text)
for script_src in eval_xpath_list(homepage_doc, "//script[contains(@src, 'layout')]/@src"):
script_resp = get(base_url + script_src)
api_keys = _GIPHY_API_KEY_RE.findall(script_resp.text)
if api_keys:
api_key = random.choice(api_keys)
CACHE.set("api_key", api_key, expire=60 * 60 * 6) # 6 hours
return api_key
raise SearxEngineAPIException("failed to extract api keys")
def request(query: str, params: "OnlineParams") -> None:
args = {
"q": query,
"api_key": _get_api_key(),
"limit": page_size,
"offset": (params["pageno"] - 1) * page_size,
"type": giphy_categ,
}
params["url"] = f"{api_url}/v1/{giphy_categ}/search?{urlencode(args)}"
def response(resp: "SXNG_Response"):
res = EngineResults()
result: dict[str, t.Any]
for result in resp.json()["data"]:
img = result['images']['original']
formats = [
ImageRef(url=img["mp4"], subtype="mp4"), # type: ignore
ImageRef(url=img["webp"], subtype="webp"), # type: ignore
]
thumb = (
result["images"].get("downsized")
or result["images"].get("downsized_medium")
or result["images"].get("downsized_small")
or result["images"].get("downsized_large")
)
res.add(
res.types.Image(
title=result["title"],
content=", ".join(result.get("tags", [])),
url=result["url"],
thumbnail_src=thumb.get("url") or img["url"],
img_src=img["url"],
resolution=f"{img['width']}x{img['height']}",
img_format="GIF",
formats=formats,
author=result["username"],
filesize=humanize_bytes(int(img["size"])),
source=result.get("source_tld") or "",
)
)
return res

9
searx/engines/gmx.py
View File

@@ -10,12 +10,10 @@ import time
import typing as t
from urllib.parse import urlencode
from lxml import html
from searx.result_types import EngineResults
from searx.exceptions import SearxEngineCaptchaException
from searx.extended_types import SXNG_Response
from searx.utils import extr, gen_useragent, html_to_text, eval_xpath
from searx.utils import extr, gen_useragent, html_to_text
from searx.network import get
if t.TYPE_CHECKING:
@@ -42,11 +40,6 @@ time_range_map = {"day": "d", "week": "w", "month": "m", "year": "y"}
def _get_page_hash(query: str, page: int, headers: dict[str, str]) -> str:
resp = get(f"{base_url}/web/result?q={query}&page={page}", headers=headers)
# detect captcha (if any)
doc = html.fromstring(resp.text)
if eval_xpath(doc, "//*[@id='spam-messages']"):
raise SearxEngineCaptchaException()
# the text we search for looks like:
# load("/desk?lang="+eV.p.param['hl']+"&q="+eV['p']['q_encode']+"&page=5&h=aa45603&t=177582576&origin=web&comp=web_serp_pag&p=gmx-com&sp=&lr="+eV.p.param['lr0']+"&mkt="+eV.p.param['mkt0']+"&family="+eV.p.param['familyFilter']+"&fcons="+eV.p.perm.fCons,"google", "eMMO", "eMH","eMP"); # pylint: disable=line-too-long
return extr(resp.text, "&h=", "&t=")

5
searx/engines/google.py
View File

@@ -32,6 +32,7 @@ from searx.utils import (
eval_xpath_getindex,
eval_xpath_list,
extract_text,
gen_gsa_useragent,
)
if t.TYPE_CHECKING:
@@ -56,7 +57,6 @@ max_page = 50
.. _Google max 50 pages: https://github.com/searxng/searxng/issues/2982
"""
time_range_support = True
language_support = True
safesearch = True
time_range_dict = {"day": "d", "week": "w", "month": "m", "year": "y"}
@@ -196,7 +196,7 @@ def get_google_info(params: "OnlineParams", eng_traits: EngineTraits) -> dict[st
# https://developers.google.com/custom-search/docs/xml_results_appendices#interfaceLanguages
# https://github.com/searxng/searxng/issues/2515#issuecomment-1607150817
ret_val["params"]["hl"] = f"{lang_code}"
ret_val["params"]["hl"] = f"{lang_code}-{country}"
# lr parameter:
# The lr (language restrict) parameter restricts search results to
@@ -267,6 +267,7 @@ def get_google_info(params: "OnlineParams", eng_traits: EngineTraits) -> dict[st
# HTTP headers
ret_val["headers"]["Accept"] = "*/*"
ret_val["headers"]["User-Agent"] = gen_gsa_useragent()
# Cookies

190
searx/engines/google_cse.py
View File

@@ -1,190 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Google Custom Search Engine"""
import datetime
import typing as t
from json import loads
from urllib.parse import urlencode
from searx.enginelib import EngineCache
from searx.exceptions import SearxEngineAPIException, SearxEngineTooManyRequestsException
from searx.network import get
from searx.result_types import EngineResults, Result, MainResult, Image
from searx.engines.google import fetch_traits # pylint: disable=unused-import
from searx.engines.google import filter_mapping, get_google_info
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://www.google.com",
"wikidata_id": "Q2233943",
"official_api_documentation": "https://developers.google.com/custom-search/docs/element",
"use_official_api": False,
"require_api_key": False,
"results": "JSONP",
"description": "Platform for creating custom search engines based on Google Search.",
}
categories = ["general", "web"]
paging = True
max_page = 5
page_size = 20
time_range_support = True
language_support = True
safesearch = True
GoogleCategType = t.Literal["", "image"]
google_categ: GoogleCategType = ""
"""Google CSE category. Set to ``""`` for web search."""
CX = "partner-pub-8993703457585266:4862972284" # blackle.com
CACHE: EngineCache
def setup(engine_settings: dict[str, t.Any]) -> bool:
global CACHE # pylint: disable=global-statement
if google_categ not in t.get_args(GoogleCategType):
raise ValueError("invalid google cse category: %s" % google_categ)
CACHE = EngineCache(engine_settings["name"])
return True
def _cse_token() -> dict[str, str]:
token: dict[str, str] = CACHE.get(CX)
if token:
return token
resp = get(f"https://www.google.com/cse/cse.js?cx={CX}", timeout=10)
if not resp.ok:
raise SearxEngineAPIException("failed to obtain cse token")
end = resp.text.rfind("});")
start = resp.text.rfind("({")
opts: dict[str, str] = loads(resp.text[start + 1 : end + 1])
cse_tok = opts.get("cse_token")
if not cse_tok:
raise SearxEngineAPIException("failed to obtain cse token")
exp = opts.get("exp")
token = {
"cse_tok": cse_tok,
"cselibv": opts.get("cselibVersion", ""),
"exp": ",".join(exp) if exp else "",
}
CACHE.set(CX, token, expire=3600)
return token
def _get_start_and_end_date_str(time_range: str) -> tuple[str, str]:
time_range_map = {"day": 1, "week": 7, "month": 30, "year": 365}
end_date = datetime.datetime.now()
start_date = end_date - datetime.timedelta(days=time_range_map[time_range])
return start_date.strftime("%Y%m%d"), end_date.strftime("%Y%m%d")
def request(query: str, params: "OnlineParams") -> None:
token = _cse_token()
google_info = get_google_info(params, traits)
info: dict[str, str] = google_info["params"]
args = {
"rsz": "filtered_cse",
"num": str(page_size),
"hl": info["hl"],
"cselibv": token["cselibv"],
"cx": CX,
"q": query,
"safe": filter_mapping[params["safesearch"]],
"cse_tok": token["cse_tok"],
"callback": "_",
"rurl": "",
"searchtype": google_categ,
}
if params["time_range"]:
start_date, end_date = _get_start_and_end_date_str(params["time_range"])
args["sort"] = f"date:r:{start_date}:{end_date}"
if info.get("lr"):
args["lr"] = info["lr"]
if info.get("cr"):
args["cr"] = info["cr"]
if google_info["country"] not in (None, "ZZ"):
args["gl"] = google_info["country"]
if token["exp"]:
args["exp"] = token["exp"]
start = (params["pageno"] - 1) * page_size
if start:
args["start"] = str(start)
params["url"] = "https://cse.google.com/cse/element/v1?" + urlencode(args)
params["cookies"] = google_info["cookies"]
params["headers"].update(google_info["headers"])
params["headers"]["Referer"] = "https://cse.google.com/"
def response(resp: "SXNG_Response") -> EngineResults:
json_resp = resp.text[resp.text.find("{") : resp.text.rfind("}") + 1]
data = loads(json_resp)
# not the real types, but a sufficient approximation
item: dict[str, str]
error: dict[str, str | int]
if error := data.get("error"):
message = error.get("message", "unknown error")
if error.get("code") == 429:
raise SearxEngineTooManyRequestsException(message=f"google cse: {message}")
raise SearxEngineAPIException(f"google cse: {message}")
results = EngineResults()
for item in data.get("results", []):
res: Result | None
if google_categ == "":
res = web_item(item)
elif google_categ == "image":
res = img_item(item)
if res is not None:
results.add(res)
return results
def web_item(item: dict[str, str]) -> MainResult | None:
url = item.get("unescapedUrl")
if not url:
return None
return MainResult(
url=url,
title=item.get("titleNoFormatting", ""),
content=item.get("contentNoFormatting", ""),
thumbnail=item.get("richSnippet", {}).get("cseThumbnail", {}).get("src", ""), # type: ignore
)
def img_item(item: dict[str, str]) -> Image | None:
resolution = ""
if item.get("height") and item.get("width"):
resolution = f"{item['width']}x{item['height']}"
return Image(
url=item["originalContextUrl"],
title=item.get("titleNoFormatting", ""),
content=item.get("contentNoFormatting", ""),
img_src=item["unescapedUrl"],
thumbnail_src=item["tbUrl"],
resolution=resolution,
img_format=item["fileFormat"].split("/")[-1],
)

1
searx/engines/google_images.py
View File

@@ -43,7 +43,6 @@ max_page = 50
"""
time_range_support = True
language_support = True
safesearch = True
filter_mapping = {0: 'images', 1: 'active', 2: 'active'}

1
searx/engines/google_news.py
View File

@@ -66,7 +66,6 @@ about = {
categories = ["news"]
paging = False
time_range_support = False
language_support = True
# Google-News results are always *SafeSearch*. Option 'safesearch' is set to
# False here.

127
searx/engines/heexy.py
View File

@@ -1,127 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""Heexy_ is a minimalist search engine that focuses on privacy.
Although it also supports news and videos, these are not implemented here
because they usually return no result to very few irrelevant ones.
It seems to use Bing internally, as the image thumbnails are loaded from Bing.
.. _Heexy: https://docs.heexy.org/introduction
"""
from urllib.parse import urlencode
import typing as t
from searx.enginelib import EngineCache
from searx.network import get
from searx.exceptions import SearxEngineAPIException, SearxEngineAccessDeniedException
from searx.result_types import EngineResults
from searx.utils import gen_useragent
if t.TYPE_CHECKING:
from searx.extended_types import SXNG_Response
from searx.search.processors import OnlineParams
about = {
"website": "https://heexy.org",
"wikidata_id": None,
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
paging = True
safesearch = True
categories = ["general"]
heexy_categ = "web"
"""Category to search in. Can be either "web" or "image"."""
base_url = "https://heexy.org"
api_url = "https://seapi.heexy.org"
safe_search_map = {0: "off", 1: "on", 2: "on"}
CACHE: EngineCache
"""Cache for storing the ``X-Data-Cacheft`` token (acts like an API key)."""
def setup(engine_settings: dict[str, t.Any]) -> bool:
global CACHE # pylint: disable=global-statement
if heexy_categ not in ("web", "image"):
raise ValueError("invalid search category: %s" % heexy_categ)
CACHE = EngineCache(engine_settings["name"])
return True
def _get_api_token(query: str) -> str:
"""The API token is independent of the search query. We just need any query
to obtain it initially, and don't hardcode it here to decrease chances of
getting blocked. The token must be passed as ``X-Data-Cacheft`` header."""
cached_token: str = CACHE.get("token")
if cached_token:
return cached_token
resp = get(
f"{base_url}/search?q={query}", headers={"User-Agent": gen_useragent(), "Accept-Language": "en-US,en:q=0.9"}
)
if not resp.ok:
raise SearxEngineAPIException("failed to obtain request token: invalid response code")
token = resp.cookies["cacheft"]
if not token:
raise SearxEngineAPIException("failed to obtain request token: no token found")
CACHE.set("token", token, expire=3 * 60)
return token
def request(query: str, params: "OnlineParams") -> None:
args = {
"q": query,
"page": params["pageno"],
"safe": safe_search_map[params["safesearch"]],
}
if params["searxng_locale"] != "all":
args["lang"] = params["searxng_locale"].split("-")[0]
params["url"] = f"{api_url}/search/{heexy_categ}?{urlencode(args)}"
params["headers"]["Origin"] = api_url
params["cookies"]["cacheft"] = _get_api_token(query)
def response(resp: "SXNG_Response"):
res = EngineResults()
json_resp = resp.json()
if not json_resp["success"]:
raise SearxEngineAccessDeniedException()
result: dict[str, str]
for result in json_resp["results"]:
if heexy_categ == "web":
res.add(
res.types.MainResult(
url=result["url"],
title=result["title"],
content=result["description"],
)
)
elif heexy_categ == "image":
res.add(
res.types.Image(
title=result["description"],
url=result["url"],
thumbnail_src=result["image"],
img_src=result["rawImage"],
)
)
return res

2
searx/engines/huggingface.py
View File

@@ -91,7 +91,7 @@ def response(resp) -> EngineResults:
published_date = None
try:
published_date = datetime.fromisoformat(entry["createdAt"].rstrip("Z"))
published_date = datetime.strptime(entry["createdAt"], "%Y-%m-%dT%H:%M:%S.%fZ")
except (ValueError, TypeError):
pass

2
searx/engines/il_post.py
View File

@@ -34,8 +34,8 @@ about = {
"use_official_api": True,
"require_api_key": False,
"results": "JSON",
"language": "it",
}
language = "it"
def request(query, params):

2
searx/engines/ina.py
View File

@@ -16,8 +16,8 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": 'HTML',
"language": 'fr',
}
language = "fr"
# engine dependent config
categories = ['videos']

4
searx/engines/iqiyi.py
View File

@@ -14,9 +14,9 @@ about = {
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
"language": "zh",
}
language = "zh"
paging = True
time_range_support = True
categories = ["videos"]
@@ -43,7 +43,7 @@ def _result(video: dict[str, typing.Any], album_info: dict[str, typing.Any]):
release_time = album_info.get("releaseTime", {}).get("value")
if release_time:
try:
published_date = datetime.fromisoformat(release_time)
published_date = datetime.strptime(release_time, "%Y-%m-%d")
except (ValueError, TypeError):
pass

88
searx/engines/iseek.py
View File

@@ -1,88 +0,0 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
"""iseek_ is a search engine by the AI company Vantage Labs LLC,
that focuses on medical and educational applicances.
Although it's an AI company, it doesn't include any AI stuff in its results.
.. _iseek : https://www.iseek.ai/
"""
import base64
from hashlib import sha256
import typing as t
from urllib.parse import urlencode
from searx.result_types import EngineResults
if t.TYPE_CHECKING:
from searx.search.processors import OnlineParams
from searx.extended_types import SXNG_Response
about = {
"website": 'https://www.iseek.com',
"wikidata_id": None,
"official_api_documentation": None,
"use_official_api": False,
"require_api_key": False,
"results": "JSON",
}
categories = ["general"]
paging = True
base_url = "https://api.iseek.com"
page_size = 10
def _get_new_token(query: str, pageno: int) -> str:
"""Create a new ``qToken``. This reduced the time for fetching subsequent pages
from 4 seconds to 200ms when testing."""
# The website uses a random value as qToken for the first page. For our use case,
# it's easier if the qToken can be deterministically re-calculated based on the search query,
# so that we can the same result when calling _get_new_token for the second, third, ... page
#
# var qToken = Math.ceil(Math.random() * parseInt("ZZZZ", 36)).toString(36);
# while (qToken.length < 4) qToken = '0' + qToken;
# qToken = qToken + "_" + pageno
query_hash = sha256(query.encode()).digest()
hash_start = base64.b64encode(query_hash).decode()[0:4]
return f"{hash_start}_{pageno}"
def request(query: str, params: "OnlineParams"):
offset = (params["pageno"] - 1) * page_size
# always seems to find 20 results max
if offset >= 20:
params["url"] = None
return
args = {
"q": query,
"key": "core-web",
"num": str(page_size),
"off": offset,
"rSort": "__metasearch_score_d:desc",
# it supports many more fields, but none of them are really relevant
"names": "title_t,content_txt,url_s",
"qNames": "title_t",
"qToken": _get_new_token(query, params["pageno"]),
}
params["url"] = f"{base_url}/search?{urlencode(args)}"
def response(resp: "SXNG_Response") -> EngineResults:
res = EngineResults()
for group in resp.json()["data"]:
group: dict[str, t.Any]
for result in group["doclist"]["docs"]:
result: dict[str, str]
res.add(
res.types.MainResult(
url=result["url_s"],
title=result["title_t"],
content="".join(result["content_txt"]),
)
)
return res

6
searx/engines/jisho.py
View File

@@ -13,8 +13,8 @@ about = {
"use_official_api": True,
"require_api_key": False,
"results": 'JSON',
"language": 'ja',
}
language = "ja"
categories = ['dictionaries']
paging = False
@@ -110,8 +110,8 @@ def get_infobox(alt_forms, result_url, definitions):
# definitions
infobox_content.append(
'''
<small><a href="https://www.edrdg.org/wiki/index.php/JMdict-EDICT_Dictionary_Project">JMdict</a>
and <a href="https://www.edrdg.org/enamdict/enamdict_doc.html">JMnedict</a>
<small><a href="https://www.edrdg.org/wiki/index.php/JMdict-EDICT_Dictionary_Project">JMdict</a>
and <a href="https://www.edrdg.org/enamdict/enamdict_doc.html">JMnedict</a>
by <a href="https://www.edrdg.org/edrdg/licence.html">EDRDG</a>, CC BY-SA 3.0.</small>
<ul>
'''

Some files were not shown because too many files have changed in this diff Show More