Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
63 changes: 63 additions & 0 deletions .github/workflows/lint.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
# Nothing ran these checks automatically before, so a defect they were
# written to catch could still be merged. They are cheap and need no
# dependencies beyond the standard library.
name: lint

on:
pull_request:
push:
branches: [master]

permissions:
contents: read

jobs:
tools:
name: tools
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: actions/setup-python@v5
with:
python-version: '3.12'

- name: Unit tests
run: python -m unittest discover -s tools -p 'test_*.py' -v

- name: Every eol.rst matches versions.json
run: python tools/update_eol.py --check

headings:
name: section titles
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: actions/setup-python@v5
with:
python-version: '3.12'

# reStructuredText requires the overline and underline of a section
# title to be the same length. Where they differ docutils refuses to
# build the section and the next one silently becomes the page title,
# without failing the build. Scoped to the versions versions.json
# calls current, so it follows a release instead of needing an edit,
# and so the archived versions -- which still carry mismatches and are
# consolidated by canonical anyway -- do not block unrelated work.
- name: Section title rules in the current documentation versions
run: |
set -eu
versions=$(python -c \
"import json; d = json.load(open('versions.json')); \
print(d['release'], d['development'])")
dirs=""
for v in $versions; do
for d in */"$v"; do
[ -d "$d" ] && dirs="$dirs $d"
done
done
if [ -z "$dirs" ]; then
echo "versions.json names no directory that exists" >&2
exit 1
fi
echo "linting:$dirs"
python tools/check_headings.py $dirs
2 changes: 1 addition & 1 deletion zh-cn/15.8/config/llm-gemini.rst
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ Google Gemini配置(AI 搜索 / RAG)
概述
====

本页说明如何配置 ``fess-llm-gemini`` 插件,以便 |Fess| 使用 Google Gemini 实现其**AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题。|Fess| 会调用 Google AI API(Generative Language API),使用 Gemini 模型对已爬取的文档执行 RAG。
本页说明如何配置 ``fess-llm-gemini`` 插件,以便 |Fess| 使用 Google Gemini 实现其\ **AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题。|Fess| 会调用 Google AI API(Generative Language API),使用 Gemini 模型对已爬取的文档执行 RAG。

Google Gemini是Google公司提供的最先进的大型语言模型(LLM)。
|Fess| 可以使用Google AI API(Generative Language API)通过Gemini模型实现AI搜索模式功能。
Expand Down
2 changes: 1 addition & 1 deletion zh-cn/15.8/config/llm-ollama.rst
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ Ollama配置(本地 LLM / RAG)
概述
====

本页说明如何配置 ``fess-llm-ollama`` 插件,以便 |Fess| 使用本地部署的 Ollama 模型实现其**AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题,且无需将数据发送至外部 API。|Fess| 会调用本地 Ollama API,对已爬取的文档执行 RAG。
本页说明如何配置 ``fess-llm-ollama`` 插件,以便 |Fess| 使用本地部署的 Ollama 模型实现其\ **AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题,且无需将数据发送至外部 API。|Fess| 会调用本地 Ollama API,对已爬取的文档执行 RAG。

Ollama是用于在本地环境运行大型语言模型(LLM)的开源平台。
|Fess| 的Ollama集成功能以插件 ``fess-llm-ollama`` 的形式提供,适合在私有环境中使用。
Expand Down
2 changes: 1 addition & 1 deletion zh-cn/15.8/config/llm-openai.rst
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ OpenAI配置(AI 搜索 / RAG)
概述
====

本页说明如何配置 ``fess-llm-openai`` 插件,以便 |Fess| 使用 OpenAI 实现其**AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题。|Fess| 会调用 OpenAI API,使用 GPT 模型对已爬取的文档执行 RAG。
本页说明如何配置 ``fess-llm-openai`` 插件,以便 |Fess| 使用 OpenAI 实现其\ **AI搜索模式(RAG:Retrieval-Augmented Generation)** —— 基于引用来源,直接从您的企业搜索索引回答自然语言问题。|Fess| 会调用 OpenAI API,使用 GPT 模型对已爬取的文档执行 RAG。

OpenAI是提供GPT-4等高性能大型语言模型(LLM)的云服务。
|Fess| 可以使用OpenAI API实现AI搜索模式功能。
Expand Down
Loading