release: bump version to 0.45.0

This commit is contained in:
rayd1o
2026-04-29 23:43:54 +08:00
parent 9dafbf4f6e
commit ba54545ac7
32 changed files with 602 additions and 29 deletions

View File

@@ -55,6 +55,30 @@ Parameters:
| `--allow-lan` | Enable LAN access |
| `--verbose` | Show more command output during execution |
### AI Provider Environment and Builds
AI Provider runtime configuration can live in `aiprovider/.env` or in matching variables in `~/.zshrc`. `planet.sh` reads simple `export AI_...=...` / `AI_...=...` lines and passes them to the container at startup.
Changing model, API key, or base URL does not rebuild the image. Restart only AI Provider to pick up runtime configuration changes:
```bash
./planet.sh restart -a
```
For complex shell expansion in `~/.zshrc`, opt in explicitly:
```bash
PLANET_LOAD_ZSHRC_ENV=source ./planet.sh start -a
```
To ignore `~/.zshrc` during troubleshooting:
```bash
PLANET_LOAD_ZSHRC_ENV=0 ./planet.sh start -a
```
The AI Provider Docker build context is intentionally limited to the files required by the service, and `uv sync` uses a BuildKit cache mount so dependency downloads are reused after the first build.
### Stop
```bash

View File

@@ -24,6 +24,12 @@ This script checks and syncs common dependencies, and generates if missing:
- `aiprovider/.env`
- `frontend/.env.local`
Personal AI Provider configuration can also live in `~/.zshrc`. `planet.sh` reads simple `export AI_...=...` / `AI_...=...` lines and passes them to the AI Provider container. After changing model, key, or base URL, restart only AI Provider:
```bash
./planet.sh restart -a
```
## 1. Start Services
From the repository root: