- Code\rebar_pipeline, review_pipeline, pier-program을 정션에서 실제 폴더로 전환 - Code\rhino_mcp 및 CLAUDE_rhino배경.md 추가, .mcp.json은 PYTHONPATH로 새 위치 사용 - pier-program 내부 .git 제거(이력은 backup\pier-program_20261028.bundle에 보관) - worktree 가이드 반영(CLAUDE.local.md, HANDOFF.md) - D:\Rhino-MCP 경로 참조를 새 위치로 치환 Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
1.1 KiB
1.1 KiB
DWG deployment runtime
The Linux/amd64 image includes LibreDWG 0.11.3876, the same decoder version/package as the local host. Geometry/rebar generation code is unchanged. The build downloads a fixed conda-forge artifact and verifies its SHA-256 before installing only dwg2dxf, its shared library, GPL license and source metadata. The runtime does not depend on a host PC path or install Conda/Python bindings.
- License: GPL-3.0-or-later, installed at
/opt/libredwg/licenses/COPYING. - Corresponding upstream source: https://github.com/LibreDWG/libredwg/releases/download/0.11.3876/libredwg-0.11.3876.tar.gz
- Source SHA-256:
ba65e77adb3099bec9f8f6c2ec2c6d97457a148be0a485d7fd3c00679f332e52. - Package recipe and source references:
/opt/libredwg/licenses/package-source.yaml.
Run the existing private deploy-server.bat on the deployment server again.
It fast-forwards the clean clone and rebuilds the image. Existing named model
and AI-auth volumes remain in place. This is not automatic deployment on push.
DWG conversion success is not full drawing/model approval.