Compare commits

...

4 Commits

Author SHA1 Message Date
Andy Lee
55f0973524 fix: leann mcp search cwd & interactive issues 2025-08-19 02:19:06 -07:00
GitHub Actions
738f1dbab8 chore: release v0.3.1 2025-08-19 05:56:45 +00:00
yichuan520030910320
37d990d51c [feature] fix cli 2025-08-18 22:55:43 -07:00
Andy Lee
a6f07a54f1 fix: Use uv venv for Arch Linux CI wheel installation (#69)
- Use astral-sh/setup-uv@v4 action for consistency with other jobs
- Create virtual environment with uv venv to bypass PEP 668 restrictions
- Install wheels using uv pip install for faster dependency resolution
- Maintain tool consistency across the entire CI pipeline

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-authored-by: Claude <noreply@anthropic.com>
2025-08-16 21:32:19 -07:00
9 changed files with 3871 additions and 3825 deletions

View File

@@ -87,7 +87,7 @@ jobs:
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v4
- uses: actions/checkout@v5
with:
ref: ${{ inputs.ref }}
submodules: recursive
@@ -98,7 +98,7 @@ jobs:
python-version: ${{ matrix.python }}
- name: Install uv
uses: astral-sh/setup-uv@v4
uses: astral-sh/setup-uv@v6
- name: Install system dependencies (Ubuntu)
if: runner.os == 'Linux'
@@ -322,24 +322,29 @@ jobs:
pacman -S --noconfirm python python-pip gcc git zlib openssl
- name: Download ALL wheel artifacts from this run
uses: actions/download-artifact@v4
uses: actions/download-artifact@v5
with:
# Don't specify name, download all artifacts
path: ./wheels
- name: Install wheels (pip automatically picks matching tags from wheels directory)
- name: Install uv
uses: astral-sh/setup-uv@v6
- name: Create virtual environment and install wheels
run: |
python -m pip install --upgrade pip
pip install --find-links wheels leann-core
pip install --find-links wheels leann-backend-hnsw
pip install --find-links wheels leann-backend-diskann
pip install --find-links wheels leann
uv venv
source .venv/bin/activate || source .venv/Scripts/activate
uv pip install --find-links wheels leann-core
uv pip install --find-links wheels leann-backend-hnsw
uv pip install --find-links wheels leann-backend-diskann
uv pip install --find-links wheels leann
- name: Import & tiny runtime check
env:
OMP_NUM_THREADS: 1
MKL_NUM_THREADS: 1
run: |
source .venv/bin/activate || source .venv/Scripts/activate
python - <<'PY'
import leann
import leann_backend_hnsw as h

View File

@@ -183,6 +183,9 @@ class Benchmark:
start_time = time.time()
with torch.no_grad():
self.model(input_ids=input_ids, attention_mask=attention_mask)
# mps sync
if torch.backends.mps.is_available():
torch.mps.synchronize()
end_time = time.time()
return end_time - start_time

View File

@@ -4,8 +4,8 @@ build-backend = "scikit_build_core.build"
[project]
name = "leann-backend-diskann"
version = "0.3.0"
dependencies = ["leann-core==0.3.0", "numpy", "protobuf>=3.19.0"]
version = "0.3.1"
dependencies = ["leann-core==0.3.1", "numpy", "protobuf>=3.19.0"]
[tool.scikit-build]
# Key: simplified CMake path

View File

@@ -6,10 +6,10 @@ build-backend = "scikit_build_core.build"
[project]
name = "leann-backend-hnsw"
version = "0.3.0"
version = "0.3.1"
description = "Custom-built HNSW (Faiss) backend for the Leann toolkit."
dependencies = [
"leann-core==0.3.0",
"leann-core==0.3.1",
"numpy",
"pyzmq>=23.0.0",
"msgpack>=1.0.0",

View File

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
[project]
name = "leann-core"
version = "0.3.0"
version = "0.3.1"
description = "Core API and plugin system for LEANN"
readme = "README.md"
requires-python = ">=3.9"

View File

@@ -206,6 +206,11 @@ Examples:
default="global",
help="Pruning strategy (default: global)",
)
search_parser.add_argument(
"--non-interactive",
action="store_true",
help="Non-interactive mode: automatically select index without prompting",
)
# Ask command
ask_parser = subparsers.add_parser("ask", help="Ask questions")
@@ -405,13 +410,9 @@ Examples:
print("💡 Get started:")
print(" leann build my-docs --docs ./documents")
else:
projects_count = len(
[
p
for p in valid_projects
if (p / ".leann" / "indexes").exists()
and list((p / ".leann" / "indexes").iterdir())
]
# Count only projects that have at least one discoverable index
projects_count = sum(
1 for p in valid_projects if len(self._discover_indexes_in_project(p)) > 0
)
print(f"📊 Total: {total_indexes} indexes across {projects_count} projects")
@@ -461,26 +462,35 @@ Examples:
)
# 2. Apps format: *.leann.meta.json files anywhere in the project
cli_indexes_dir = project_path / ".leann" / "indexes"
for meta_file in project_path.rglob("*.leann.meta.json"):
if meta_file.is_file():
# Extract index name from filename (remove .leann.meta.json extension)
index_name = meta_file.name.replace(".leann.meta.json", "")
# Skip CLI-built indexes (which store meta under .leann/indexes/<name>/)
try:
if cli_indexes_dir.exists() and cli_indexes_dir in meta_file.parents:
continue
except Exception:
pass
# Use the parent directory name as the app index display name
display_name = meta_file.parent.name
# Extract file base used to store files
file_base = meta_file.name.replace(".leann.meta.json", "")
# Apps indexes are considered complete if the .leann.meta.json file exists
status = ""
# Calculate total size of all related files
# Calculate total size of all related files (use file base)
size_mb = 0
try:
index_dir = meta_file.parent
for related_file in index_dir.glob(f"{index_name}.leann*"):
for related_file in index_dir.glob(f"{file_base}.leann*"):
size_mb += related_file.stat().st_size / (1024 * 1024)
except (OSError, PermissionError):
pass
indexes.append(
{
"name": index_name,
"name": display_name,
"type": "app",
"status": status,
"size_mb": size_mb,
@@ -534,13 +544,79 @@ Examples:
if not project_path.exists():
continue
# 1) CLI-format index under .leann/indexes/<name>
index_dir = project_path / ".leann" / "indexes" / index_name
if index_dir.exists():
is_current = project_path == current_path
matches.append(
{"project_path": project_path, "index_dir": index_dir, "is_current": is_current}
{
"project_path": project_path,
"index_dir": index_dir,
"is_current": is_current,
"kind": "cli",
}
)
# 2) App-format indexes
# We support two ways of addressing apps:
# a) by the file base (e.g., `pdf_documents`)
# b) by the parent directory name (e.g., `new_txt`)
seen_app_meta = set()
# 2a) by file base
for meta_file in project_path.rglob(f"{index_name}.leann.meta.json"):
if meta_file.is_file():
# Skip CLI-built indexes' meta under .leann/indexes
try:
cli_indexes_dir = project_path / ".leann" / "indexes"
if cli_indexes_dir.exists() and cli_indexes_dir in meta_file.parents:
continue
except Exception:
pass
is_current = project_path == current_path
key = (str(project_path), str(meta_file))
if key in seen_app_meta:
continue
seen_app_meta.add(key)
matches.append(
{
"project_path": project_path,
"files_dir": meta_file.parent,
"meta_file": meta_file,
"is_current": is_current,
"kind": "app",
"display_name": meta_file.parent.name,
"file_base": meta_file.name.replace(".leann.meta.json", ""),
}
)
# 2b) by parent directory name
for meta_file in project_path.rglob("*.leann.meta.json"):
if meta_file.is_file() and meta_file.parent.name == index_name:
# Skip CLI-built indexes' meta under .leann/indexes
try:
cli_indexes_dir = project_path / ".leann" / "indexes"
if cli_indexes_dir.exists() and cli_indexes_dir in meta_file.parents:
continue
except Exception:
pass
is_current = project_path == current_path
key = (str(project_path), str(meta_file))
if key in seen_app_meta:
continue
seen_app_meta.add(key)
matches.append(
{
"project_path": project_path,
"files_dir": meta_file.parent,
"meta_file": meta_file,
"is_current": is_current,
"kind": "app",
"display_name": meta_file.parent.name,
"file_base": meta_file.name.replace(".leann.meta.json", ""),
}
)
# Sort: current project first, then by project name
matches.sort(key=lambda x: (not x["is_current"], x["project_path"].name))
return matches
@@ -548,8 +624,8 @@ Examples:
def _remove_single_match(self, match, index_name: str, force: bool):
"""Handle removal when only one match is found"""
project_path = match["project_path"]
index_dir = match["index_dir"]
is_current = match["is_current"]
kind = match.get("kind", "cli")
if is_current:
location_info = "current project"
@@ -560,7 +636,10 @@ Examples:
print(f"✅ Found 1 index named '{index_name}':")
print(f" {emoji} Location: {location_info}")
print(f" 📍 Path: {project_path}")
if kind == "cli":
print(f" 📍 Path: {project_path / '.leann' / 'indexes' / index_name}")
else:
print(f" 📍 Meta: {match['meta_file']}")
if not force:
if not is_current:
@@ -572,9 +651,22 @@ Examples:
print(" ❌ Removal cancelled.")
return False
return self._delete_index_directory(
index_dir, index_name, project_path if not is_current else None
)
if kind == "cli":
return self._delete_index_directory(
match["index_dir"],
index_name,
project_path if not is_current else None,
is_app=False,
)
else:
return self._delete_index_directory(
match["files_dir"],
match.get("display_name", index_name),
project_path if not is_current else None,
is_app=True,
meta_file=match.get("meta_file"),
app_file_base=match.get("file_base"),
)
def _remove_from_multiple_matches(self, matches, index_name: str, force: bool):
"""Handle removal when multiple matches are found"""
@@ -585,19 +677,34 @@ Examples:
for i, match in enumerate(matches, 1):
project_path = match["project_path"]
is_current = match["is_current"]
kind = match.get("kind", "cli")
if is_current:
print(f" {i}. 🏠 Current project")
print(f" 📍 {project_path}")
print(f" {i}. 🏠 Current project ({'CLI' if kind == 'cli' else 'APP'})")
else:
print(f" {i}. 📂 {project_path.name}")
print(f" 📍 {project_path}")
print(f" {i}. 📂 {project_path.name} ({'CLI' if kind == 'cli' else 'APP'})")
# Show path details
if kind == "cli":
print(f" 📍 {project_path / '.leann' / 'indexes' / index_name}")
else:
print(f" 📍 {match['meta_file']}")
# Show size info
try:
size_mb = sum(
f.stat().st_size for f in match["index_dir"].iterdir() if f.is_file()
) / (1024 * 1024)
if kind == "cli":
size_mb = sum(
f.stat().st_size for f in match["index_dir"].iterdir() if f.is_file()
) / (1024 * 1024)
else:
file_base = match.get("file_base")
size_mb = 0.0
if file_base:
size_mb = sum(
f.stat().st_size
for f in match["files_dir"].glob(f"{file_base}.leann*")
if f.is_file()
) / (1024 * 1024)
print(f" 📦 Size: {size_mb:.1f} MB")
except (OSError, PermissionError):
pass
@@ -621,8 +728,8 @@ Examples:
if 0 <= choice_idx < len(matches):
selected_match = matches[choice_idx]
project_path = selected_match["project_path"]
index_dir = selected_match["index_dir"]
is_current = selected_match["is_current"]
kind = selected_match.get("kind", "cli")
location = "current project" if is_current else f"'{project_path.name}' project"
print(f" 🎯 Selected: Remove from {location}")
@@ -635,9 +742,22 @@ Examples:
print(" ❌ Confirmation failed. Removal cancelled.")
return False
return self._delete_index_directory(
index_dir, index_name, project_path if not is_current else None
)
if kind == "cli":
return self._delete_index_directory(
selected_match["index_dir"],
index_name,
project_path if not is_current else None,
is_app=False,
)
else:
return self._delete_index_directory(
selected_match["files_dir"],
selected_match.get("display_name", index_name),
project_path if not is_current else None,
is_app=True,
meta_file=selected_match.get("meta_file"),
app_file_base=selected_match.get("file_base"),
)
else:
print(" ❌ Invalid choice. Removal cancelled.")
return False
@@ -647,21 +767,65 @@ Examples:
return False
def _delete_index_directory(
self, index_dir: Path, index_name: str, project_path: Optional[Path] = None
self,
index_dir: Path,
index_display_name: str,
project_path: Optional[Path] = None,
is_app: bool = False,
meta_file: Optional[Path] = None,
app_file_base: Optional[str] = None,
):
"""Actually delete the index directory"""
"""Delete a CLI index directory or APP index files safely."""
try:
import shutil
if is_app:
removed = 0
errors = 0
# Delete only files that belong to this app index (based on file base)
pattern_base = app_file_base or ""
for f in index_dir.glob(f"{pattern_base}.leann*"):
try:
f.unlink()
removed += 1
except Exception:
errors += 1
# Best-effort: also remove the meta file if specified and still exists
if meta_file and meta_file.exists():
try:
meta_file.unlink()
removed += 1
except Exception:
errors += 1
shutil.rmtree(index_dir)
if project_path:
print(f"Index '{index_name}' removed from {project_path.name}")
if removed > 0 and errors == 0:
if project_path:
print(
f"App index '{index_display_name}' removed from {project_path.name}"
)
else:
print(f"✅ App index '{index_display_name}' removed successfully")
return True
elif removed > 0 and errors > 0:
print(
f"⚠️ App index '{index_display_name}' partially removed (some files couldn't be deleted)"
)
return True
else:
print(
f"❌ No files found to remove for app index '{index_display_name}' in {index_dir}"
)
return False
else:
print(f"✅ Index '{index_name}' removed successfully")
return True
import shutil
shutil.rmtree(index_dir)
if project_path:
print(f"✅ Index '{index_display_name}' removed from {project_path.name}")
else:
print(f"✅ Index '{index_display_name}' removed successfully")
return True
except Exception as e:
print(f"❌ Error removing index '{index_name}': {e}")
print(f"❌ Error removing index '{index_display_name}': {e}")
return False
def load_documents(
@@ -1085,13 +1249,101 @@ Examples:
async def search_documents(self, args):
index_name = args.index_name
query = args.query
index_path = self.get_index_path(index_name)
if not self.index_exists(index_name):
print(
f"Index '{index_name}' not found. Use 'leann build {index_name} --docs <dir> [<dir2> ...]' to create it."
)
return
# First try to find the index in current project
index_path = self.get_index_path(index_name)
if self.index_exists(index_name):
# Found in current project, use it
pass
else:
# Search across all registered projects (like list_indexes does)
all_matches = self._find_all_matching_indexes(index_name)
if not all_matches:
print(
f"Index '{index_name}' not found. Use 'leann build {index_name} --docs <dir> [<dir2> ...]' to create it."
)
return
elif len(all_matches) == 1:
# Found exactly one match, use it
match = all_matches[0]
if match["kind"] == "cli":
index_path = str(match["index_dir"] / "documents.leann")
else:
# App format: use the meta file to construct the path
meta_file = match["meta_file"]
file_base = match["file_base"]
index_path = str(meta_file.parent / f"{file_base}.leann")
project_info = (
"current project"
if match["is_current"]
else f"project '{match['project_path'].name}'"
)
print(f"Using index '{index_name}' from {project_info}")
else:
# Multiple matches found
if args.non_interactive:
# Non-interactive mode: automatically select the best match
# Priority: current project first, then first available
current_matches = [m for m in all_matches if m["is_current"]]
if current_matches:
match = current_matches[0]
location_desc = "current project"
else:
match = all_matches[0]
location_desc = f"project '{match['project_path'].name}'"
if match["kind"] == "cli":
index_path = str(match["index_dir"] / "documents.leann")
else:
meta_file = match["meta_file"]
file_base = match["file_base"]
index_path = str(meta_file.parent / f"{file_base}.leann")
print(
f"Found {len(all_matches)} indexes named '{index_name}', using index from {location_desc}"
)
else:
# Interactive mode: ask user to choose
print(f"Found {len(all_matches)} indexes named '{index_name}':")
for i, match in enumerate(all_matches, 1):
project_path = match["project_path"]
is_current = match["is_current"]
kind = match.get("kind", "cli")
if is_current:
print(
f" {i}. 🏠 Current project ({'CLI' if kind == 'cli' else 'APP'})"
)
else:
print(
f" {i}. 📂 {project_path.name} ({'CLI' if kind == 'cli' else 'APP'})"
)
try:
choice = input(f"Which index to search? (1-{len(all_matches)}): ").strip()
choice_idx = int(choice) - 1
if 0 <= choice_idx < len(all_matches):
match = all_matches[choice_idx]
if match["kind"] == "cli":
index_path = str(match["index_dir"] / "documents.leann")
else:
meta_file = match["meta_file"]
file_base = match["file_base"]
index_path = str(meta_file.parent / f"{file_base}.leann")
project_info = (
"current project"
if match["is_current"]
else f"project '{match['project_path'].name}'"
)
print(f"Using index '{index_name}' from {project_info}")
else:
print("Invalid choice. Aborting search.")
return
except (ValueError, KeyboardInterrupt):
print("Invalid input. Aborting search.")
return
searcher = LeannSearcher(index_path=index_path)
results = searcher.search(

View File

@@ -94,7 +94,7 @@ def handle_request(request):
},
}
# Build simplified command
# Build simplified command with non-interactive flag for MCP compatibility
cmd = [
"leann",
"search",
@@ -102,6 +102,7 @@ def handle_request(request):
args["query"],
f"--top-k={args.get('top_k', 5)}",
f"--complexity={args.get('complexity', 32)}",
"--non-interactive",
]
result = subprocess.run(cmd, capture_output=True, text=True)

View File

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
[project]
name = "leann"
version = "0.3.0"
version = "0.3.1"
description = "LEANN - The smallest vector index in the world. RAG Everything with LEANN!"
readme = "README.md"
requires-python = ">=3.9"

7313
uv.lock generated
View File

File diff suppressed because it is too large Load Diff