benchbox 0.1.0__py3-none-any.whl → 0.1.1__py3-none-any.whl

benchbox/release/workflow.py

@@ -75,13 +75,27 @@ TESTS_DIR_EXCLUDES: Sequence[str] = (
 # The _sources directory contains build artifacts and data files that should not
 # be included in releases. Only query templates needed for qgen/dsqgen are included.
 _SOURCES_INCLUDE_PATHS: Sequence[str] = (
+    # Top-level documentation
+    "README.md",
+    # Compilation infrastructure (for users who need to build from source)
+    "compilation",
     # TPC-H query templates (required for qgen)
     "tpc-h/dbgen/queries",  # 22 SQL template files
     "tpc-h/dbgen/variants",  # 5 variant SQL files
     "tpc-h/dbgen/dists.dss",  # Distribution specifications for qgen
+    # TPC-H answer files (for validation tests)
+    "tpc-h/dbgen/answers",
+    # TPC-H patch documentation
+    "tpc-h/PATCHES.md",
+    "tpc-h/stdout-support.patch",
     # TPC-DS query templates (required for dsqgen)
     "tpc-ds/query_templates",  # Template files (*.tpl, *.lst)
     "tpc-ds/query_variants",  # Variant template files (*.tpl)
+    # TPC-DS answer sets (for validation tests)
+    "tpc-ds/answer_sets",
+    # TPC-DS patch documentation
+    "tpc-ds/PATCHES.md",
+    "tpc-ds/stdout-support.patch",
 )
 
 # Global patterns to exclude from all directories
@@ -109,6 +123,9 @@ CLEANUP_PATTERNS: Sequence[str] = (
     ".nox",  # Nox environments
     "dist",  # Build output (created fresh by build step)
     "build",  # Build output
+    "_sources",  # Always rebuild from whitelist to prevent stale accumulation
+    "_project",  # Development-only working files
+    "benchmark_runs",  # Execution outputs (should never be in releases)
 )
 
 SANITIZED_README = Path("release/README.public.md")

benchbox/utils/VERSION_MANAGEMENT.md

@@ -4,7 +4,7 @@
 
 This document outlines the version management strategy for BenchBox, ensuring consistency across all version references and providing clear guidelines for version updates.
 
-Current release: `v0.1.0`.
+Current release: `v0.1.1`.
 
 ## Version Sources
 

benchbox/utils/cloud_storage.py

@@ -16,6 +16,8 @@ from pathlib import Path
 from typing import Any, List, Protocol, Union, runtime_checkable
 from urllib.parse import urlparse
 
+from benchbox.utils.dependencies import get_install_command, get_package_install_message
+
 try:
     from cloudpathlib import CloudPath
     from cloudpathlib.exceptions import (
@@ -270,7 +272,7 @@ class DatabricksVolumeAdapter:
             from databricks.sdk import WorkspaceClient  # type: ignore
         except Exception as e:  # pragma: no cover - optional dependency
             raise ImportError(
-                "databricks-sdk required for DatabricksVolumeAdapter. Install with: uv add databricks-sdk"
+                get_package_install_message("databricks-sdk", "databricks-sdk required for DatabricksVolumeAdapter.")
             ) from e
 
         if workspace_client is not None:
@@ -436,9 +438,7 @@ def create_path_handler(path: Union[str, Path]) -> Union[Path, CloudPath, Databr
         return Path(path)
 
     if CloudPath is None:
-        raise ImportError(
-            "cloudpathlib is required for cloud storage paths. Install with: uv add benchbox --extra cloudstorage"
-        )
+        raise ImportError(f"cloudpathlib is required for cloud storage paths. {get_install_command('cloudstorage')}")
 
     try:
         return CloudPath(str(path))  # type: ignore
@@ -461,7 +461,9 @@ def get_remote_fs_adapter(remote_path: str) -> RemoteFileSystemAdapter:
             return DatabricksVolumeAdapter(ws)
         except ImportError as e:
             raise ImportError(
-                "databricks-sdk required for Databricks UC Volume operations. Install with: uv add databricks-sdk"
+                get_package_install_message(
+                    "databricks-sdk", "databricks-sdk required for Databricks UC Volume operations."
+                )
             ) from e
         except Exception as e:
             raise RuntimeError(

benchbox/utils/compression.py

@@ -14,6 +14,8 @@ from abc import ABC, abstractmethod
 from pathlib import Path
 from typing import BinaryIO, Optional, TextIO, Union, cast
 
+from benchbox.utils.file_format import detect_compression as _detect_compression_from_path
+
 try:
     import zstandard as zstd
 
@@ -329,19 +331,17 @@ class CompressionManager:
     def detect_compression(self, path: Path) -> Optional[str]:
         """Detect compression type from file extension.
 
+        Uses centralized detection from benchbox.utils.file_format which
+        recognizes all compression extensions (.zst, .gz, .bz2, .xz, .lz4, .snappy).
+
         Args:
             path: File path to analyze
 
         Returns:
-            Detected compression type or None if uncompressed
+            Detected compression type or "none" if uncompressed
         """
-        suffix = path.suffix.lower()
-        if suffix == ".gz":
-            return "gzip"
-        elif suffix == ".zst":
-            return "zstd"
-        else:
-            return "none"
+        compression_type = _detect_compression_from_path(path)
+        return compression_type if compression_type is not None else "none"
 
     def get_compression_info(self, input_path: Path, output_path: Path) -> dict[str, Union[int, float]]:
         """Get compression information comparing two files.

benchbox/utils/compression_mixin.py

@@ -12,6 +12,7 @@ from pathlib import Path
 from typing import Union
 
 from .compression import CompressionError, CompressionManager
+from .file_format import strip_compression_suffix
 
 
 class CompressionMixin:
@@ -156,7 +157,7 @@ class CompressionMixin:
 
             if str(file_path).endswith(extension):
                 # This is the compressed file, look for original
-                original_path = Path(str(file_path)[: -len(extension)])
+                original_path = strip_compression_suffix(file_path)
                 if not original_path.exists():
                     # Original doesn't exist, skip this file
                     continue

benchbox/utils/data_validation.py

@@ -15,6 +15,8 @@ from datetime import datetime, timezone
 from pathlib import Path
 from typing import Optional, Union
 
+from benchbox.utils.file_format import detect_compression, strip_compression_suffix
+
 logger = logging.getLogger(__name__)
 
 
@@ -279,17 +281,17 @@ class BenchmarkDataValidator:
         return total_rows
 
     def _count_rows_paths(self, files: list[Path]) -> int:
-        """Count rows across a list of file paths (supports .dat, .tbl, .csv, .gz; best-effort for .zst)."""
+        """Count rows across a list of file paths (supports .dat, .tbl, .csv with compression)."""
         total = 0
         for f in files:
-            name = f.name.lower()
             try:
-                if name.endswith(".gz"):
+                compression = detect_compression(f)
+                if compression == "gzip":
                     import gzip
 
                     with gzip.open(f, "rt") as g:
                         total += sum(1 for _ in g)
-                elif name.endswith(".zst"):
+                elif compression == "zstd":
                     try:
                         import zstandard as zstd  # type: ignore
 
@@ -303,7 +305,18 @@ class BenchmarkDataValidator:
                         # If zstandard is not available, skip row counting for this file
                         logger.debug(f"Skipping zstd row count for {f}")
                         continue
+                elif compression == "bzip2":
+                    import bz2
+
+                    with bz2.open(f, "rt") as b:
+                        total += sum(1 for _ in b)
+                elif compression == "xz":
+                    import lzma
+
+                    with lzma.open(f, "rt") as x:
+                        total += sum(1 for _ in x)
                 else:
+                    # Uncompressed or unsupported compression (lz4, snappy)
                     with open(f, "rb") as fh:
                         total += sum(1 for _ in fh)
             except Exception as e:
@@ -337,9 +350,9 @@ class BenchmarkDataValidator:
         for ext in [".dat", ".dat.gz", ".dat.zst"]:
             for pf in data_dir.glob(f"{table_name}_*.{ext.split('.')[-1]}"):
                 name = pf.name
-                # Normalize compression suffix handling
-                core = name[:-3] if name.endswith(".gz") else (name[:-4] if name.endswith(".zst") else name)
-                stem = Path(core).stem  # removes .dat
+                # Use centralized suffix stripping for compression handling
+                stripped = strip_compression_suffix(Path(name))
+                stem = stripped.stem  # removes .dat
                 if stem.startswith(f"{table_name}_"):
                     suffix = stem[len(f"{table_name}_") :]
                     parts = suffix.split("_")
@@ -550,13 +563,9 @@ class BenchmarkDataValidator:
         # Group by table name (stem up to first dot for tbl/dat, handle chunk names)
         for fp in candidates:
             name = fp.name
-            base = name
-            # Remove compression suffix
-            if base.endswith(".gz"):
-                base = base[:-3]
-            elif base.endswith(".zst"):
-                base = base[:-4]
-            stem = Path(base).stem  # removes .tbl/.dat/.csv
+            # Remove compression suffix using centralized utility
+            stripped = strip_compression_suffix(Path(name))
+            stem = stripped.stem  # removes .tbl/.dat/.csv
             # Normalize chunked variants by table prefix
             table = stem
             # For known patterns like table_N_M or table.tbl.N preserve table name before first underscore/dot-digit

benchbox/utils/dependencies.py

@@ -97,6 +97,44 @@ def get_install_command(extra: str) -> str:
         return f'uv pip install "benchbox[{resolved_extra}]"'
 
 
+def get_package_install_message(packages: str, description: str = "") -> str:
+    """Generate a user-friendly install message for raw Python packages.
+
+    Provides both standalone (pip) and project (uv add) installation options.
+
+    Args:
+        packages: Space-separated package names (e.g., 'boto3' or 'azure-storage-blob azure-identity')
+        description: Optional description of what the packages are for
+
+    Returns:
+        Multi-line install message with both options
+    """
+    prefix = f"{description} " if description else ""
+    return (
+        f"{prefix}Install with:\n  pip install {packages}  # standalone\n  uv add {packages}       # inside a project"
+    )
+
+
+def get_extra_install_message(extra_name: str, description: str = "") -> str:
+    """Generate a user-friendly install message for BenchBox extras.
+
+    Provides both standalone (pip install) and project (uv add) installation options.
+
+    Args:
+        extra_name: The BenchBox extra name (e.g., 'cloud', 'mcp', 'databricks')
+        description: Optional description prefix
+
+    Returns:
+        Multi-line install message with both options
+    """
+    prefix = f"{description}\n" if description else ""
+    return (
+        f"{prefix}Install with:\n"
+        f"  pip install 'benchbox[{extra_name}]'  # pip/venv\n"
+        f"  uv add benchbox --extra {extra_name}  # uv project"
+    )
+
+
 class DependencyInfo:
     """Information about a platform dependency group."""
 
@@ -116,6 +154,14 @@ class DependencyInfo:
         self.use_cases = use_cases
         self.platforms = platforms
 
+    def get_install_message(self) -> str:
+        """Get a context-aware install message with both standalone and project options.
+
+        Returns:
+            Multi-line install message showing both installation options.
+        """
+        return get_extra_install_message(self.name)
+
 
 # Structured installation guidance for documentation and CLI matrix output
 class InstallationScenario:
@@ -701,13 +747,7 @@ def get_dependency_error_message(platform: str, missing_packages: list[str]) ->
         f"  Extra: benchbox[{dep_info.name}]",
         f"  Required packages: {', '.join(missing_packages)}",
         "",
-        "Install with (recommended):",
-        f"  {dep_info.install_command}",
-        "",
-        "Alternative (pip-compatible):",
-        f'  uv pip install "benchbox[{dep_info.name}]"',
-        f'  python -m pip install "benchbox[{dep_info.name}]"',
-        f'  pipx install "benchbox[{dep_info.name}]"',
+        dep_info.get_install_message(),
         "",
         "This extra provides:",
         f"  • {dep_info.description}",