ado-git-repo-insights 1.2.1__py3-none-any.whl → 2.7.4__py3-none-any.whl

ado_git_repo_insights/transform/csv_generator.py

@@ -1,132 +1,132 @@
-"""CSV generator for PowerBI-compatible output.
-
-Generates CSVs that are:
-- Schema-compliant (exact columns, exact order - Invariants 1-4)
-- Deterministic (same DB → same bytes - Adjustment 3)
-"""
-
-from __future__ import annotations
-
-import logging
-from pathlib import Path
-from typing import TYPE_CHECKING
-
-import pandas as pd
-
-from ..persistence.models import CSV_SCHEMAS, SORT_KEYS
-
-if TYPE_CHECKING:
-    from ..persistence.database import DatabaseManager
-
-logger = logging.getLogger(__name__)
-
-
-class CSVGenerationError(Exception):
-    """CSV generation failed."""
-
-
-class CSVGenerator:
-    """Generates PowerBI-compatible CSV files from SQLite.
-
-    Invariant 1: CSV schema is a hard contract.
-    Invariant 3: CSV output must be deterministic.
-    """
-
-    def __init__(self, db: DatabaseManager, output_dir: Path) -> None:
-        """Initialize the CSV generator.
-
-        Args:
-            db: Database manager instance.
-            output_dir: Directory for CSV output files.
-        """
-        self.db = db
-        self.output_dir = output_dir
-
-    def generate_all(self) -> dict[str, int]:
-        """Generate all CSV files.
-
-        Returns:
-            Dict mapping table names to row counts.
-
-        Raises:
-            CSVGenerationError: If generation fails.
-        """
-        self.output_dir.mkdir(parents=True, exist_ok=True)
-
-        results: dict[str, int] = {}
-
-        for table_name, columns in CSV_SCHEMAS.items():
-            try:
-                count = self._generate_table(table_name, columns)
-                results[table_name] = count
-                logger.info(f"Generated {table_name}.csv: {count} rows")
-            except Exception as e:
-                raise CSVGenerationError(
-                    f"Failed to generate {table_name}.csv: {e}"
-                ) from e
-
-        return results
-
-    def _generate_table(self, table_name: str, columns: list[str]) -> int:
-        """Generate a single CSV file.
-
-        Args:
-            table_name: Name of the table/CSV.
-            columns: Expected column order (contract).
-
-        Returns:
-            Number of rows written.
-        """
-        # Query the table
-        column_list = ", ".join(columns)
-        df = pd.read_sql_query(
-            f"SELECT {column_list} FROM {table_name}",  # noqa: S608
-            self.db.connection,
-        )
-
-        # Ensure column order matches contract exactly (Invariant 1)
-        df = df[columns]
-
-        # Deterministic row ordering (Adjustment 3)
-        sort_keys = SORT_KEYS.get(table_name, columns[:1])
-        df = df.sort_values(by=sort_keys, ascending=True)
-
-        # Write CSV with deterministic settings
-        output_path = self.output_dir / f"{table_name}.csv"
-        df.to_csv(
-            output_path,
-            index=False,
-            encoding="utf-8",
-            lineterminator="\n",  # Unix line endings for consistency
-            date_format="%Y-%m-%dT%H:%M:%S",  # Consistent datetime format
-        )
-
-        return len(df)
-
-    def validate_schemas(self) -> bool:
-        """Validate that generated CSVs match expected schemas.
-
-        Returns:
-            True if all schemas valid.
-
-        Raises:
-            CSVGenerationError: If any schema mismatch.
-        """
-        for table_name, expected_columns in CSV_SCHEMAS.items():
-            csv_path = self.output_dir / f"{table_name}.csv"
-
-            if not csv_path.exists():
-                raise CSVGenerationError(f"Missing CSV: {csv_path}")
-
-            df = pd.read_csv(csv_path, nrows=0)  # Just read headers
-            actual_columns = list(df.columns)
-
-            if actual_columns != expected_columns:
-                raise CSVGenerationError(
-                    f"Schema mismatch in {table_name}.csv:\n"
-                    f"  Expected: {expected_columns}\n"
-                    f"  Actual:   {actual_columns}"
-                )
-
-        logger.info("All CSV schemas validated successfully")
-        return True
+"""CSV generator for PowerBI-compatible output.
+
+Generates CSVs that are:
+- Schema-compliant (exact columns, exact order - Invariants 1-4)
+- Deterministic (same DB → same bytes - Adjustment 3)
+"""
+
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+import pandas as pd
+
+from ..persistence.models import CSV_SCHEMAS, SORT_KEYS
+
+if TYPE_CHECKING:
+    from ..persistence.database import DatabaseManager
+
+logger = logging.getLogger(__name__)
+
+
+class CSVGenerationError(Exception):
+    """CSV generation failed."""
+
+
+class CSVGenerator:
+    """Generates PowerBI-compatible CSV files from SQLite.
+
+    Invariant 1: CSV schema is a hard contract.
+    Invariant 3: CSV output must be deterministic.
+    """
+
+    def __init__(self, db: DatabaseManager, output_dir: Path) -> None:
+        """Initialize the CSV generator.
+
+        Args:
+            db: Database manager instance.
+            output_dir: Directory for CSV output files.
+        """
+        self.db = db
+        self.output_dir = output_dir
+
+    def generate_all(self) -> dict[str, int]:
+        """Generate all CSV files.
+
+        Returns:
+            Dict mapping table names to row counts.
+
+        Raises:
+            CSVGenerationError: If generation fails.
+        """
+        self.output_dir.mkdir(parents=True, exist_ok=True)
+
+        results: dict[str, int] = {}
+
+        for table_name, columns in CSV_SCHEMAS.items():
+            try:
+                count = self._generate_table(table_name, columns)
+                results[table_name] = count
+                logger.info(f"Generated {table_name}.csv: {count} rows")
+            except Exception as e:
+                raise CSVGenerationError(
+                    f"Failed to generate {table_name}.csv: {e}"
+                ) from e
+
+        return results
+
+    def _generate_table(self, table_name: str, columns: list[str]) -> int:
+        """Generate a single CSV file.
+
+        Args:
+            table_name: Name of the table/CSV.
+            columns: Expected column order (contract).
+
+        Returns:
+            Number of rows written.
+        """
+        # Query the table
+        column_list = ", ".join(columns)
+        df = pd.read_sql_query(
+            f"SELECT {column_list} FROM {table_name}",  # noqa: S608
+            self.db.connection,
+        )
+
+        # Ensure column order matches contract exactly (Invariant 1)
+        df = df[columns]
+
+        # Deterministic row ordering (Adjustment 3)
+        sort_keys = SORT_KEYS.get(table_name, columns[:1])
+        df = df.sort_values(by=sort_keys, ascending=True)
+
+        # Write CSV with deterministic settings
+        output_path = self.output_dir / f"{table_name}.csv"
+        df.to_csv(
+            output_path,
+            index=False,
+            encoding="utf-8",
+            lineterminator="\n",  # Unix line endings for consistency
+            date_format="%Y-%m-%dT%H:%M:%S",  # Consistent datetime format
+        )
+
+        return len(df)
+
+    def validate_schemas(self) -> bool:
+        """Validate that generated CSVs match expected schemas.
+
+        Returns:
+            True if all schemas valid.
+
+        Raises:
+            CSVGenerationError: If any schema mismatch.
+        """
+        for table_name, expected_columns in CSV_SCHEMAS.items():
+            csv_path = self.output_dir / f"{table_name}.csv"
+
+            if not csv_path.exists():
+                raise CSVGenerationError(f"Missing CSV: {csv_path}")
+
+            df = pd.read_csv(csv_path, nrows=0)  # Just read headers
+            actual_columns = list(df.columns)
+
+            if actual_columns != expected_columns:
+                raise CSVGenerationError(
+                    f"Schema mismatch in {table_name}.csv:\n"
+                    f"  Expected: {expected_columns}\n"
+                    f"  Actual:   {actual_columns}"
+                )
+
+        logger.info("All CSV schemas validated successfully")
+        return True

ado_git_repo_insights/utils/__init__.py

@@ -1 +1 @@
-"""Utilities module for shared helper functions."""
+"""Utilities module for shared helper functions."""

ado_git_repo_insights/utils/datetime_utils.py

@@ -1,101 +1,101 @@
-"""Datetime utilities for ado-git-repo-insights.
-
-Ported from the original generate_raw_data.py to ensure identical behavior.
-"""
-
-from __future__ import annotations
-
-import logging
-from datetime import datetime
-
-logger = logging.getLogger(__name__)
-
-
-def parse_iso_datetime(date_str: str | None) -> datetime | None:
-    """Parse ISO 8601 datetime strings from ADO API.
-
-    Handles 7-digit microseconds and 'Z' suffix quirks from ADO API responses.
-    Preserved from original implementation for compatibility.
-
-    Args:
-        date_str: ISO 8601 datetime string, or None.
-
-    Returns:
-        Parsed datetime, or None if parsing fails or input is None.
-
-    Examples:
-        >>> parse_iso_datetime("2024-01-15T10:30:45.1234567Z")
-        datetime.datetime(2024, 1, 15, 10, 30, 45, 123456)
-        >>> parse_iso_datetime(None)
-        None
-    """
-    if not date_str:
-        return None
-
-    try:
-        # Remove trailing 'Z' (Zulu/UTC indicator)
-        date_str = date_str.rstrip("Z")
-
-        if "." in date_str:
-            # ADO API sometimes returns 7-digit microseconds, Python only supports 6
-            date_part, microseconds = date_str.split(".")
-            microseconds = microseconds[:6]  # Truncate to 6 digits
-            date_str = f"{date_part}.{microseconds}"
-            return datetime.strptime(date_str, "%Y-%m-%dT%H:%M:%S.%f")
-        else:
-            # No microseconds
-            return datetime.strptime(date_str, "%Y-%m-%dT%H:%M:%S")
-
-    except ValueError as e:
-        logger.warning(f"Failed to parse date '{date_str}': {e}")
-        return None
-
-
-def calculate_cycle_time_minutes(
-    creation_date: str | None, closed_date: str | None
-) -> float | None:
-    """Calculate PR cycle time in minutes.
-
-    Cycle time is the duration from PR creation to closure.
-    Minimum value is 1 minute to avoid zero/negative values.
-
-    Args:
-        creation_date: ISO 8601 creation date string.
-        closed_date: ISO 8601 closed date string.
-
-    Returns:
-        Cycle time in minutes (minimum 1.0), or None if dates are invalid.
-
-    Examples:
-        >>> calculate_cycle_time_minutes(
-        ...     "2024-01-15T10:00:00Z",
-        ...     "2024-01-15T10:30:00Z"
-        ... )
-        30.0
-    """
-    created = parse_iso_datetime(creation_date)
-    closed = parse_iso_datetime(closed_date)
-
-    if created and closed:
-        delta_seconds = (closed - created).total_seconds()
-        minutes = delta_seconds / 60
-        # Minimum 1 minute, rounded to 2 decimal places
-        return max(1.0, round(minutes, 2))
-
-    return None
-
-
-def format_date_for_api(dt: datetime) -> str:
-    """Format a datetime for ADO API queries.
-
-    Args:
-        dt: Datetime to format.
-
-    Returns:
-        ISO 8601 formatted string with 'Z' suffix.
-
-    Examples:
-        >>> format_date_for_api(datetime(2024, 1, 15, 10, 30, 0))
-        '2024-01-15T10:30:00Z'
-    """
-    return dt.strftime("%Y-%m-%dT%H:%M:%SZ")
+"""Datetime utilities for ado-git-repo-insights.
+
+Ported from the original generate_raw_data.py to ensure identical behavior.
+"""
+
+from __future__ import annotations
+
+import logging
+from datetime import datetime
+
+logger = logging.getLogger(__name__)
+
+
+def parse_iso_datetime(date_str: str | None) -> datetime | None:
+    """Parse ISO 8601 datetime strings from ADO API.
+
+    Handles 7-digit microseconds and 'Z' suffix quirks from ADO API responses.
+    Preserved from original implementation for compatibility.
+
+    Args:
+        date_str: ISO 8601 datetime string, or None.
+
+    Returns:
+        Parsed datetime, or None if parsing fails or input is None.
+
+    Examples:
+        >>> parse_iso_datetime("2024-01-15T10:30:45.1234567Z")
+        datetime.datetime(2024, 1, 15, 10, 30, 45, 123456)
+        >>> parse_iso_datetime(None)
+        None
+    """
+    if not date_str:
+        return None
+
+    try:
+        # Remove trailing 'Z' (Zulu/UTC indicator)
+        date_str = date_str.rstrip("Z")
+
+        if "." in date_str:
+            # ADO API sometimes returns 7-digit microseconds, Python only supports 6
+            date_part, microseconds = date_str.split(".")
+            microseconds = microseconds[:6]  # Truncate to 6 digits
+            date_str = f"{date_part}.{microseconds}"
+            return datetime.strptime(date_str, "%Y-%m-%dT%H:%M:%S.%f")
+        else:
+            # No microseconds
+            return datetime.strptime(date_str, "%Y-%m-%dT%H:%M:%S")
+
+    except ValueError as e:
+        logger.warning(f"Failed to parse date '{date_str}': {e}")
+        return None
+
+
+def calculate_cycle_time_minutes(
+    creation_date: str | None, closed_date: str | None
+) -> float | None:
+    """Calculate PR cycle time in minutes.
+
+    Cycle time is the duration from PR creation to closure.
+    Minimum value is 1 minute to avoid zero/negative values.
+
+    Args:
+        creation_date: ISO 8601 creation date string.
+        closed_date: ISO 8601 closed date string.
+
+    Returns:
+        Cycle time in minutes (minimum 1.0), or None if dates are invalid.
+
+    Examples:
+        >>> calculate_cycle_time_minutes(
+        ...     "2024-01-15T10:00:00Z",
+        ...     "2024-01-15T10:30:00Z"
+        ... )
+        30.0
+    """
+    created = parse_iso_datetime(creation_date)
+    closed = parse_iso_datetime(closed_date)
+
+    if created and closed:
+        delta_seconds = (closed - created).total_seconds()
+        minutes = delta_seconds / 60
+        # Minimum 1 minute, rounded to 2 decimal places
+        return max(1.0, round(minutes, 2))
+
+    return None
+
+
+def format_date_for_api(dt: datetime) -> str:
+    """Format a datetime for ADO API queries.
+
+    Args:
+        dt: Datetime to format.
+
+    Returns:
+        ISO 8601 formatted string with 'Z' suffix.
+
+    Examples:
+        >>> format_date_for_api(datetime(2024, 1, 15, 10, 30, 0))
+        '2024-01-15T10:30:00Z'
+    """
+    return dt.strftime("%Y-%m-%dT%H:%M:%SZ")