Initial upload: shell-history-semantic-search v0.1.0

src/shell_history_search/cli/commands.py

@@ -0,0 +1,109 @@
+import click
+import json
+import logging
+from datetime import datetime
+
+from ..core import SearchEngine, IndexingService
+
+logger = logging.getLogger(__name__)
+
+
+@click.group()
+@click.option("-v", "--verbose", count=True, help="Increase verbosity")
+def cli(verbose: int) -> None:
+    log_level = logging.WARNING
+    if verbose == 1:
+        log_level = logging.INFO
+    elif verbose >= 2:
+        log_level = logging.DEBUG
+    logging.basicConfig(level=log_level, format="%(levelname)s: %(message)s")
+
+
+@cli.command()
+@click.option("--shell", type=str, help="Index only specific shell (bash, zsh, fish)")
+def index(shell: str | None) -> None:
+    click.echo("Indexing shell history...")
+
+    indexing_service = IndexingService()
+
+    shell_type = shell.lower() if shell else None
+    result = indexing_service.index_shell_history(shell_type)
+
+    click.echo(
+        f"Indexed {result['total_indexed']} commands "
+        f"(skipped {result['total_skipped']} duplicates)"
+    )
+
+
+@cli.command()
+@click.argument("query")
+@click.option("--limit", "-n", type=int, default=10, help="Number of results to return")
+@click.option("--shell", type=str, help="Filter by shell type (bash, zsh, fish)")
+@click.option("--json", "json_output", is_flag=True, help="Output as JSON")
+def search(query: str, limit: int, shell: str | None, json_output: bool) -> None:
+    search_engine = SearchEngine()
+
+    shell_type = shell.lower() if shell else None
+
+    results = search_engine.search(query, limit=limit, shell_type=shell_type)
+
+    if not results:
+        click.echo("No results found.")
+        return
+
+    if json_output:
+        output = [
+            {
+                "command": r.command,
+                "shell": r.shell_type,
+                "timestamp": r.timestamp,
+                "similarity": round(r.similarity, 4),
+            }
+            for r in results
+        ]
+        click.echo(json.dumps(output, indent=2))
+    else:
+        for i, result in enumerate(results, 1):
+            timestamp_str = ""
+            if result.timestamp:
+                dt = datetime.fromtimestamp(result.timestamp)
+                timestamp_str = dt.strftime("%Y-%m-%d %H:%M")
+
+            similarity_pct = result.similarity * 100
+            click.echo(f"{i}. [{result.shell_type}] {timestamp_str}")
+            click.echo(f"   {result.command}")
+            click.echo(f"   Similarity: {similarity_pct:.1f}%")
+            click.echo()
+
+
+@cli.command()
+@click.option("--json", "json_output", is_flag=True, help="Output as JSON")
+def stats(json_output: bool) -> None:
+    search_engine = SearchEngine()
+    stats_data = search_engine.get_stats()
+
+    if json_output:
+        click.echo(json.dumps(stats_data, indent=2))
+    else:
+        click.echo("Shell History Search Statistics")
+        click.echo("=" * 40)
+        click.echo(f"Total commands: {stats_data['total_commands']}")
+        click.echo(f"Total embeddings: {stats_data['total_embeddings']}")
+        click.echo(f"Embedding model: {stats_data['embedding_model']}")
+        click.echo(f"Embedding dimension: {stats_data['embedding_dim']}")
+        click.echo()
+        click.echo("Commands by shell:")
+        for shell_type, count in stats_data["shell_counts"].items():
+            click.echo(f"  {shell_type}: {count}")
+
+
+@cli.command()
+@click.confirmation_option(prompt="Are you sure you want to clear all indexed data?")
+def clear() -> None:
+    search_engine = SearchEngine()
+    search_engine.clear_all()
+    click.echo("All indexed data has been cleared.")
+
+
+if __name__ == "__main__":
+    cli()