pq-peek 0.1.1__py3-none-any.whl

pq_peek/__init__.py

@@ -0,0 +1,2 @@
+def hello() -> str:
+    return "Hello from pq-peek!"

pq_peek/__main__.py

@@ -0,0 +1,4 @@
+from .main import app
+
+if __name__ == "__main__":
+    app()

pq_peek/main.py

@@ -0,0 +1,98 @@
+from pathlib import Path
+
+import polars as pl
+import typer
+from rich import print as rprint
+from rich.console import Console
+from rich.table import Table
+
+app = typer.Typer(help="A fast CLI tool to inspect Parquet files")
+console = Console()
+
+def get_scan(file_path: Path) -> pl.LazyFrame:
+    verify_path(file_path)
+    return pl.scan_parquet(file_path)
+
+def verify_path(file_path: Path):
+    """
+    Verifies that the file exists and checks the correct ending 
+    """
+    if not file_path.exists():
+        rprint(f"[bold red]Error:[/bold red] File was not found: '{file_path}'")
+        raise typer.Exit(code=1)
+    if file_path.suffix != ".parquet":
+        rprint("[bold yellow]Warning:[/bold yellow] File does not end with '.parquet'")
+
+@app.command()
+def schema(file_path: Path):
+    """
+    Displays the schema (column names, types) of the parquet file
+    Uses Lazy Loading (scan_parquet) to save memory
+    """
+    try:
+        file = get_scan(file_path)
+        schema = file.collect_schema()
+
+        table = Table(title=f"Schema: {file_path.name}")
+        table.add_column("Column name", style="cyan", no_wrap=True)
+        table.add_column("Type", style="magenta")
+
+        for name, dtype in schema.items():
+            table.add_row(name, str(dtype))
+
+        console.print(table)
+        rprint(f"\n[green]Amount of columns:[/green] {len(schema)}")
+    except Exception as failure:
+        rprint(f"[bold red]Failure while reading the parquet file:[/bold red] {failure}")
+        raise typer.Exit(code=1)
+
+@app.command()
+def head(file_path: Path, n: int = typer.Option(5, help="Amount of rows")):
+    """
+    Displays the first n rows of the parquet file
+    """
+    try:
+        file = get_scan(file_path)
+        df = file.limit(n).collect()
+
+        if df.is_empty():
+            rprint("[yellow]File is empty.[/yellow]")
+            return
+        
+        table = Table(title=f"Preview {file_path.name} ({n} rows)")
+
+        for column in df.columns:
+            table.add_column(column, overflow="fold")
+
+        for row in df.iter_rows():
+            str_row = [str(x) for x in row]
+            table.add_row(*str_row)
+        
+        console.print(table)
+    except Exception as failure:
+        rprint(f"[bold red]Failure while reading the parquet file:[/bold red] {failure}")
+        raise typer.Exit(code=1)
+
+
+
+@app.command()
+def stats(file_path: Path):
+    """
+    Displays stats of the parquet file (min, max, nulls)
+    Uses Polars Query Engine for parallel computation
+    """
+    try:
+        with console.status("[bold green]Calculating stats...[/bold green]"):
+            file = get_scan(file_path)
+
+            stats_df = file.describe()
+
+        console.print(stats_df)
+
+    except Exception as failure:
+        rprint(f"[bold red]Failure while calculating stats:[/bold red] {failure}")
+        raise typer.Exit(code=1)
+
+
+if __name__ == "__main__":
+    app()

pq_peek-0.1.1.dist-info/METADATA

@@ -0,0 +1,47 @@
+Metadata-Version: 2.3
+Name: pq-peek
+Version: 0.1.1
+Summary: A blazing fast CLI tool to inspect Parquet files using Polars.
+Author: Lars Lewerenz
+Author-email: Lars Lewerenz <git.lars.lewerenz@gmail.com>
+Requires-Dist: polars>=1.37.1
+Requires-Dist: rich>=14.3.1
+Requires-Dist: typer>=0.21.1
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+
+pq-peek 🦆
+
+A blazing fast, memory-efficient CLI tool to inspect large Parquet files directly in the terminal.
+Built with *Polars*, *Typer*, and *Rich*. Managed via *uv*.
+
+## Install (uv)
+
+```bash
+uv pip install pq-peek
+```
+
+## CLI usage
+
+```bash
+pq-peek schema /path/to/file.parquet
+pq-peek head /path/to/file.parquet --n 5
+pq-peek stats /path/to/file.parquet
+```
+
+## Module usage
+
+```bash
+python -m pq_peek schema /path/to/file.parquet
+```
+
+## Build and publish (uv)
+
+```bash
+uv build
+uv publish
+```
+
+## Publishing notes
+
+CI publishing uses GitHub's Trusted Publisher OIDC. See `PUBLISHING.md` for the full release steps.

pq_peek-0.1.1.dist-info/RECORD

@@ -0,0 +1,8 @@
+pq_peek/__init__.py,sha256=JYq5WlD6oEtrkztuq_nXw2yyh0f8TE_tT5Icqj6NzPw,53
+pq_peek/__main__.py,sha256=RCZmmoCNOWC7rAfIDm_LaymsybXIzE6McYbUEEkf9P8,60
+pq_peek/main.py,sha256=MVmxFoA49dGKLuoDz7iNrV5bOLtcXUwBOeoWGBq9lFA,2899
+pq_peek/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+pq_peek-0.1.1.dist-info/WHEEL,sha256=fAguSjoiATBe7TNBkJwOjyL1Tt4wwiaQGtNtjRPNMQA,80
+pq_peek-0.1.1.dist-info/entry_points.txt,sha256=1UxcnCZp2LTe0m4lJeiqct4aIEecydBoYd03rh0EfnY,46
+pq_peek-0.1.1.dist-info/METADATA,sha256=_Nxm_rYOKGEg_mcJqppW_5yWvX6C-XU1d4uHS68Jz5M,980
+pq_peek-0.1.1.dist-info/RECORD,,

pq_peek-0.1.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: uv 0.9.28
+Root-Is-Purelib: true
+Tag: py3-none-any

pq_peek-0.1.1.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+pq-peek = pq_peek.main:app
+