sumry 0.2.0__tar.gz → 0.3.1__tar.gz

{sumry-0.2.0/src/sumry.egg-info → sumry-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sumry
-Version: 0.2.0
+Version: 0.3.1
 Summary: A data summarization tool for analyzing various file formats
 Author: Sparkgeo
 License: MIT

{sumry-0.2.0 → sumry-0.3.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sumry"
-version = "0.2.0"
+version = "0.3.1"
 description = "A data summarization tool for analyzing various file formats"
 readme = "README.md"
 requires-python = ">=3.12"

{sumry-0.2.0 → sumry-0.3.1}/src/sumry/cli.py

@@ -93,7 +93,7 @@ def main(
         
         # Display sample records if requested
         if sample is not None or show_sample:
-            sample_count = sample if sample and sample > 0 else 5
+            sample_count = sample if sample is not None and sample != 0 else 5
             display_sample_records(file_path, file_type, sample_count)
         
     except Exception as e:
@@ -102,28 +102,51 @@ def main(
 
 
 def display_sample_records(file_path: Path, file_type: str, sample_count: int):
-    """Display sample records from the file."""
+    """Display sample records from the file.
+
+    Args:
+        sample_count: Number of records to show. Positive for first N, negative for last N.
+    """
     import pandas as pd
     import geopandas as gpd
-    
-    console.print(f"\n[bold cyan]Sample Records (first {sample_count}):[/bold cyan]")
-    
+
+    # Determine if we're showing first or last records
+    show_last = sample_count < 0
+    abs_count = abs(sample_count)
+    position = "last" if show_last else "first"
+
+    console.print(f"\n[bold cyan]Sample Records ({position} {abs_count}):[/bold cyan]")
+
     try:
         # Read the data based on file type
         if file_type == "CSV":
-            df = pd.read_csv(file_path, nrows=sample_count)
+            if show_last:
+                df = pd.read_csv(file_path).tail(abs_count)
+            else:
+                df = pd.read_csv(file_path, nrows=abs_count)
         elif file_type == "Excel":
-            df = pd.read_excel(file_path, nrows=sample_count)
+            if show_last:
+                df = pd.read_excel(file_path).tail(abs_count)
+            else:
+                df = pd.read_excel(file_path, nrows=abs_count)
         elif file_type == "Parquet":
-            df = pd.read_parquet(file_path).head(sample_count)
+            full_df = pd.read_parquet(file_path)
+            df = full_df.tail(abs_count) if show_last else full_df.head(abs_count)
         elif file_type in ["GeoJSON", "Shapefile", "GeoParquet"]:
             # For spatial files, use geopandas
             if file_type == "GeoJSON":
-                gdf = gpd.read_file(file_path, rows=sample_count)
+                if show_last:
+                    gdf = gpd.read_file(file_path).tail(abs_count)
+                else:
+                    gdf = gpd.read_file(file_path, rows=abs_count)
             elif file_type == "Shapefile":
-                gdf = gpd.read_file(file_path, rows=sample_count)
+                if show_last:
+                    gdf = gpd.read_file(file_path).tail(abs_count)
+                else:
+                    gdf = gpd.read_file(file_path, rows=abs_count)
             else:  # GeoParquet
-                gdf = gpd.read_parquet(file_path).head(sample_count)
+                full_gdf = gpd.read_parquet(file_path)
+                gdf = full_gdf.tail(abs_count) if show_last else full_gdf.head(abs_count)
 
             # Replace geometry column with placeholder
             df = gdf.copy()

{sumry-0.2.0 → sumry-0.3.1}/src/sumry/formats/xlsx.py

@@ -82,18 +82,33 @@ class XLSX(BaseFormat):
         
         # Read sample data from the specified layer
         try:
-            sample_query = f"""
-                SELECT * FROM st_read('{self.file}', layer='{self.layer}')
-                LIMIT {count}
-            """
-            df = duckdb.sql(sample_query).df()
-            
+            abs_count = abs(count)
+
+            if count < 0:
+                # Get last N records by using row number and ordering
+                sample_query = f"""
+                    SELECT * EXCLUDE (_rn) FROM (
+                        SELECT *, ROW_NUMBER() OVER () as _rn
+                        FROM st_read('{self.file}', layer='{self.layer}')
+                    ) ORDER BY _rn DESC LIMIT {abs_count}
+                """
+                df = duckdb.sql(sample_query).df()
+                # Reverse to show in original order
+                df = df.iloc[::-1].reset_index(drop=True)
+            else:
+                sample_query = f"""
+                    SELECT * FROM st_read('{self.file}', layer='{self.layer}')
+                    LIMIT {abs_count}
+                """
+                df = duckdb.sql(sample_query).df()
+
             if df.empty:
                 console.print(f"[yellow]No data found in layer '{self.layer}'[/yellow]")
                 return
-            
+
+            position = "last" if count < 0 else "first"
             # Create a Rich table
-            table = Table(title=f"Sample data from '{self.layer}' (showing {min(count, len(df))} of {count} requested records)")
+            table = Table(title=f"Sample data from '{self.layer}' (showing {position} {len(df)} records)")
             
             # Add columns to the table
             for col in df.columns:

{sumry-0.2.0 → sumry-0.3.1}/src/sumry/main.py

@@ -18,7 +18,7 @@ def info(
 def sample(
     file: str = typer.Argument(..., help="Source file."),
     layer: str = typer.Argument(None, help="Layer name."),
-    count: int = typer.Option(5, "--count", "-n", help="Number of sample records to display."),
+    count: int = typer.Option(5, "--count", "-n", help="Number of sample records to display. Use negative values to show records from the end (e.g., -5 for last 5 records)."),
 ):
     xlsx = XLSX(file, layer=layer)
     xlsx.sample(count=count)

{sumry-0.2.0 → sumry-0.3.1/src/sumry.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sumry
-Version: 0.2.0
+Version: 0.3.1
 Summary: A data summarization tool for analyzing various file formats
 Author: Sparkgeo
 License: MIT