dataframe-textual 1.0.0__py3-none-any.whl → 1.2.0__py3-none-any.whl

dataframe_textual/__init__.py

@@ -2,7 +2,7 @@
 
 from .data_frame_help_panel import DataFrameHelpPanel
 from .data_frame_table import DataFrameTable, History
-from .data_frame_viewer import DataFrameViewer, _load_dataframe
+from .data_frame_viewer import DataFrameViewer
 from .table_screen import FrequencyScreen, RowDetailScreen, TableScreen
 from .yes_no_screen import (
     ConfirmScreen,
@@ -31,5 +31,4 @@ __all__ = [
     "FilterScreen",
     "FreezeScreen",
     "OpenFileScreen",
-    "_load_dataframe",
 ]

dataframe_textual/__main__.py

@@ -4,6 +4,7 @@ import argparse
 import sys
 from pathlib import Path
 
+from .common import load_dataframe
 from .data_frame_viewer import DataFrameViewer
 
 SUPPORTED_FORMATS = ["csv", "excel", "tsv", "parquet", "json", "ndjson"]
@@ -22,13 +23,14 @@ def main() -> None:
         SystemExit: If invalid arguments are provided or required files are missing.
     """
     parser = argparse.ArgumentParser(
+        prog="dv",
         description="Interactive terminal based viewer/editor for tabular data (e.g., CSV/Excel).",
         formatter_class=argparse.RawDescriptionHelpFormatter,
         epilog="Examples:\n"
-        "  dataframe-textual data.csv\n"
-        "  dataframe-textual file1.csv file2.csv file3.csv\n"
-        "  dataframe-textual data.xlsx  (opens all sheets in tabs)\n"
-        "  cat data.csv | dataframe-textual --format csv\n",
+        "  %(prog)s data.csv\n"
+        "  %(prog)s file1.csv file2.csv file3.csv\n"
+        "  %(prog)s data.xlsx  (opens each sheet in separate tab)\n"
+        "  cat data.csv | %(prog)s --format csv\n",
     )
     parser.add_argument("files", nargs="*", help="Files to view (or read from stdin)")
     parser.add_argument(
@@ -57,7 +59,8 @@ def main() -> None:
         parser.print_help()
         sys.exit(1)
 
-    app = DataFrameViewer(*filenames, file_format=args.format, has_header=not args.no_header)
+    sources = load_dataframe(filenames, file_format=args.format, has_header=not args.no_header)
+    app = DataFrameViewer(*sources)
     app.run()
 
 

dataframe_textual/common.py

@@ -1,7 +1,9 @@
 """Common utilities and constants for dataframe_viewer."""
 
 import re
+import sys
 from dataclasses import dataclass
+from pathlib import Path
 from typing import Any
 
 import polars as pl
@@ -111,6 +113,26 @@ def DtypeConfig(dtype: pl.DataType) -> DtypeClass:
         return STYLES[pl.Unknown]
 
 
+def format_float(value: float, thousand_separator: bool = False, precision: int = 2) -> str:
+    """Format a float value, keeping integers without decimal point.
+
+    Args:
+        val: The float value to format.
+        thousand_separator: Whether to include thousand separators. Defaults to False.
+
+    Returns:
+        The formatted float as a string.
+    """
+
+    if (val := int(value)) == value:
+        return f"{val:,}" if thousand_separator else str(val)
+    else:
+        if precision > 0:
+            return f"{value:,.{precision}f}" if thousand_separator else f"{value:.{precision}f}"
+        else:
+            return f"{value:,f}" if thousand_separator else str(value)
+
+
 def format_row(vals, dtypes, apply_justify=True, thousand_separator=False) -> list[Text]:
     """Format a single row with proper styling and justification.
 
@@ -135,8 +157,8 @@ def format_row(vals, dtypes, apply_justify=True, thousand_separator=False) -> li
             text_val = NULL_DISPLAY
         elif dc.gtype == "integer" and thousand_separator:
             text_val = f"{val:,}"
-        elif dc.gtype == "float" and thousand_separator:
-            text_val = f"{val:,}"
+        elif dc.gtype == "float":
+            text_val = format_float(val, thousand_separator)
         else:
             text_val = str(val)
 
@@ -318,3 +340,122 @@ def validate_expr(term: str, df: pl.DataFrame, current_col_idx: int) -> pl.Expr
             raise ValueError(f"Failed to evaluate expression `{expr_str}`: {e}") from e
     except Exception as ve:
         raise ValueError(f"Failed to validate expression `{term}`: {ve}") from ve
+
+
+def load_dataframe(
+    filenames: list[str], file_format: str | None = None, has_header: bool = True
+) -> list[tuple[pl.LazyFrame, str, str]]:
+    """Load DataFrames from file specifications.
+
+    Handles loading from multiple files, single files, or stdin. For Excel files,
+    loads all sheets as separate entries. For other formats, loads as single file.
+
+    Args:
+        filenames: List of filenames to load. If single filename is "-", read from stdin.
+        file_format: Optional format specifier for input files (e.g., 'csv', 'excel').
+        has_header: Whether the input files have a header row. Defaults to True.
+
+    Returns:
+        List of tuples of (LazyFrame, filename, tabname) ready for display.
+    """
+    sources = []
+
+    prefix_sheet = len(filenames) > 1
+
+    for filename in filenames:
+        sources.extend(load_file(filename, prefix_sheet=prefix_sheet, file_format=file_format, has_header=has_header))
+    return sources
+
+
+def load_file(
+    filename: str,
+    first_sheet: bool = False,
+    prefix_sheet: bool = False,
+    file_format: str | None = None,
+    has_header: bool = True,
+) -> list[tuple[pl.LazyFrame, str, str]]:
+    """Load a single file and return list of sources.
+
+    For Excel files, when `first_sheet` is True, returns only the first sheet. Otherwise, returns one entry per sheet.
+    For other files or multiple files, returns one entry per file.
+
+    Args:
+        filename: Path to file to load.
+        first_sheet: If True, only load first sheet for Excel files. Defaults to False.
+        prefix_sheet: If True, prefix filename to sheet name as the tab name for Excel files. Defaults to False.
+        file_format: Optional format specifier (i.e., 'csv', 'excel', 'tsv', 'parquet', 'json', 'ndjson') for input files.
+                     By default, infers from file extension.
+        has_header: Whether the input files have a header row. Defaults to True.
+
+    Returns:
+        List of tuples of (LazyFrame, filename, tabname).
+    """
+    sources = []
+
+    if filename == "-":
+        import os
+        from io import StringIO
+
+        # Read from stdin into memory first (stdin is not seekable)
+        stdin_data = sys.stdin.read()
+        lf = pl.scan_csv(StringIO(stdin_data), has_header=has_header, separator="," if file_format == "csv" else "\t")
+
+        # Reopen stdin to /dev/tty for proper terminal interaction
+        try:
+            tty = open("/dev/tty")
+            os.dup2(tty.fileno(), sys.stdin.fileno())
+        except (OSError, FileNotFoundError):
+            pass
+
+        sources.append((lf, f"stdin.{file_format}" if file_format else "stdin", "stdin"))
+        return sources
+
+    filepath = Path(filename)
+
+    if file_format == "csv":
+        lf = pl.scan_csv(filename, has_header=has_header)
+        sources.append((lf, filename, filepath.stem))
+    elif file_format == "excel":
+        if first_sheet:
+            # Read only the first sheet for multiple files
+            lf = pl.read_excel(filename).lazy()
+            sources.append((lf, filename, filepath.stem))
+        else:
+            # For single file, expand all sheets
+            sheets = pl.read_excel(filename, sheet_id=0)
+            for sheet_name, df in sheets.items():
+                tabname = f"{filepath.stem}_{sheet_name}" if prefix_sheet else sheet_name
+                sources.append((df.lazy(), filename, tabname))
+    elif file_format == "tsv":
+        lf = pl.scan_csv(filename, has_header=has_header, separator="\t")
+        sources.append((lf, filename, filepath.stem))
+    elif file_format == "parquet":
+        lf = pl.scan_parquet(filename)
+        sources.append((lf, filename, filepath.stem))
+    elif file_format == "json":
+        df = pl.read_json(filename)
+        sources.append((df, filename, filepath.stem))
+    elif file_format == "ndjson":
+        lf = pl.scan_ndjson(filename)
+        sources.append((lf, filename, filepath.stem))
+    else:
+        ext = filepath.suffix.lower()
+        if ext == ".csv":
+            file_format = "csv"
+        elif ext in (".xlsx", ".xls"):
+            file_format = "excel"
+        elif ext in (".tsv", ".tab"):
+            file_format = "tsv"
+        elif ext == ".parquet":
+            file_format = "parquet"
+        elif ext == ".json":
+            file_format = "json"
+        elif ext == ".ndjson":
+            file_format = "ndjson"
+        else:
+            # Default to TSV
+            file_format = "tsv"
+
+        sources.extend(load_file(filename, first_sheet, prefix_sheet, file_format, has_header))
+
+    return sources