workpytools 0.1.4__py3-none-any.whl

workpytools/processing/profiler.py

@@ -0,0 +1,277 @@
+from __future__ import annotations
+
+import argparse
+import logging
+from pathlib import Path
+
+from workpytools.common.clipboard import copy_text_to_clipboard, get_clipboard_text
+from workpytools.common.tabular import (
+    DEFAULT_EMPTY_VALUES,
+    ColumnProfile,
+    Table,
+    format_top,
+    load_tables,
+    profile_columns,
+    read_csv_like,
+)
+from workpytools.processing.base import Processor
+
+logger = logging.getLogger(__name__)
+
+_HEADER = (
+    "source",
+    "sheet",
+    "column",
+    "rows",
+    "filled",
+    "empty",
+    "fill_rate",
+    "unique",
+    "unique_rate",
+    "is_filled",
+    "is_unique",
+    "key_score",
+    "top",
+)
+
+
+def _profile_row(source: str, sheet: str, profile: ColumnProfile) -> list[str]:
+    return [
+        source,
+        sheet,
+        profile.column,
+        str(profile.rows),
+        str(profile.filled),
+        str(profile.empty),
+        f"{profile.fill_rate:.4f}",
+        str(profile.unique),
+        f"{profile.unique_rate:.4f}",
+        str(profile.is_filled),
+        str(profile.is_unique),
+        f"{profile.key_score:.4f}",
+        format_top(profile.top),
+    ]
+
+
+class ProfilerProcessor(Processor):
+    """Profile every column of tabular data (CSV/TSV/Excel/JSON): fill rate,
+    uniqueness, top values, and a key-likelihood score. Reads from a file or,
+    with no argument, from the clipboard (e.g. a range copied from Excel).
+    """
+
+    name = "profiler"
+    help = "表形式データの各列をプロファイルする（欠損・一意性・頻度上位・主キー候補）"
+
+    def add_arguments(self, parser: argparse.ArgumentParser) -> None:
+        parser.add_argument(
+            "path",
+            nargs="*",
+            default=[],
+            help="対象ファイル（複数・グロブ可）。省略時はクリップボードのTSVを読む",
+        )
+        parser.add_argument("-o", "--output", default=None, help="出力先パス（.tsv / .xlsx）")
+        parser.add_argument(
+            "--clip", action="store_true", help="結果をクリップボードにコピーする"
+        )
+        parser.add_argument(
+            "--view", action="store_true", help="結果をブラウザでHTML表示する"
+        )
+        parser.add_argument("--sep", default=None, help="区切り文字（拡張子判定を上書き）")
+        parser.add_argument(
+            "--header",
+            type=int,
+            default=0,
+            help="ヘッダー行のインデックス（0始まり、既定0）",
+        )
+        parser.add_argument(
+            "--no-header", action="store_true", help="ヘッダーなしとして扱う"
+        )
+        parser.add_argument(
+            "--top", type=int, default=10, help="頻度上位の表示件数（既定10、0で非表示）"
+        )
+        parser.add_argument(
+            "--empty-values",
+            default=None,
+            help="空とみなす値のカンマ区切りリスト（既定値を置き換える）",
+        )
+        parser.add_argument(
+            "--no-default-empty-values",
+            action="store_true",
+            help="既定の空値リストを無効化し、Noneと空文字だけを空とみなす",
+        )
+
+    def run(self, args: argparse.Namespace) -> int:
+        empty_values = self._resolve_empty_values(args)
+
+        sources = self._resolve_sources(args)
+        has_excel = any(
+            Path(s).suffix.lower() == ".xlsx" for s, _ in sources if s != "(clipboard)"
+        )
+
+        rows: list[list[str]] = [list(_HEADER)] if has_excel else [
+            [h for h in _HEADER if h != "sheet"]
+        ]
+
+        for source_label, table_by_sheet in sources:
+            for sheet_name, table in table_by_sheet.items():
+                if not table.rows:
+                    label = f"{source_label} [{sheet_name}]" if sheet_name else source_label
+                    logger.warning("データ行が0件のためスキップします: %s", label)
+                    continue
+                for profile in profile_columns(table, top_n=args.top, empty_values=empty_values):
+                    row = _profile_row(source_label, sheet_name, profile)
+                    if not has_excel:
+                        row = [v for v, h in zip(row, _HEADER, strict=True) if h != "sheet"]
+                    rows.append(row)
+
+        if len(rows) <= 1:
+            raise SystemExit("プロファイル対象の列がありません（入力が空の可能性があります）")
+
+        self._emit(rows, args)
+        return 0
+
+    def _resolve_empty_values(self, args: argparse.Namespace) -> frozenset[object]:
+        if args.no_default_empty_values:
+            base: frozenset[object] = frozenset({None, ""})
+        else:
+            base = DEFAULT_EMPTY_VALUES
+
+        if args.empty_values:
+            return frozenset(v.strip() for v in args.empty_values.split(","))
+        return base
+
+    def _resolve_sources(
+        self, args: argparse.Namespace
+    ) -> list[tuple[str, dict[str, Table]]]:
+        if not args.path:
+            try:
+                text = get_clipboard_text()
+            except Exception as exc:
+                raise SystemExit("クリップボードにテキストがありません") from exc
+            sep = args.sep if args.sep is not None else "\t"
+            header_row = None if args.no_header else args.header
+            table = read_csv_like(text, sep, header_row)
+            return [("(clipboard)", {"": table})]
+
+        from glob import glob
+
+        results = []
+        header_row = None if args.no_header else args.header
+        for pattern in args.path:
+            matches = glob(pattern) or [pattern]
+            for filename in matches:
+                path = Path(filename)
+                try:
+                    tables = load_tables(path, args.sep, header_row)
+                except ValueError as exc:
+                    raise SystemExit(str(exc)) from exc
+                except OSError as exc:
+                    raise SystemExit(f"ファイルを読み込めません: {path}") from exc
+                results.append((str(path), tables))
+
+        if not results:
+            raise SystemExit("対象ファイルが見つかりません")
+        return results
+
+    def _emit(self, rows: list[list[str]], args: argparse.Namespace) -> None:
+        tsv_text = "\n".join("\t".join(row) for row in rows)
+
+        if args.clip:
+            copy_text_to_clipboard(tsv_text)
+            print("プロファイル結果をコピーしました")
+            return
+
+        if args.view:
+            self._view_in_browser(rows)
+            return
+
+        if args.output:
+            self._write_output(rows, tsv_text, args.output)
+            return
+
+        print(tsv_text)
+
+    def _write_output(self, rows: list[list[str]], tsv_text: str, output: str) -> None:
+        output_path = Path(output)
+        if output_path.suffix.lower() == ".xlsx":
+            self._write_excel(rows, output_path)
+        else:
+            output_path.write_text(tsv_text, encoding="utf-8")
+        logger.info("プロファイル結果を書き出しました: %s", output_path)
+        print(output_path)
+
+    def _write_excel(self, rows: list[list[str]], output_path: Path) -> None:
+        import xlsxwriter
+
+        header = rows[0]
+        bad_cols = {i for i, h in enumerate(header) if h in ("is_filled", "is_unique")}
+
+        with xlsxwriter.Workbook(str(output_path)) as workbook:
+            worksheet = workbook.add_worksheet("profile")
+            header_format = workbook.add_format(
+                {"bold": True, "align": "center", "border": 1}
+            )
+            worksheet.write_row(0, 0, header, header_format)
+
+            red_format = workbook.add_format({"bg_color": "#FFC7CE", "font_color": "#9C0006"})
+            for r, row in enumerate(rows[1:], start=1):
+                worksheet.write_row(r, 0, row)
+                for c in bad_cols:
+                    if row[c] == "False":
+                        worksheet.write(r, c, row[c], red_format)
+
+            worksheet.autofilter(0, 0, len(rows) - 1, len(header) - 1)
+
+    def _view_in_browser(self, rows: list[list[str]]) -> None:
+        from workpytools.common.browser_preview import write_and_open
+
+        html = _render_html_table(rows)
+        write_and_open(html, "workpytools_profiler_preview.html", no_open=False)
+
+
+def _render_html_table(rows: list[list[str]]) -> str:
+    import html as html_module
+
+    header, *data_rows = rows
+    bad_cols = {i for i, h in enumerate(header) if h in ("is_filled", "is_unique")}
+
+    head_html = "".join(f"<th>{html_module.escape(h)}</th>" for h in header)
+    body_html = []
+    for row in data_rows:
+        cells = []
+        for i, value in enumerate(row):
+            cls = ' class="bad"' if i in bad_cols and value == "False" else ""
+            cells.append(f"<td{cls}>{html_module.escape(value)}</td>")
+        body_html.append(f"<tr>{''.join(cells)}</tr>")
+
+    style = """\
+:root { color-scheme: light dark; }
+body { font-family: -apple-system, "Segoe UI", sans-serif; margin: 1.5rem; }
+table { border-collapse: collapse; width: 100%; }
+th, td { border: 1px solid #999; padding: 0.3rem 0.6rem; text-align: left; }
+th { position: sticky; top: 0; background: #eee; }
+td.bad { background: #ffdcdc; }
+@media (prefers-color-scheme: dark) {
+  body { background: #1e1e1e; color: #ddd; }
+  th { background: #333; }
+  td.bad { background: #5a2a2a; }
+  th, td { border-color: #555; }
+}
+"""
+
+    return f"""<!doctype html>
+<html>
+<head>
+<meta charset="utf-8">
+<meta http-equiv="Cache-Control" content="no-store">
+<title>profiler result</title>
+<style>{style}</style>
+</head>
+<body>
+<table>
+<thead><tr>{head_html}</tr></thead>
+<tbody>{"".join(body_html)}</tbody>
+</table>
+</body>
+</html>
+"""

workpytools/processing/seiretsu.py

@@ -0,0 +1,111 @@
+from __future__ import annotations
+
+import argparse
+import logging
+
+from workpytools.common.powerpoint import (
+    NoActivePresentationError,
+    PowerPointNotRunningError,
+    get_active_presentation,
+    get_running_powerpoint,
+)
+from workpytools.common.table_shapes import (
+    GRID_TOLERANCE,
+    DuplicateGridPositionError,
+    GridShape,
+    column_and_row_sizes,
+    estimate_grid,
+    grid_centers,
+)
+from workpytools.processing.base import Processor
+
+logger = logging.getLogger(__name__)
+
+_SELECTION_SHAPES = 2  # ppSelectionShapes
+
+
+class SeiretsuProcessor(Processor):
+    """Realign selected shapes into a grid by position only. Unlike tbl's
+    compose direction, this never creates a table or deletes/creates
+    shapes -- each shape keeps its size, text, and formatting; only
+    Left/Top are rewritten so shapes' centers land on grid positions.
+    """
+
+    name = "seiretsu"
+    help = "選択したシェイプを表に変換せず格子状の位置に整列する"
+
+    def add_arguments(self, parser: argparse.ArgumentParser) -> None:
+        pass
+
+    def run(self, args: argparse.Namespace) -> int:
+        try:
+            app = get_running_powerpoint()
+        except PowerPointNotRunningError as exc:
+            raise SystemExit(
+                "PowerPointでプレゼンテーションを開いた状態で実行してください"
+            ) from exc
+
+        try:
+            get_active_presentation(app)
+        except NoActivePresentationError as exc:
+            raise SystemExit(
+                "PowerPointでプレゼンテーションを開いた状態で実行してください"
+            ) from exc
+
+        try:
+            selection = app.ActiveWindow.Selection
+        except Exception as exc:
+            raise SystemExit(f"選択状態の取得中にエラーが発生しました: {exc}") from exc
+
+        if selection.Type != _SELECTION_SHAPES:
+            raise SystemExit("対象を選択してから実行してください（2つ以上選択が必要です）")
+
+        shape_range = selection.ShapeRange
+        shapes = [shape_range.Item(i) for i in range(1, shape_range.Count + 1)]
+
+        if len(shapes) < 2:
+            raise SystemExit("対象を選択してから実行してください（2つ以上選択が必要です）")
+
+        grid_shapes = [
+            GridShape(left=s.Left, top=s.Top, width=s.Width, height=s.Height, ref=s)
+            for s in shapes
+        ]
+
+        try:
+            positions, n_rows, n_cols = estimate_grid(grid_shapes, GRID_TOLERANCE)
+        except DuplicateGridPositionError as exc:
+            raise SystemExit(str(exc)) from exc
+
+        missing = n_rows * n_cols - len(positions)
+        if missing:
+            logger.warning("歯抜けのグリッド位置: %d件", missing)
+
+        col_width, row_height = column_and_row_sizes(positions, n_rows, n_cols)
+
+        overall_left = min(s.left for s in grid_shapes)
+        overall_top = min(s.top for s in grid_shapes)
+
+        center_x, center_y = grid_centers(col_width, row_height, overall_left, overall_top)
+
+        try:
+            app.StartNewUndoEntry()
+            for pos in positions:
+                shape = pos.shape.ref
+                new_left = center_x[pos.col] - pos.shape.width / 2
+                new_top = center_y[pos.row] - pos.shape.height / 2
+                shape.Left = new_left  # type: ignore[attr-defined]
+                shape.Top = new_top  # type: ignore[attr-defined]
+        except Exception as exc:
+            raise SystemExit(
+                f"PowerPointの操作中にエラーが発生しました（Ctrl+Zで開始前の状態に戻せます）: {exc}"
+            ) from exc
+
+        logger.info(
+            "対象シェイプ数=%d 推定グリッド=%d行%d列 歯抜け=%d",
+            len(shapes),
+            n_rows,
+            n_cols,
+            missing,
+        )
+        print(f"{len(shapes)}個のシェイプを{n_rows}行{n_cols}列に整列しました")
+        return 0