scitex 2.5.0__py3-none-any.whl → 2.7.0__py3-none-any.whl

scitex/io/_save_modules/_bibtex.py

@@ -4,6 +4,7 @@
 # File: ./src/scitex/io/_save_modules/_bibtex.py
 # ----------------------------------------
 import os
+
 __FILE__ = __file__
 __DIR__ = os.path.dirname(__FILE__)
 # ----------------------------------------
@@ -26,7 +27,7 @@ logger = logging.getLogger(__name__)
 def save_bibtex(obj: Union[Dict, List[Dict]], spath: str, **kwargs) -> None:
     """
     Save data as BibTeX file.
-    
+
     Args:
         obj: Can be:
             - A single BibTeX entry (dict)
@@ -39,7 +40,7 @@ def save_bibtex(obj: Union[Dict, List[Dict]], spath: str, **kwargs) -> None:
             - add_header: Add header comment (default: True)
             - append: Append to existing file (default: False)
             - include_enriched: Include enriched metadata for Paper objects (default: True)
-    
+
     The input dict should have structure:
     {
         'entry_type': 'article',
@@ -54,69 +55,75 @@ def save_bibtex(obj: Union[Dict, List[Dict]], spath: str, **kwargs) -> None:
     """
     # Check if it's a Paper or PaperCollection from scholar module
     obj_type = type(obj).__name__
-    if obj_type == 'PaperCollection':
+    if obj_type == "PaperCollection":
         # Get BibTeX content from PaperCollection
-        include_enriched = kwargs.get('include_enriched', True)
+        include_enriched = kwargs.get("include_enriched", True)
         content = obj._to_bibtex(include_enriched=include_enriched)
         # Write directly so scitex.io.save features work
-        encoding = kwargs.get('encoding', 'utf-8')
-        add_header = kwargs.get('add_header', True)
-        
-        with open(spath, 'w', encoding=encoding) as f:
+        encoding = kwargs.get("encoding", "utf-8")
+        add_header = kwargs.get("add_header", True)
+
+        with open(spath, "w", encoding=encoding) as f:
             if add_header:
                 f.write(f"% BibTeX bibliography\n")
-                f.write(f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n")
+                f.write(
+                    f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n"
+                )
                 f.write(f"% Number of entries: {len(obj._papers)}\n\n")
             f.write(content)
         return
-        
-    elif obj_type == 'Paper':
+
+    elif obj_type == "Paper":
         # Get BibTeX content from Paper
-        include_enriched = kwargs.get('include_enriched', True)
+        include_enriched = kwargs.get("include_enriched", True)
         content = obj._to_bibtex(include_enriched=include_enriched)
         # Write directly so scitex.io.save features work
-        encoding = kwargs.get('encoding', 'utf-8')
-        add_header = kwargs.get('add_header', True)
-        
-        with open(spath, 'w', encoding=encoding) as f:
+        encoding = kwargs.get("encoding", "utf-8")
+        add_header = kwargs.get("add_header", True)
+
+        with open(spath, "w", encoding=encoding) as f:
             if add_header:
                 f.write(f"% BibTeX entry\n")
-                f.write(f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n\n")
+                f.write(
+                    f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n\n"
+                )
             f.write(content)
         return
-    
+
     # Handle standard dict format
-    encoding = kwargs.get('encoding', 'utf-8')
-    add_header = kwargs.get('add_header', True)
-    append = kwargs.get('append', False)
-    
+    encoding = kwargs.get("encoding", "utf-8")
+    add_header = kwargs.get("add_header", True)
+    append = kwargs.get("append", False)
+
     # Ensure obj is a list
     if isinstance(obj, dict):
         entries = [obj]
     else:
         entries = obj
-    
+
     # Generate BibTeX content
     lines = []
-    
+
     if add_header and not append:
-        lines.extend([
-            f"% BibTeX bibliography",
-            f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}",
-            f"% Number of entries: {len(entries)}",
-            ""
-        ])
-    
+        lines.extend(
+            [
+                f"% BibTeX bibliography",
+                f"% Generated by SciTeX on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}",
+                f"% Number of entries: {len(entries)}",
+                "",
+            ]
+        )
+
     # Convert each entry
     for entry in entries:
         bibtex_str = _entry_to_bibtex(entry)
         lines.append(bibtex_str)
         lines.append("")  # Empty line between entries
-    
-    content = '\n'.join(lines)
-    
+
+    content = "\n".join(lines)
+
     # Write file
-    mode = 'a' if append else 'w'
+    mode = "a" if append else "w"
     with open(spath, mode, encoding=encoding) as f:
         f.write(content)
 
@@ -124,71 +131,69 @@ def save_bibtex(obj: Union[Dict, List[Dict]], spath: str, **kwargs) -> None:
 def _entry_to_bibtex(entry: Dict[str, Any]) -> str:
     """
     Convert a single entry dict to BibTeX format.
-    
+
     Args:
         entry: Dictionary with 'entry_type', 'key', and 'fields'
-    
+
     Returns:
         BibTeX formatted string
     """
     # Extract components
-    entry_type = entry.get('entry_type', 'misc')
-    key = entry.get('key', 'unknown')
-    fields = entry.get('fields', {})
-    
+    entry_type = entry.get("entry_type", "misc")
+    key = entry.get("key", "unknown")
+    fields = entry.get("fields", {})
+
     # Start building the entry
     lines = [f"@{entry_type}{{{key},"]
-    
+
     # Add fields
     for field_name, field_value in fields.items():
         # Escape special characters
         escaped_value = _escape_bibtex(str(field_value))
-        
+
         # Determine if we need braces or quotes
-        if '\n' in escaped_value or '{' in escaped_value or '}' in escaped_value:
+        if "\n" in escaped_value or "{" in escaped_value or "}" in escaped_value:
             # Use braces for multi-line or complex values
             lines.append(f"  {field_name} = {{{escaped_value}}},")
         else:
             # Use braces (standard)
             lines.append(f"  {field_name} = {{{escaped_value}}},")
-    
+
     # Remove trailing comma from last field
-    if lines[-1].endswith(','):
+    if lines[-1].endswith(","):
         lines[-1] = lines[-1][:-1]
-    
+
     lines.append("}")
-    
-    return '\n'.join(lines)
+
+    return "\n".join(lines)
 
 
 def _escape_bibtex(text: str) -> str:
     """
     Escape special characters for BibTeX.
-    
+
     Args:
         text: Text to escape
-    
+
     Returns:
         Escaped text safe for BibTeX
     """
     # Special characters that need escaping
     replacements = {
-        '&': r'\&',
-        '%': r'\%',
-        '$': r'\$',
-        '#': r'\#',
-        '_': r'\_',
+        "&": r"\&",
+        "%": r"\%",
+        "$": r"\$",
+        "#": r"\#",
+        "_": r"\_",
         # Don't escape braces - they're used for formatting
         # '{': r'\{',
         # '}': r'\}',
     }
-    
+
     for char, escaped in replacements.items():
         text = text.replace(char, escaped)
-    
-    return text
-
 
+    return text
 
 
-# EOF
+# EOF

scitex/io/_save_modules/_canvas.py

@@ -64,9 +64,7 @@ def save_canvas(
 
     # Validate extension
     if not str(spath).endswith(".canvas"):
-        raise ValueError(
-            f"Canvas path must end with .canvas extension: {spath}"
-        )
+        raise ValueError(f"Canvas path must end with .canvas extension: {spath}")
 
     # Handle different object types
     if isinstance(obj, (str, Path)):
@@ -143,6 +141,7 @@ def _export_canvas_figures(
     except Exception as e:
         # Log but don't fail save if export fails
         import sys
+
         print(f"Warning: Canvas export failed: {e}", file=sys.stderr)
 
 
@@ -152,9 +151,7 @@ def _copy_canvas_directory(source: Path, dest: Path) -> None:
         raise FileNotFoundError(f"Source canvas directory not found: {source}")
 
     if not (source / "canvas.json").exists():
-        raise ValueError(
-            f"Invalid canvas directory (missing canvas.json): {source}"
-        )
+        raise ValueError(f"Invalid canvas directory (missing canvas.json): {source}")
 
     # Remove destination if exists
     if dest.exists():
@@ -306,6 +303,7 @@ def _link_or_copy(source: Path, dest: Path, bundle: bool = False) -> None:
         try:
             # Use relative symlink for portability
             import os
+
             rel_path = os.path.relpath(source.resolve(), dest.parent.resolve())
             dest.symlink_to(rel_path)
         except (OSError, ValueError):
@@ -342,6 +340,7 @@ def _extract_embedded_panels(canvas_dict: Dict[str, Any], dest: Path) -> None:
         # Check for embedded JSON data (scitex type panels)
         if "panel_json" in panel:
             import json
+
             json_path = panel_dir / "panel.json"
             with open(json_path, "w") as f:
                 json.dump(panel["panel_json"], f, indent=2, default=str)

scitex/io/_save_modules/_catboost.py

@@ -7,14 +7,14 @@
 def _save_catboost(obj, spath):
     """
     Save a CatBoost model.
-    
+
     Parameters
     ----------
     obj : catboost.CatBoost
         The CatBoost model to save.
     spath : str
         Path where the CatBoost model file will be saved.
-        
+
     Returns
     -------
     None

scitex/io/_save_modules/_csv.py

@@ -65,9 +65,9 @@ def _save_csv(obj, spath: str, **kwargs) -> None:
                 pass
 
     # Set default index=False if not explicitly specified in kwargs
-    if 'index' not in kwargs:
-        kwargs['index'] = False
-        
+    if "index" not in kwargs:
+        kwargs["index"] = False
+
     # Save the file based on type
     if isinstance(obj, (pd.Series, pd.DataFrame)):
         obj.to_csv(spath, **kwargs)
@@ -86,7 +86,7 @@ def _save_csv(obj, spath: str, **kwargs) -> None:
         pd.DataFrame.from_dict(obj).to_csv(spath, **kwargs)
     else:
         # Check if it's a PaperCollection or similar object with to_dataframe method
-        if hasattr(obj, 'to_dataframe') and callable(getattr(obj, 'to_dataframe')):
+        if hasattr(obj, "to_dataframe") and callable(getattr(obj, "to_dataframe")):
             obj.to_dataframe().to_csv(spath, **kwargs)
         else:
             try:

scitex/io/_save_modules/_excel.py

@@ -5,6 +5,7 @@
 # ----------------------------------------
 from __future__ import annotations
 import os
+
 __FILE__ = __file__
 __DIR__ = os.path.dirname(__FILE__)
 # ----------------------------------------
@@ -93,18 +94,14 @@ def _apply_stats_styling(df, spath):
     wb = load_workbook(spath)
     ws = wb.active
 
-    fill_red = PatternFill(
-        start_color="FF6B6B", end_color="FF6B6B", fill_type="solid"
-    )
+    fill_red = PatternFill(start_color="FF6B6B", end_color="FF6B6B", fill_type="solid")
     fill_orange = PatternFill(
         start_color="FFA500", end_color="FFA500", fill_type="solid"
     )
     fill_yellow = PatternFill(
         start_color="FFE66D", end_color="FFE66D", fill_type="solid"
     )
-    fill_gray = PatternFill(
-        start_color="E8E8E8", end_color="E8E8E8", fill_type="solid"
-    )
+    fill_gray = PatternFill(start_color="E8E8E8", end_color="E8E8E8", fill_type="solid")
     font_bold = Font(bold=True)
 
     PVAL_VARIANTS = _generate_pval_variants()
@@ -200,9 +197,7 @@ def save_excel(obj, spath, style=True, **kwargs):
     elif isinstance(obj, np.ndarray):
         df = pd.DataFrame(obj)
     else:
-        raise ValueError(
-            f"Cannot save object of type {type(obj)} as Excel file"
-        )
+        raise ValueError(f"Cannot save object of type {type(obj)} as Excel file")
 
     # Save to Excel
     df.to_excel(spath, index=False, **kwargs)
@@ -211,4 +206,5 @@ def save_excel(obj, spath, style=True, **kwargs):
     if style and _is_statistical_results(df):
         _apply_stats_styling(df, spath)
 
+
 # EOF

scitex/io/_save_modules/_hdf5.py

@@ -4,6 +4,7 @@
 # File: /ssh:sp:/home/ywatanabe/proj/scitex_repo/src/scitex/io/_save_modules/_hdf5.py
 # ----------------------------------------
 import os
+
 __FILE__ = __file__
 __DIR__ = os.path.dirname(__FILE__)
 # ----------------------------------------
@@ -36,9 +37,7 @@ class SWMRFile:
             try:
                 if self.mode in ["w", "w-"]:
                     # For write mode, create new file
-                    self.file = h5py.File(
-                        self.filepath, self.mode, libver="latest"
-                    )
+                    self.file = h5py.File(self.filepath, self.mode, libver="latest")
                     if self.swmr:
                         self.file.swmr_mode = True
                     return self.file
@@ -64,16 +63,12 @@ class SWMRFile:
                             )
                         else:
                             # Normal append mode
-                            self.file = h5py.File(
-                                self.filepath, "r+", libver="latest"
-                            )
+                            self.file = h5py.File(self.filepath, "r+", libver="latest")
                             if self.swmr:
                                 self.file.swmr_mode = True
                     else:
                         # Create new file
-                        self.file = h5py.File(
-                            self.filepath, "w", libver="latest"
-                        )
+                        self.file = h5py.File(self.filepath, "w", libver="latest")
                         if self.swmr:
                             self.file.swmr_mode = True
                     return self.file
@@ -87,9 +82,7 @@ class SWMRFile:
 
             except (OSError, IOError) as e:
                 if "Resource temporarily unavailable" in str(e):
-                    time.sleep(
-                        0.1 * (2 ** min(5, (time.time() - start_time) / 10))
-                    )
+                    time.sleep(0.1 * (2 ** min(5, (time.time() - start_time) / 10)))
                 else:
                     raise
 
@@ -174,15 +167,11 @@ def _save_hdf5(
 
                             for part in parts[:-1]:
                                 if part:
-                                    current_group = (
-                                        current_group.require_group(part)
-                                    )
+                                    current_group = current_group.require_group(part)
 
                             final_key = parts[-1] if parts else ""
                             if final_key:
-                                target_group = current_group.create_group(
-                                    final_key
-                                )
+                                target_group = current_group.create_group(final_key)
                             else:
                                 target_group = current_group
                         else:
@@ -213,9 +202,7 @@ def _save_hdf5(
             else:
                 raise
 
-    raise RuntimeError(
-        f"Failed to save to {spath} after {max_retries} attempts"
-    )
+    raise RuntimeError(f"Failed to save to {spath} after {max_retries} attempts")
 
 
 def _save_dataset(group, name, data, compression, compression_opts, **kwargs):
@@ -271,4 +258,5 @@ def _save_dataset(group, name, data, compression, compression_opts, **kwargs):
     except Exception as e:
         print(f"Warning: Could not save dataset '{name}': {e}")
 
+
 # EOF

scitex/io/_save_modules/_html.py

@@ -18,7 +18,7 @@ import plotly
 
 def save_html(obj, spath, **kwargs):
     """Handle HTML file saving (primarily for Plotly figures).
-    
+
     Parameters
     ----------
     obj : plotly.graph_objs.Figure or str
@@ -27,22 +27,22 @@ def save_html(obj, spath, **kwargs):
         Path where HTML file will be saved
     **kwargs
         Additional keyword arguments passed to plotly.io.write_html()
-        
+
     Notes
     -----
     - Primarily designed for saving Plotly interactive figures
     - Can also save raw HTML strings
     """
-    if hasattr(obj, 'write_html'):
+    if hasattr(obj, "write_html"):
         # Plotly figure object
         obj.write_html(spath, **kwargs)
     elif isinstance(obj, str):
         # Raw HTML string
-        with open(spath, 'w') as f:
+        with open(spath, "w") as f:
             f.write(obj)
     else:
         # Try to convert to HTML using plotly
         plotly.io.write_html(obj, spath, **kwargs)
 
 
-# EOF
+# EOF

scitex/io/_save_modules/_image.py

@@ -23,8 +23,24 @@ def save_image(
     add_qr=False,
     qr_position="bottom-right",
     verbose=False,
+    save_stats=True,
     **kwargs,
 ):
+    # Determine if spath is a file-like object (e.g., BytesIO)
+    is_file_like = not isinstance(spath, str)
+
+    # Get format from file extension or kwargs
+    if is_file_like:
+        fmt = kwargs.get('format', '').lower()
+    else:
+        # Get extension without the leading dot
+        fmt = spath.lower().rsplit('.', 1)[-1] if '.' in spath else ''
+
+    # Auto-save stats BEFORE saving (obj may be deleted during save)
+    # Only for file paths, not file-like objects
+    if save_stats and not is_file_like:
+        _save_stats_from_figure(obj, spath, verbose=verbose)
+
     # Add URL to metadata if not present
     if metadata is not None:
         if verbose:
@@ -55,7 +71,7 @@ def save_image(
                 warnings.warn(f"Failed to add QR code: {e}")
 
     # png
-    if spath.endswith(".png"):
+    if fmt == 'png':
         # plotly
         if isinstance(obj, plotly.graph_objs.Figure):
             obj.write_image(file=spath, format="png")
@@ -71,7 +87,7 @@ def save_image(
         del obj
 
     # tiff
-    elif spath.endswith(".tiff") or spath.endswith(".tif"):
+    elif fmt in ('tiff', 'tif'):
         # PIL image
         if isinstance(obj, Image.Image):
             obj.save(spath)
@@ -88,7 +104,7 @@ def save_image(
         del obj
 
     # jpeg
-    elif spath.endswith(".jpeg") or spath.endswith(".jpg"):
+    elif fmt in ('jpeg', 'jpg'):
         buf = _io.BytesIO()
 
         # plotly
@@ -125,7 +141,7 @@ def save_image(
         del obj
 
     # GIF
-    elif spath.endswith(".gif"):
+    elif fmt == 'gif':
         # PIL image
         if isinstance(obj, Image.Image):
             obj.save(spath, save_all=True)
@@ -153,7 +169,7 @@ def save_image(
         del obj
 
     # SVG
-    elif spath.endswith(".svg"):
+    elif fmt == 'svg':
         # Plotly
         if isinstance(obj, plotly.graph_objs.Figure):
             obj.write_image(file=spath, format="svg")
@@ -168,7 +184,7 @@ def save_image(
         del obj
 
     # PDF
-    elif spath.endswith(".pdf"):
+    elif fmt == 'pdf':
         # Plotly
         if isinstance(obj, plotly.graph_objs.Figure):
             obj.write_image(file=spath, format="pdf")
@@ -191,8 +207,8 @@ def save_image(
                 obj.figure.savefig(spath, **save_kwargs)
         del obj
 
-    # Embed metadata if provided
-    if metadata is not None:
+    # Embed metadata if provided (only for file paths, not file-like objects)
+    if metadata is not None and not is_file_like:
         from .._metadata import embed_metadata
 
         try:
@@ -204,4 +220,85 @@ def save_image(
 
             warnings.warn(f"Failed to embed metadata: {e}")
 
+def _save_stats_from_figure(obj, spath, verbose=False):
+    """
+    Extract and save statistical annotations from a figure.
+
+    Saves to {basename}_stats.csv if stats are found.
+    """
+    try:
+        from scitex.bridge import extract_stats_from_axes
+    except ImportError:
+        return  # Bridge not available
+
+    # Get the matplotlib figure
+    fig = None
+    if hasattr(obj, "savefig"):
+        fig = obj
+    elif hasattr(obj, "figure") and hasattr(obj.figure, "savefig"):
+        fig = obj.figure
+    elif hasattr(obj, "_fig_mpl"):
+        fig = obj._fig_mpl
+
+    if fig is None:
+        return
+
+    # Extract stats from all axes
+    all_stats = []
+    try:
+        for ax in fig.axes:
+            stats = extract_stats_from_axes(ax)
+            all_stats.extend(stats)
+    except Exception:
+        return  # Silently fail if extraction fails
+
+    if not all_stats:
+        return  # No stats to save
+
+    # Build stats dataframe
+    try:
+        import pandas as pd
+
+        stats_data = []
+        for stat in all_stats:
+            row = {
+                "test_type": stat.test_type,
+                "statistic_name": stat.statistic.get("name", ""),
+                "statistic_value": stat.statistic.get("value", ""),
+                "p_value": stat.p_value,
+                "stars": stat.stars,
+            }
+            # Add effect size if available
+            if stat.effect_size:
+                row["effect_size_name"] = stat.effect_size.get("name", "")
+                row["effect_size_value"] = stat.effect_size.get("value", "")
+            # Add CI if available
+            if stat.ci_95:
+                row["ci_95_lower"] = stat.ci_95[0]
+                row["ci_95_upper"] = stat.ci_95[1]
+            # Add sample/group info if available (for consistent naming with plot CSV)
+            if stat.samples:
+                for group_name, group_info in stat.samples.items():
+                    if isinstance(group_info, dict):
+                        row[f"n_{group_name}"] = group_info.get("n")
+                        row[f"mean_{group_name}"] = group_info.get("mean")
+                        row[f"std_{group_name}"] = group_info.get("std")
+            stats_data.append(row)
+
+        stats_df = pd.DataFrame(stats_data)
+
+        # Save to {basename}_stats.csv
+        import os
+        base, ext = os.path.splitext(spath)
+        stats_path = f"{base}_stats.csv"
+        stats_df.to_csv(stats_path, index=False)
+
+        if verbose:
+            logger.info(f"  • Auto-saved stats to: {stats_path}")
+
+    except Exception as e:
+        import warnings
+        warnings.warn(f"Failed to auto-save stats: {e}")
+
+
 # EOF

scitex/io/_save_modules/_joblib.py

@@ -9,14 +9,14 @@ import joblib
 def _save_joblib(obj, spath):
     """
     Save an object using joblib serialization.
-    
+
     Parameters
     ----------
     obj : Any
         Object to serialize.
     spath : str
         Path where the joblib file will be saved.
-        
+
     Returns
     -------
     None