edsl 0.1.15__py3-none-any.whl → 0.1.40__py3-none-any.whl

edsl/results/DatasetTree.py

@@ -0,0 +1,295 @@
+from typing import Dict, List, Any, Optional, List
+
+
+def is_hashable(v):
+    try:
+        hash(v)
+        return True
+    except TypeError:
+        return False
+
+
+class TreeNode:
+    def __init__(self, key=None, value=None):
+        self.key = key
+        self.value = value
+        self.children = {}
+
+
+class Tree:
+    def __init__(self, data: "Dataset", node_order: Optional[List[str]] = None):
+        d = {}
+        for entry in data:
+            d.update(entry)
+        self.data = d
+        self.root = None
+
+        self.node_order = node_order
+
+        self.construct_tree(node_order)
+
+    def unique_values_by_keys(self) -> dict:
+        unique_values = {}
+        for key, raw_values in self.data.items():
+            values = [v if is_hashable(v) else str(v) for v in raw_values]
+            unique_values[key] = list(set(values))
+        return unique_values
+
+    def construct_tree(self, node_order: Optional[List[str]] = None):
+        # Validate node_order
+        if node_order is None:
+            unique_values = self.unique_values_by_keys()
+            # Sort keys by number of unique values
+            node_order = sorted(
+                unique_values, key=lambda k: len(unique_values[k]), reverse=True
+            )
+        else:
+            if not set(node_order).issubset(set(self.data.keys())):
+                invalid_keys = set(node_order) - set(self.data.keys())
+                raise ValueError(f"Invalid keys in node_order: {invalid_keys}")
+
+        self.root = TreeNode()
+
+        for i in range(len(self.data[node_order[0]])):
+            current = self.root
+            for level in node_order[:-1]:
+                value = self.data[level][i]
+                if not is_hashable(value):
+                    value = str(value)
+                if value not in current.children:
+                    current.children[value] = TreeNode(key=level, value=value)
+                current = current.children[value]
+
+            leaf_key = node_order[-1]
+            leaf_value = self.data[leaf_key][i]
+            if not is_hashable(leaf_value):
+                leaf_value = str(leaf_value)
+            if leaf_value not in current.children:
+                current.children[leaf_value] = TreeNode(key=leaf_key, value=leaf_value)
+
+    def __repr__(self):
+        if self.node_order is not None:
+            return f"Tree(Dataset({self.data}), node_order={self.node_order})"
+        else:
+            return f"Tree(Dataset({self.data}))"
+
+    def print_tree(
+        self, node: Optional[TreeNode] = None, level: int = 0, print_keys: bool = False
+    ):
+        if node is None:
+            node = self.root
+            if node is None:
+                print("Tree has not been constructed yet.")
+                return
+
+        if node.value is not None:
+            if print_keys and node.key is not None:
+                print("  " * level + f"{node.key}: {node.value}")
+            else:
+                print("  " * level + str(node.value))
+        for child in node.children.values():
+            self.print_tree(child, level + 1, print_keys)
+
+    def to_docx(self, filename: Optional[str] = None):
+        if filename is None:
+            filename = "tree_structure.docx"
+
+        from docx import Document
+        from docx.shared import Inches, Pt
+        from docx.enum.text import WD_ALIGN_PARAGRAPH
+        from docx.enum.style import WD_STYLE_TYPE
+
+        doc = Document()
+
+        # Create styles for headings
+        for i in range(1, 10):  # Up to 9 levels of headings
+            style_name = f"Heading {i}"
+            if style_name not in doc.styles:
+                doc.styles.add_style(style_name, WD_STYLE_TYPE.PARAGRAPH)
+
+        # Get or create the 'Body Text' style
+        if "Body Text" not in doc.styles:
+            body_style = doc.styles.add_style("Body Text", WD_STYLE_TYPE.PARAGRAPH)
+        else:
+            body_style = doc.styles["Body Text"]
+
+        body_style.font.size = Pt(11)
+
+        self._add_to_docx(doc, self.root, 0)
+        import base64
+        from io import BytesIO
+        import base64
+
+        # Save document to bytes buffer
+        doc_buffer = BytesIO()
+        doc.save(doc_buffer)
+        doc_buffer.seek(0)
+
+        base64_string = base64.b64encode(doc_buffer.getvalue()).decode("utf-8")
+        from edsl.scenarios.FileStore import FileStore
+
+        # Create and return FileStore instance
+        return FileStore(
+            path="tree_structure.docx",  # Default name
+            mime_type="application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+            binary=True,
+            suffix="docx",
+            base64_string=base64_string,
+        )
+        # doc.save(filename)
+        # from edsl.utilities.utilities import file_notice
+        # file_notice(filename)
+
+    def _repr_html_(self):
+        """Returns an interactive HTML representation of the tree with collapsible sections."""
+
+        # Generate a unique ID for this tree instance
+        import uuid
+
+        tree_id = f"tree_{uuid.uuid4().hex[:8]}"
+
+        styles = f"""
+        <div class="{tree_id}">
+            <style>
+                .{tree_id} details {{
+                    margin-left: 20px;
+                }}
+                .{tree_id} summary {{
+                    cursor: pointer;
+                    margin: 2px 0;
+                }}
+                .{tree_id} .value {{
+                    font-family: monospace;
+                    background: #f5f5f5;
+                    padding: 2px 6px;
+                    border-radius: 3px;
+                    margin: 1px 0;
+                }}
+                .{tree_id} .key {{
+                    color: #666;
+                    font-style: italic;
+                }}
+            </style>
+        """
+
+        def node_to_html(node, level=0, print_keys=True):
+            if node is None:
+                return "Tree has not been constructed yet."
+
+            html = []
+
+            if node.value is not None:
+                # Create the node content
+                content = []
+                if print_keys and node.key is not None:
+                    content.append(f'<span class="key">{node.key}: </span>')
+                content.append(f'<span class="value">{node.value}</span>')
+                content_html = "".join(content)
+
+                if node.children:
+                    # Node with children
+                    html.append(f'<details {"open" if level < 1 else ""}>')
+                    html.append(f"<summary>{content_html}</summary>")
+                    for child in node.children.values():
+                        html.append(node_to_html(child, level + 1, print_keys))
+                    html.append("</details>")
+                else:
+                    # Leaf node
+                    html.append(f"<div>{content_html}</div>")
+            else:
+                # Root node with no value
+                if node.children:
+                    for child in node.children.values():
+                        html.append(node_to_html(child, level, print_keys))
+
+            return "\n".join(html)
+
+        tree_html = node_to_html(self.root)
+        return f"{styles}{tree_html}</div>"
+
+    # def _repr_html_(self):
+    #     """Returns an HTML representation of the tree, following the same logic as print_tree."""
+    #     styles = """
+    #     <style>
+    #         .tree-container {
+    #             font-family: monospace;
+    #             white-space: pre;
+    #             margin: 10px;
+    #         }
+    #     </style>
+    #     """
+
+    #     def node_to_html(node, level=0, print_keys=False):
+    #         if node is None:
+    #             node = self.root
+    #             if node is None:
+    #                 return "Tree has not been constructed yet."
+
+    #         html = []
+    #         if node.value is not None:
+    #             indent = "&nbsp;" * 2 * level  # Using &nbsp; for HTML spaces
+    #             if print_keys and node.key is not None:
+    #                 html.append(f"{indent}{node.key}: {node.value}<br>")
+    #             else:
+    #                 html.append(f"{indent}{node.value}<br>")
+
+    #         for child in node.children.values():
+    #             html.append(node_to_html(child, level + 1, print_keys))
+
+    #         return "".join(html)
+
+    #     tree_html = node_to_html(self.root)
+    #     return f'<div class="tree-container">{tree_html}</div>{styles}'
+
+    def _add_to_docx(self, doc, node: TreeNode, level: int):
+        if node.value is not None:
+            if level == 0:
+                doc.add_heading(str(node.value), level=level + 1)
+            elif node.children:  # If the node has children, it's not the last level
+                para = doc.add_paragraph(str(node.value))
+                para.style = f"Heading {level+1}"
+            else:  # If the node has no children, it's the last level (body text)
+                para = doc.add_paragraph(str(node.value))
+                para.style = "Body Text"
+
+        # Process child nodes (moved outside the if block)
+        for child in node.children.values():
+            self._add_to_docx(doc, child, level + 1)
+
+
+# Example usage (commented out)
+"""
+from edsl.results.Dataset import Dataset
+
+data = Dataset(
+    [
+        {"continent": ["North America", "Asia", "Europe", "North America", "Asia"]},
+        {"country": ["US", "China", "France", "Canada", "Japan"]},
+        {"city": ["New York", "Beijing", "Paris", "Toronto", "Tokyo"]},
+        {"population": [8419000, 21540000, 2161000, 2930000, 13960000]},
+    ]
+)
+
+tree = Tree(data)
+
+try:
+    tree.construct_tree(["continent", "country", "city", "population"])
+    print("Tree without key names:")
+    tree.print_tree()
+    print("\nTree with key names:")
+    tree.print_tree(print_keys=True)
+except ValueError as e:
+    print(f"Error: {e}")
+
+# Demonstrating validation
+try:
+    tree.construct_tree(["continent", "country", "invalid_key"])
+except ValueError as e:
+    print(f"\nValidation Error: {e}")
+
+tree = Tree(data)
+tree.construct_tree(["continent", "country", "city", "population"])
+tree.print_tree(print_keys=True)
+tree.to_docx("tree_structure.docx")
+print("DocX file 'tree_structure.docx' has been created.")
+"""

edsl/results/MarkdownToDocx.py

@@ -0,0 +1,122 @@
+from typing import Optional
+import subprocess
+import os
+from pathlib import Path
+import tempfile
+
+
+class MarkdownToDocx:
+    def __init__(self, markdown_content: str, filename: Optional[str] = None):
+        """
+        Initialize the converter with markdown content.
+
+        Args:
+            markdown_content (str): The markdown content to be converted
+        """
+        self.markdown_content = markdown_content
+        self.filename = filename
+        self._check_pandoc()
+
+    def _check_pandoc(self):
+        """Check if pandoc is installed and accessible."""
+        try:
+            subprocess.run(["pandoc", "--version"], capture_output=True, check=True)
+        except (subprocess.CalledProcessError, FileNotFoundError):
+            raise RuntimeError(
+                "Pandoc is not installed or not found in PATH. "
+                "Please install pandoc before using this converter."
+            )
+
+    def convert(self, output_path: str, **options) -> bool:
+        """
+        Convert the markdown content to DOCX.
+
+        Args:
+            output_path (str): Path where the DOCX should be saved
+            **options: Additional conversion options
+                reference_doc (str): Path to reference docx for styling
+                toc (bool): Include table of contents (default: False)
+                number_sections (bool): Number sections (default: False)
+                highlight_style (str): Code highlighting style (default: "tango")
+
+        Returns:
+            bool: True if conversion was successful, False otherwise
+        """
+        # Ensure output directory exists
+        output_path = Path(output_path)
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+
+        # Build pandoc command
+        cmd = ["pandoc", "-f", "markdown", "-t", "docx", "-o", str(output_path)]
+
+        # Add reference doc if provided
+        if "reference_doc" in options:
+            ref_doc = Path(options["reference_doc"])
+            if ref_doc.exists():
+                cmd.extend(["--reference-doc", str(ref_doc)])
+            else:
+                print(f"Warning: Reference document {ref_doc} not found")
+
+        # Add optional parameters
+        if options.get("toc", False):
+            cmd.append("--toc")
+
+        if options.get("number_sections", False):
+            cmd.append("--number-sections")
+
+        if "highlight_style" in options:
+            cmd.extend(["--highlight-style", options["highlight_style"]])
+
+        try:
+            # Run pandoc command
+            result = subprocess.run(
+                cmd,
+                input=self.markdown_content,
+                text=True,
+                capture_output=True,
+                check=True,
+            )
+            return True
+        except subprocess.CalledProcessError as e:
+            print(f"Error converting markdown to DOCX: {e.stderr}")
+            return False
+
+    def preview(self) -> str:
+        """
+        Generate a temporary DOCX and return its path.
+
+        Returns:
+            str: Path to the temporary DOCX file
+        """
+        temp_dir = tempfile.mkdtemp()
+        if self.filename:
+            temp_docx = os.path.join(temp_dir, self.filename)
+        else:
+            temp_docx = os.path.join(temp_dir, "preview.docx")
+
+        if self.convert(temp_docx):
+            from edsl.scenarios.FileStore import FileStore
+
+            return FileStore(path=temp_docx)
+
+        return None
+
+    def create_template(self, output_path: str) -> bool:
+        """
+        Create a reference DOCX template that can be modified for styling.
+
+        Args:
+            output_path (str): Path where the template should be saved
+
+        Returns:
+            bool: True if template was created successfully, False otherwise
+        """
+        try:
+            cmd = ["pandoc", "--print-default-data-file", "reference.docx"]
+
+            with open(output_path, "wb") as f:
+                result = subprocess.run(cmd, stdout=f, check=True)
+            return True
+        except subprocess.CalledProcessError as e:
+            print(f"Error creating template: {e.stderr}")
+            return False

edsl/results/MarkdownToPDF.py

@@ -0,0 +1,111 @@
+from typing import Optional
+import subprocess
+import os
+from pathlib import Path
+import tempfile
+
+
+class MarkdownToPDF:
+    def __init__(self, markdown_content: str, filename: Optional[str] = None):
+        """
+        Initialize the converter with markdown content.
+
+        Args:
+            markdown_content (str): The markdown content to be converted
+        """
+        self.markdown_content = markdown_content
+        self.filename = filename
+        self._check_pandoc()
+        # self.convert()
+
+    def _check_pandoc(self):
+        """Check if pandoc is installed and accessible."""
+        try:
+            subprocess.run(["pandoc", "--version"], capture_output=True, check=True)
+        except (subprocess.CalledProcessError, FileNotFoundError):
+            raise RuntimeError(
+                "Pandoc is not installed or not found in PATH. "
+                "Please install pandoc before using this converter."
+            )
+
+    def convert(self, output_path: str, **options) -> bool:
+        """
+        Convert the markdown content to PDF.
+
+        Args:
+            output_path (str): Path where the PDF should be saved
+            **options: Additional conversion options
+                margin (str): Page margin (default: "1in")
+                font_size (str): Font size (default: "12pt")
+                toc (bool): Include table of contents (default: False)
+                number_sections (bool): Number sections (default: False)
+                highlight_style (str): Code highlighting style (default: "tango")
+
+        Returns:
+            bool: True if conversion was successful, False otherwise
+        """
+        # Ensure output directory exists
+        output_path = Path(output_path)
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+
+        # Build pandoc command with default options
+        cmd = [
+            "pandoc",
+            "-f",
+            "markdown",
+            "-o",
+            str(output_path),
+            "--pdf-engine=xelatex",
+            "--variable",
+            f'geometry:margin={options.get("margin", "1in")}',
+            "--variable",
+            f'fontsize={options.get("font_size", "12pt")}',
+        ]
+
+        # Add font only if specifically provided
+        if "font" in options:
+            cmd.extend(["--variable", f'mainfont={options["font"]}'])
+
+        # Add optional parameters
+        if options.get("toc", False):
+            cmd.append("--toc")
+
+        if options.get("number_sections", False):
+            cmd.append("--number-sections")
+
+        if "highlight_style" in options:
+            cmd.extend(["--highlight-style", options["highlight_style"]])
+
+        try:
+            # Run pandoc command
+            result = subprocess.run(
+                cmd,
+                input=self.markdown_content,
+                text=True,
+                capture_output=True,
+                check=True,
+            )
+            return True
+        except subprocess.CalledProcessError as e:
+            print(f"Error converting markdown to PDF: {e.stderr}")
+            return False
+
+    def preview(self) -> str:
+        """
+        Generate a temporary PDF and return its path.
+
+        Returns:
+            str: Path to the temporary PDF file
+        """
+        temp_dir = tempfile.mkdtemp()
+        if self.filename:
+            temp_pdf = os.path.join(temp_dir, f"{self.filename}.pdf")
+        else:
+            temp_pdf = os.path.join(temp_dir, "preview.pdf")
+
+        if self.convert(temp_pdf):
+            from edsl.scenarios.FileStore import FileStore
+
+            return FileStore(temp_pdf)
+
+        return None