code-to-txt 0.1.0__py3-none-any.whl

code_to_txt/__init__.py

@@ -0,0 +1,3 @@
+from .code_to_txt import CodeToText
+
+__all__ = ["CodeToText"]

code_to_txt/cli.py

@@ -0,0 +1,96 @@
+import click
+from pathlib import Path
+from .code_to_txt import CodeToText
+
+
+@click.command()
+@click.argument("path", type=click.Path(exists=True), default=".")
+@click.option(
+    "-o",
+    "--output",
+    default="code_output.txt",
+    help="Output file path",
+    type=click.Path(),
+)
+@click.option(
+    "-e",
+    "--extensions",
+    multiple=True,
+    help="File extensions to include (e.g., .py .js). Can be specified multiple times.",
+)
+@click.option(
+    "-x",
+    "--exclude",
+    multiple=True,
+    help="Patterns to exclude (gitignore style). Can be specified multiple times.",
+)
+@click.option(
+    "--no-gitignore",
+    is_flag=True,
+    help="Don't respect .gitignore files",
+)
+@click.option(
+    "--no-tree",
+    is_flag=True,
+    help="Don't include directory tree in output",
+)
+@click.option(
+    "--separator",
+    default="=" * 80,
+    help="Separator between files",
+)
+def main(path, output, extensions, exclude, no_gitignore, no_tree, separator):
+    """
+    Convert code files to a single text file for easy LLM consumption.
+
+    PATH: Directory to scan (default: current directory)
+
+    Examples:
+
+        # Convert all code files in current directory
+        code-to-txt
+
+        # Convert specific directory to custom output
+        code-to-txt ./my-project -o project.txt
+
+        # Only include Python and JavaScript files
+        code-to-txt -e .py -e .js
+
+        # Exclude test files
+        code-to-txt -x "tests/*" -x "*.test.js"
+
+        # Don't use .gitignore and don't show tree
+        code-to-txt --no-gitignore --no-tree
+    """
+    click.echo(f"Scanning: {path}")
+
+    include_extensions = set(extensions) if extensions else None
+
+    codetotxt = CodeToText(
+        root_path=path,
+        output_file=output,
+        include_extensions=include_extensions,
+        exclude_patterns=list(exclude),
+        gitignore=not no_gitignore,
+    )
+
+    try:
+        num_files = codetotxt.convert(
+            add_tree=not no_tree,
+            separator=separator,
+        )
+
+        output_path = Path(output).resolve()
+        click.echo(f"Successfully processed {num_files} files")
+        click.echo(f"Output saved to: {output_path}")
+
+        size_kb = output_path.stat().st_size / 1024
+        click.echo(f"File size: {size_kb:.2f} KB")
+
+    except Exception as e:
+        click.echo(f"Error: {e}", err=True)
+        raise click.Abort()
+
+
+if __name__ == "__main__":
+    main()

code_to_txt/code_to_txt.py

@@ -0,0 +1,180 @@
+import os
+from pathlib import Path
+from typing import List, Optional, Set
+import pathspec
+
+
+class CodeToText:
+    DEFAULT_IGNORE = {
+        "__pycache__",
+        "*.pyc",
+        "*.pyo",
+        "*.pyd"
+    }
+
+    DEFAULT_EXTENSIONS = {
+        ".py", ".js", ".ts", ".jsx", ".tsx", ".java", ".c", ".cpp", ".h"
+    }
+
+    def __init__(
+            self,
+            root_path: str,
+            output_file: str = "output.txt",
+            include_extensions: Optional[Set[str]] = None,
+            exclude_patterns: Optional[List[str]] = None,
+            gitignore: bool = True,
+    ):
+        """
+        Initialize the instance of CodeToText.
+
+        Args:
+            root_path: Root directory to scan
+            output_file: Output file path
+            include_extensions: Set of file extensions to include (with dots)
+            exclude_patterns: List of patterns to exclude (gitignore style)
+            gitignore: Whether to respect .gitignore files
+        """
+        self.root_path = Path(root_path).resolve()
+        self.output_file = output_file
+        self.include_extensions = include_extensions or self.DEFAULT_EXTENSIONS
+        self.exclude_patterns = exclude_patterns or []
+        self.gitignore = gitignore
+        self.spec = None
+
+        if self.gitignore:
+            self._load_gitignore()
+
+    def _load_gitignore(self):
+        """Load .gitignore patterns if present."""
+        gitignore_path = self.root_path / ".gitignore"
+        patterns = list(self.DEFAULT_IGNORE)
+
+        if gitignore_path.exists():
+            with open(gitignore_path, "r") as f:
+                for line in f:
+                    line = line.strip()
+                    if line and not line.startswith("#"):
+                        patterns.append(line)
+
+        patterns.extend(self.exclude_patterns)
+        self.spec = pathspec.PathSpec.from_lines("gitignore", patterns)
+
+    def _should_include_file(self, file_path: Path) -> bool:
+        """Check if a file should be included."""
+        # Check extension
+        if file_path.suffix not in self.include_extensions:
+            return False
+
+        if self.spec:
+            relative_path = file_path.relative_to(self.root_path)
+            if self.spec.match_file(str(relative_path)):
+                return False
+
+        return True
+
+    def _get_files(self) -> List[Path]:
+        """Get all files to process."""
+        files = []
+        for root, dirs, filenames in os.walk(self.root_path):
+            root_path = Path(root)
+
+            if self.spec:
+                relative_root = root_path.relative_to(self.root_path)
+                dirs[:] = [
+                    d for d in dirs
+                    if not self.spec.match_file(str(relative_root / d))
+                ]
+
+            for filename in filenames:
+                file_path = root_path / filename
+                if self._should_include_file(file_path):
+                    files.append(file_path)
+
+        return sorted(files)
+
+    def convert(self, add_tree: bool = True, separator: str = "=" * 80) -> int:
+        """
+        Convert files to single text file.
+
+        Args:
+            add_tree: Whether to add directory tree at the beginning
+            separator: Separator between files
+
+        Returns:
+            Number of files processed
+        """
+        files = self._get_files()
+
+        with open(self.output_file, "w", encoding="utf-8") as out:
+            out.write(f"Code Export from: {self.root_path}\n")
+            out.write(f"Total files: {len(files)}\n")
+            out.write(f"{separator}\n\n")
+
+            if add_tree:
+                out.write("DIRECTORY TREE:\n")
+                out.write(separator + "\n")
+                out.write(self._generate_tree())
+                out.write(f"\n{separator}\n\n")
+
+            for i, file_path in enumerate(files, 1):
+                relative_path = file_path.relative_to(self.root_path)
+
+                out.write(f"FILE {i}/{len(files)}: {relative_path}\n")
+                out.write(separator + "\n")
+
+                try:
+                    with open(file_path, "r", encoding="utf-8") as f:
+                        content = f.read()
+                    out.write(content)
+                except UnicodeDecodeError:
+                    out.write(f"[Binary file - skipped]\n")
+                except Exception as e:
+                    out.write(f"[Error reading file: {e}]\n")
+
+                out.write(f"\n{separator}\n\n")
+
+        return len(files)
+
+    def _generate_tree(self) -> str:
+        """Generate a directory tree representation."""
+        tree_lines = []
+        files = self._get_files()
+
+        dir_structure = {}
+        for file_path in files:
+            relative_path = file_path.relative_to(self.root_path)
+            parts = relative_path.parts
+
+            current = dir_structure
+            for part in parts[:-1]:
+                if part not in current:
+                    current[part] = {}
+                current = current[part]
+
+            if "__files__" not in current:
+                current["__files__"] = []
+            current["__files__"].append(parts[-1])
+
+        def print_tree(structure, prefix="", is_last=True):
+            items = []
+            for key in sorted(structure.keys()):
+                if key != "__files__":
+                    items.append((key, True))  # directory
+
+            if "__files__" in structure:
+                for file in sorted(structure["__files__"]):
+                    items.append((file, False))  # file
+
+            for i, (name, is_dir) in enumerate(items):
+                is_last_item = i == len(items) - 1
+                connector = "└── " if is_last_item else "├── "
+                tree_lines.append(f"{prefix}{connector}{name}{'/' if is_dir else ''}")
+
+                if is_dir:
+                    extension = "    " if is_last_item else "│   "
+                    print_tree(structure[name], prefix + extension, is_last_item)
+
+        tree_lines.append(f"{self.root_path.name}/")
+        print_tree(dir_structure)
+
+        return "\n".join(tree_lines)

code_to_txt-0.1.0.dist-info/METADATA

@@ -0,0 +1,24 @@
+Metadata-Version: 2.4
+Name: code-to-txt
+Version: 0.1.0
+Summary: Convert code files to a single text file for LLM consumption
+License: MIT
+Author: Andrii Sonsiadlo
+Author-email: andrii.sonsiadlo@gmail.com
+Requires-Python: >=3.10
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Requires-Dist: click (>=8.3.1,<9.0.0)
+Requires-Dist: gitpython (>=3.1.46,<4.0.0)
+Requires-Dist: pathspec (>=1.0.4,<2.0.0)
+Description-Content-Type: text/markdown
+
+# CodeToTxt
+
+A small Python package to convert code to text.
+

code_to_txt-0.1.0.dist-info/RECORD

@@ -0,0 +1,7 @@
+code_to_txt/__init__.py,sha256=0_iks7Uz24B1pc-Na1n8C97vgMms6haaFNqIRkpq_Cg,62
+code_to_txt/cli.py,sha256=EEm87GRTgk1YorSfoLl2IR2NUjG0f_fkgEQ5YnkxpQ8,2409
+code_to_txt/code_to_txt.py,sha256=HKXiTXMUxM612f7SVHcoxsra6uK6SVV6_-UY49jtR2w,6089
+code_to_txt-0.1.0.dist-info/METADATA,sha256=QEbEcYTZonOO00fLkq3EFuIWBgaGWQLsxnGV3UY1O7I,807
+code_to_txt-0.1.0.dist-info/WHEEL,sha256=3ny-bZhpXrU6vSQ1UPG34FoxZBp3lVcvK0LkgUz6VLk,88
+code_to_txt-0.1.0.dist-info/entry_points.txt,sha256=jPT0g_nryiuAd0E496deFZAhdscNLXiUmUdD3KGN3iA,52
+code_to_txt-0.1.0.dist-info/RECORD,,

code_to_txt-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: poetry-core 2.3.0
+Root-Is-Purelib: true
+Tag: py3-none-any

code_to_txt-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+code-to-txt=code_to_txt.cli:main
+