docconvert-local 2.0.0__tar.gz

docconvert_local-2.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 DocConvert
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

docconvert_local-2.0.0/PKG-INFO

@@ -0,0 +1,253 @@
+Metadata-Version: 2.4
+Name: docconvert-local
+Version: 2.0.0
+Summary: Clean Excel & Word → Markdown for RAG pipelines. Works 100% offline — no API keys, no cloud upload.
+Author: DocConvert Contributors
+License: MIT
+Project-URL: Homepage, https://github.com/leop017/DocConvert
+Project-URL: BugTracker, https://github.com/leop017/DocConvert/issues
+Keywords: document,conversion,excel,word,markdown
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Office/Business :: Office Suites
+Classifier: Topic :: Text Processing :: Markup :: HTML
+Classifier: Topic :: Text Processing :: Markup :: Markdown
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: python-docx>=1.1.0
+Requires-Dist: openpyxl>=3.1.0
+Requires-Dist: pandas>=2.0.0
+Requires-Dist: mammoth>=1.6.0
+Requires-Dist: beautifulsoup4>=4.12.0
+Requires-Dist: markdownify>=0.11.0
+Requires-Dist: xlrd>=2.0.0
+Provides-Extra: test
+Requires-Dist: mypy>=1.0; extra == "test"
+Requires-Dist: xlwt; extra == "test"
+Provides-Extra: doc
+Requires-Dist: textract>=1.6.0; platform_system != "Windows" and extra == "doc"
+Provides-Extra: build
+Requires-Dist: pyinstaller>=6.0; extra == "build"
+Provides-Extra: all
+Requires-Dist: docconvert-local[build,doc,test]; extra == "all"
+Dynamic: license-file
+
+# DocConvert
+
+> Clean Markdown & structured data from Excel & Word — built for RAG pipelines and LLM workflows.
+> Works 100% offline. No API keys. No data leaves your machine.
+
+[![PyPI version](https://img.shields.io/pypi/v/docconvert-local.svg)](https://pypi.org/project/docconvert-local/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/)
+[![Build](https://github.com/leop017/DocConvert/actions/workflows/ci.yml/badge.svg)](https://github.com/leop017/DocConvert/actions)
+[![Downloads](https://static.pepy.tech/badge/docconvert-local/month)](https://pepy.tech/project/docconvert-local)
+[![Stars](https://img.shields.io/github/stars/leop017/DocConvert?style=social)](https://github.com/leop017/DocConvert/stargazers)
+
+## Why DocConvert
+
+Most document-to-Markdown tools convert the file — they don't **clean** it. Raw outputs are full of page numbers, duplicate headers, and whitespace noise that eats your context window and dilutes retrieval quality.
+
+DocConvert was built for people who feed documents into LLMs and need every token to count.
+
+| Feature                              |       DocConvert      | MarkItDown | Pandas + python-docx |
+| ------------------------------------ | :-------------------: | :--------: | :------------------: |
+| Legacy `.doc` support                |           ✅           |      ❌     |           ❌          |
+| Excel merged cells (rowspan/colspan) |           ✅           |     ⚠️     |        Manual        |
+| Configurable cleaning pipeline       | ✅ 4 rules, toggle any |      ❌     |           ❌          |
+| Batch + specific sheet selection     |           ✅           |      ✅     |           ❌          |
+| Desktop GUI (no terminal needed)     |           ✅           |      ❌     |           ❌          |
+| PDF / PPT / audio support            |           ❌           |      ✅     |           ❌          |
+| MCP server / Claude integration      |           ❌           |      ✅     |           ❌          |
+| 100% offline, no cloud dependency    |           ✅           |      ✅     |           ✅          |
+
+**Choose DocConvert if:** you work with Excel/Word documents inside an organization, need legacy `.doc` support, or want a configurable cleaning pipeline before feeding docs into a RAG system.
+
+**Choose MarkItDown if:** you need PDF, PPT, images, or audio conversion, or want MCP/Claude Desktop integration out of the box.
+
+## Use Cases
+
+* **RAG ingestion** — clean Excel financial reports and Word contracts into Markdown ready for embedding
+
+* **LLM context prep** — strip page numbers, duplicates, and noise before chunking
+
+* **Offline compliance** — convert sensitive documents without uploading to any cloud service
+
+* **Batch automation** — convert entire folders of reports into a structured directory
+
+## Installation
+
+```bash
+pip install docconvert-local
+```
+
+Optional extras:
+
+```bash
+# Legacy .doc support (Linux / macOS only)
+pip install docconvert-local[doc]
+
+# Full feature set including build tools
+pip install docconvert-local[all]
+```
+
+## Quick Start
+
+### GUI (interactive)
+
+```bash
+python main.py
+```
+
+### CLI (batch / scripting)
+
+```bash
+# Single file → clean Markdown
+python main.py convert input.xlsx --format md
+
+# Batch convert with enhanced cleaning (recommended for RAG)
+python main.py convert input.docx --format md --enhanced
+
+# Multiple files → HTML into output/
+python main.py convert file1.xlsx file2.docx --format html -o ./output
+
+# Pick specific Excel sheets → JSON
+python main.py convert input.xlsx --format json --sheet "Sheet1" --sheet "Sheet2"
+```
+
+### Python API
+
+```python
+from docconvert.controller import ConversionController
+
+controller = ConversionController()
+results = controller.convert_files(
+    files=["input.xlsx", "report.docx"],
+    output_fmt="md",
+    enhanced_md=True,
+)
+
+for name, path, error in results:
+    if error:
+        print(f"Failed: {name} – {error}")
+    else:
+        print(f"OK: {name} → {path}")
+```
+
+### RAG Pipeline Integration
+
+```python
+from docconvert.controller import ConversionController
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+# Convert and clean
+controller = ConversionController()
+docs = []
+for name, path, error in controller.convert_files(
+    files=["contracts/*.docx"],
+    output_fmt="md",
+    enhanced_md=True,
+):
+    if not error:
+        with open(path) as f:
+            docs.append(f.read())
+
+# Chunk and embed — noise already removed
+splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+chunks = splitter.split_text("\n".join(docs))
+```
+
+## Output Preview
+
+**Input** — an Excel sheet with merged cells:
+
+|   Region  |  Q1 |  Q2 |
+| :-------: | :-: | :-: |
+| **North** | 120 | 150 |
+| **South** |  90 | 200 |
+
+→ **Markdown output** (auto-cleaned):
+
+```markdown
+## Region    Q1    Q2
+North       120   150
+South        90   200
+```
+
+→ **JSON output**:
+
+```json
+{
+  "Region": ["North", "South"],
+  "Q1": [120, 90],
+  "Q2": [150, 200]
+}
+```
+
+## Smart Cleaning Pipeline
+
+The `--enhanced` flag runs a configurable cleaning pass that removes common document noise before output. Each rule is independently toggleable:
+
+```python
+from docconvert.config import AppConfig
+
+config = AppConfig(
+    cleaning_rules={
+        "remove_page_numbers": True,    # strips 1, 2, 3… and "Page X of Y"
+        "remove_duplicate_headers": True,  # deduplicates repeating section titles
+        "remove_empty_lines": True,     # collapses excessive blank lines
+        "normalize_spaces": True,       # single-spaces text, preserves tables
+    }
+)
+```
+
+All four rules are enabled by default with `--enhanced`. Set any to `False` to keep the raw output.
+
+## Features
+
+* **Excel** — Sheet selection, merged cells (rowspan/colspan), HTML / Markdown / JSON
+
+* **Word** — `.docx` via `python-docx` + `mammoth`, legacy `.doc` via `textract`
+
+* **Smart Markdown** — Removes page numbers, duplicate headers, collapses blank lines; all rules configurable
+
+* **GUI** — Tkinter desktop app with file list, preview, progress bar, overwrite protection
+
+* **CLI** — One-line batch conversion via argparse
+
+* **Python API** — Programmatic control with full type hints
+
+* **Executable releases** — Download a standalone `.exe` for Windows / macOS / Linux, no Python install needed
+
+## Releases (no Python needed)
+
+Standalone executables for Windows, macOS, and Linux are built automatically on each tag push. Download them from [Releases](https://github.com/leop017/DocConvert/releases).
+
+## Project Layout
+
+```
+docconvert/
+  converters/     # Excel / Word / .doc readers
+  cleaners/       # Markdown cleaning pipeline
+  exporters/      # HTML / Markdown / JSON output
+  controller/     # Orchestration, async, overwrite checks
+  gui/            # Tkinter desktop app
+  parsers/, chunkers/  # Extension points
+tests/
+main.py           # GUI / CLI entry point
+```
+
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md).
+
+## License
+
+MIT

docconvert_local-2.0.0/README.md

@@ -0,0 +1,212 @@
+# DocConvert
+
+> Clean Markdown & structured data from Excel & Word — built for RAG pipelines and LLM workflows.
+> Works 100% offline. No API keys. No data leaves your machine.
+
+[![PyPI version](https://img.shields.io/pypi/v/docconvert-local.svg)](https://pypi.org/project/docconvert-local/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/)
+[![Build](https://github.com/leop017/DocConvert/actions/workflows/ci.yml/badge.svg)](https://github.com/leop017/DocConvert/actions)
+[![Downloads](https://static.pepy.tech/badge/docconvert-local/month)](https://pepy.tech/project/docconvert-local)
+[![Stars](https://img.shields.io/github/stars/leop017/DocConvert?style=social)](https://github.com/leop017/DocConvert/stargazers)
+
+## Why DocConvert
+
+Most document-to-Markdown tools convert the file — they don't **clean** it. Raw outputs are full of page numbers, duplicate headers, and whitespace noise that eats your context window and dilutes retrieval quality.
+
+DocConvert was built for people who feed documents into LLMs and need every token to count.
+
+| Feature                              |       DocConvert      | MarkItDown | Pandas + python-docx |
+| ------------------------------------ | :-------------------: | :--------: | :------------------: |
+| Legacy `.doc` support                |           ✅           |      ❌     |           ❌          |
+| Excel merged cells (rowspan/colspan) |           ✅           |     ⚠️     |        Manual        |
+| Configurable cleaning pipeline       | ✅ 4 rules, toggle any |      ❌     |           ❌          |
+| Batch + specific sheet selection     |           ✅           |      ✅     |           ❌          |
+| Desktop GUI (no terminal needed)     |           ✅           |      ❌     |           ❌          |
+| PDF / PPT / audio support            |           ❌           |      ✅     |           ❌          |
+| MCP server / Claude integration      |           ❌           |      ✅     |           ❌          |
+| 100% offline, no cloud dependency    |           ✅           |      ✅     |           ✅          |
+
+**Choose DocConvert if:** you work with Excel/Word documents inside an organization, need legacy `.doc` support, or want a configurable cleaning pipeline before feeding docs into a RAG system.
+
+**Choose MarkItDown if:** you need PDF, PPT, images, or audio conversion, or want MCP/Claude Desktop integration out of the box.
+
+## Use Cases
+
+* **RAG ingestion** — clean Excel financial reports and Word contracts into Markdown ready for embedding
+
+* **LLM context prep** — strip page numbers, duplicates, and noise before chunking
+
+* **Offline compliance** — convert sensitive documents without uploading to any cloud service
+
+* **Batch automation** — convert entire folders of reports into a structured directory
+
+## Installation
+
+```bash
+pip install docconvert-local
+```
+
+Optional extras:
+
+```bash
+# Legacy .doc support (Linux / macOS only)
+pip install docconvert-local[doc]
+
+# Full feature set including build tools
+pip install docconvert-local[all]
+```
+
+## Quick Start
+
+### GUI (interactive)
+
+```bash
+python main.py
+```
+
+### CLI (batch / scripting)
+
+```bash
+# Single file → clean Markdown
+python main.py convert input.xlsx --format md
+
+# Batch convert with enhanced cleaning (recommended for RAG)
+python main.py convert input.docx --format md --enhanced
+
+# Multiple files → HTML into output/
+python main.py convert file1.xlsx file2.docx --format html -o ./output
+
+# Pick specific Excel sheets → JSON
+python main.py convert input.xlsx --format json --sheet "Sheet1" --sheet "Sheet2"
+```
+
+### Python API
+
+```python
+from docconvert.controller import ConversionController
+
+controller = ConversionController()
+results = controller.convert_files(
+    files=["input.xlsx", "report.docx"],
+    output_fmt="md",
+    enhanced_md=True,
+)
+
+for name, path, error in results:
+    if error:
+        print(f"Failed: {name} – {error}")
+    else:
+        print(f"OK: {name} → {path}")
+```
+
+### RAG Pipeline Integration
+
+```python
+from docconvert.controller import ConversionController
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+# Convert and clean
+controller = ConversionController()
+docs = []
+for name, path, error in controller.convert_files(
+    files=["contracts/*.docx"],
+    output_fmt="md",
+    enhanced_md=True,
+):
+    if not error:
+        with open(path) as f:
+            docs.append(f.read())
+
+# Chunk and embed — noise already removed
+splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+chunks = splitter.split_text("\n".join(docs))
+```
+
+## Output Preview
+
+**Input** — an Excel sheet with merged cells:
+
+|   Region  |  Q1 |  Q2 |
+| :-------: | :-: | :-: |
+| **North** | 120 | 150 |
+| **South** |  90 | 200 |
+
+→ **Markdown output** (auto-cleaned):
+
+```markdown
+## Region    Q1    Q2
+North       120   150
+South        90   200
+```
+
+→ **JSON output**:
+
+```json
+{
+  "Region": ["North", "South"],
+  "Q1": [120, 90],
+  "Q2": [150, 200]
+}
+```
+
+## Smart Cleaning Pipeline
+
+The `--enhanced` flag runs a configurable cleaning pass that removes common document noise before output. Each rule is independently toggleable:
+
+```python
+from docconvert.config import AppConfig
+
+config = AppConfig(
+    cleaning_rules={
+        "remove_page_numbers": True,    # strips 1, 2, 3… and "Page X of Y"
+        "remove_duplicate_headers": True,  # deduplicates repeating section titles
+        "remove_empty_lines": True,     # collapses excessive blank lines
+        "normalize_spaces": True,       # single-spaces text, preserves tables
+    }
+)
+```
+
+All four rules are enabled by default with `--enhanced`. Set any to `False` to keep the raw output.
+
+## Features
+
+* **Excel** — Sheet selection, merged cells (rowspan/colspan), HTML / Markdown / JSON
+
+* **Word** — `.docx` via `python-docx` + `mammoth`, legacy `.doc` via `textract`
+
+* **Smart Markdown** — Removes page numbers, duplicate headers, collapses blank lines; all rules configurable
+
+* **GUI** — Tkinter desktop app with file list, preview, progress bar, overwrite protection
+
+* **CLI** — One-line batch conversion via argparse
+
+* **Python API** — Programmatic control with full type hints
+
+* **Executable releases** — Download a standalone `.exe` for Windows / macOS / Linux, no Python install needed
+
+## Releases (no Python needed)
+
+Standalone executables for Windows, macOS, and Linux are built automatically on each tag push. Download them from [Releases](https://github.com/leop017/DocConvert/releases).
+
+## Project Layout
+
+```
+docconvert/
+  converters/     # Excel / Word / .doc readers
+  cleaners/       # Markdown cleaning pipeline
+  exporters/      # HTML / Markdown / JSON output
+  controller/     # Orchestration, async, overwrite checks
+  gui/            # Tkinter desktop app
+  parsers/, chunkers/  # Extension points
+tests/
+main.py           # GUI / CLI entry point
+```
+
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md).
+
+## License
+
+MIT

docconvert_local-2.0.0/docconvert/__init__.py

@@ -0,0 +1,15 @@
+"""DocConvert - 文档转换工具
+
+The GUI is imported lazily so CLI / library use (``python main.py convert``
+or ``from docconvert.controller import ConversionController``) does not
+require the Tkinter package.
+"""
+
+__all__ = ["DocConvertApp"]
+
+
+def __getattr__(name: str):
+    if name == "DocConvertApp":
+        from docconvert.gui.app import DocConvertApp
+        return DocConvertApp
+    raise AttributeError(f"module 'docconvert' has no attribute {name!r}")

docconvert_local-2.0.0/docconvert/chunkers/__init__.py

@@ -0,0 +1,5 @@
+from docconvert.chunkers.table_chunker import BaseChunker
+
+__all__ = [
+    "BaseChunker",
+]

docconvert_local-2.0.0/docconvert/chunkers/table_chunker.py

@@ -0,0 +1,10 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+from typing import Any
+
+
+class BaseChunker(ABC):
+    @abstractmethod
+    def chunk(self, content: Any, **kwargs) -> list[Any]:
+        ...

docconvert_local-2.0.0/docconvert/cleaners/__init__.py

@@ -0,0 +1,7 @@
+from docconvert.cleaners.base import BaseCleaner
+from docconvert.cleaners.word_md import WordMdCleaner
+
+__all__ = [
+    "BaseCleaner",
+    "WordMdCleaner",
+]

docconvert_local-2.0.0/docconvert/cleaners/base.py

@@ -0,0 +1,10 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+
+
+class BaseCleaner(ABC):
+
+    @abstractmethod
+    def clean(self, content: str, **kwargs) -> str:
+        ...