compilable exe

Author: tjtanaa

docs/Windows.md

@@ -15,10 +15,11 @@
 
 2. Setup python dependencies
     ```
-    cd unstructured-api
-    python -m pip install -r requirements/base.txt
-    python -c "import nltk; nltk.download('punkt')"
-    python -c "import nltk; nltk.download('averaged_perceptron_tagger')"
+    cd <path/to/repo/>
+    python -m pip install -r requirements/win-base.txt // uvloop does not support Windows
+    python -c "import nltk; nltk.download('punkt', download_dir='nltk_data')"
+    python -c "import nltk; nltk.download('averaged_perceptron_tagger', download_dir='nltk_data')"
+    pip install httpx htmlBuilder pydantic_settings
     ```
 
 3. Launch the app
@@ -40,7 +41,7 @@
     python -m pip install -r requirements/win-base.txt // uvloop does not support Windows
     python -c "import nltk; nltk.download('punkt', download_dir='nltk_data')"
     python -c "import nltk; nltk.download('averaged_perceptron_tagger', download_dir='nltk_data')"
-    pip install httpx htmlBuilder
+    pip install httpx htmlBuilder pydantic_settings
     ```
 4. Download `sqlite3.dll` as a patch from [link](https://www.sqlite.org/download.html). Download [sqlite-dll-win-x64-3460000.zip](https://www.sqlite.org/2024/sqlite-dll-win-x64-3460000.zip). Somehow python native `sqlite3` does not work. Place it at the root of repository. Unzip the zip file. 
 5.  Convert the unstructuredioapi repo into a python package by updating the `pyproject.toml`

prepline_general/api/app.py

@@ -5,8 +5,8 @@
 import logging
 import os
 
-from .general import router as general_router
-from .openapi import set_custom_openapi
+from prepline_general.api.general import router as general_router
+from prepline_general.api.openapi import set_custom_openapi
 
 logger = logging.getLogger("unstructured_api")
 

pyproject.toml

@@ -1,3 +1,24 @@
+
+[build-system]
+# setuptools-scm considers all files tracked by git to be data files
+requires = ["setuptools>=62.0", "setuptools-scm"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "prepline_general"
+description = "UnstructuredIO API"
+readme = "README.md"
+requires-python = "~=3.10"
+# keywords = ["one", "two"]
+license = { text = "Proprietary" }
+classifiers = [ # https://pypi.org/classifiers/
+    "Development Status :: 3 - Alpha",
+    "Programming Language :: Python :: 3 :: Only",
+    "Intended Audience :: Information Technology",
+    "Operating System :: Unix",
+]
+version="0.0.68"
+
 [tool.black]
 line-length = 100
 
@@ -35,3 +56,6 @@ known-first-party = [
     "unstructured",
     "unstructured_inference",
 ]
+
+[tool.setuptools.packages.find]
+where = ["."]

unstructuredio_api.py

@@ -0,0 +1,33 @@
+import os
+from pydantic_settings import BaseSettings, SettingsConfigDict
+
+
+class Config(BaseSettings):
+    model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8", extra="ignore")
+    unstapi_port: int = 6989
+    unstapi_host: str = "0.0.0.0"
+    unstapi_workers: int = 1
+    log_config: str = os.path.join("_internal", "config", "logger_config.yaml")
+
+
+config = Config()
+
+if __name__ == "__main__":
+    import uvicorn
+
+    # if os.name == "nt":
+    from multiprocessing import freeze_support
+
+    freeze_support()
+    print("The system is Windows.")
+    # else:
+    #     print("The system is not Windows.")
+
+    uvicorn.run(
+        "prepline_general.api.app:app",
+        reload=False,
+        port=config.unstapi_port,
+        host=config.unstapi_host,
+        workers=config.unstapi_workers,
+        log_config=config.log_config,
+    )

unstructuredio_api.spec

@@ -0,0 +1,75 @@
+# -*- mode: python ; coding: utf-8 -*-
+
+import os
+from pathlib import Path
+import sys
+from PyInstaller.utils.hooks import collect_all
+
+binaries_list = [
+    ('C:\\Program Files\\LibreOffice\\program', 'libreoffice'), # modify this to point to where the LibreOffice is installed
+    (Path('sqlite-dll-win-x64-3460000/sqlite3.dll').as_posix(), '.'), # modify this to point to where you unzip the sqlite3.dll
+    (Path('sqlite-dll-win-x64-3460000/sqlite3.def').as_posix(), '.'), # modify this to point to where you unzip the sqlite3.def
+
+]
+
+datas_list = [
+    (Path('logger_config.yaml').as_posix(), 'config'), # modify this to point to where the repo is
+    (Path('nltk_data').as_posix(), 'nltk_data') # modify this to point to where nltk download the nltk data
+]
+
+hiddenimports_list = []
+
+def add_package(package_name):
+    datas, binaries, hiddenimports = collect_all(package_name)
+    datas_list.extend(datas)
+    binaries_list.extend(binaries)
+    hiddenimports_list.extend(hiddenimports)
+
+# Collect all resources from the package_name
+add_package('unstructured')
+add_package('effdet')
+add_package('onnxruntime')
+add_package('encodings')
+add_package('prepline_general')
+
+a = Analysis(
+    [Path('unstructuredio_api.py').as_posix()],
+    pathex=[],
+    binaries=binaries_list,
+    datas=datas_list,
+    hiddenimports=hiddenimports_list,
+    hookspath=[],
+    hooksconfig={},
+    runtime_hooks=[],
+    excludes=[],
+    noarchive=False,
+    optimize=0,
+)
+pyz = PYZ(a.pure)
+
+exe = EXE(
+    pyz,
+    a.scripts,
+    [],
+    exclude_binaries=True,
+    name='unstructuredio_api',
+    debug=False,
+    bootloader_ignore_signals=False,
+    strip=False,
+    upx=True,
+    console=True,
+    disable_windowed_traceback=False,
+    argv_emulation=False,
+    target_arch=None,
+    codesign_identity=None,
+    entitlements_file=None,
+)
+coll = COLLECT(
+    exe,
+    a.binaries,
+    a.datas,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    name='unstructuredio_api',
+)