doctra 0.1.1__tar.gz → 0.3.0__tar.gz

{doctra-0.1.1/doctra.egg-info → doctra-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: doctra
-Version: 0.1.1
+Version: 0.3.0
 Summary: Parse, extract, and analyze documents with ease
 Home-page: https://github.com/AdemBoukhris457/Doctra
 Author: Adem Boukhris
@@ -241,6 +241,8 @@ Provides-Extra: openai
 Requires-Dist: openai>=1.0.0; extra == "openai"
 Provides-Extra: gemini
 Requires-Dist: google-generativeai>=0.3.0; extra == "gemini"
+Provides-Extra: anthropic
+Requires-Dist: anthropic>=0.40.0; extra == "anthropic"
 Provides-Extra: dev
 Requires-Dist: pytest>=6.0; extra == "dev"
 Requires-Dist: pytest-cov>=2.0; extra == "dev"
@@ -329,7 +331,7 @@ parser = StructuredPDFParser()
 # Parser with VLM for structured data extraction
 parser = StructuredPDFParser(
     use_vlm=True,
-    vlm_provider="openai",  # or "gemini"
+    vlm_provider="openai",  # or "gemini" or "anthropic" or "openrouter"
     vlm_api_key="your_api_key_here"
 )
 
@@ -344,7 +346,7 @@ parser = StructuredPDFParser(
     # VLM Settings
     use_vlm=True,
     vlm_provider="openai",
-    vlm_model="gpt-4o",
+    vlm_model="gpt-5",
     vlm_api_key="your_api_key",
     
     # Layout Detection Settings
@@ -406,7 +408,7 @@ parser = ChartTablePDFParser(
     # VLM Settings
     use_vlm=True,
     vlm_provider="openai",
-    vlm_model="gpt-4o",
+    vlm_model="gpt-5",
     vlm_api_key="your_api_key",
     
     # Layout Detection Settings
@@ -545,7 +547,7 @@ parser = StructuredPDFParser(
     use_vlm=True,
     vlm_provider="openai",
     vlm_api_key="your_openai_api_key",
-    vlm__model="gpt-4o",
+    vlm__model="gpt-5",
     layout_model_name="PP-DocLayout_plus-L",
     dpi=300,  # Higher DPI for better quality
     min_score=0.5,  # Higher confidence threshold
@@ -623,4 +625,41 @@ parser.display_pages_with_boxes("document.pdf")
 - **Pandas**: Data manipulation
 - **OpenPyXL**: Excel file generation
 - **Google Generative AI**: For Gemini VLM integration
-- **OpenAI**: For GPT-4 VLM integration
+- **OpenAI**: For GPT-5 VLM integration
+
+## 🖥️ Web Interface (Gradio)
+
+You can try Doctra in a simple web UI powered by Gradio.
+
+### Run locally
+
+```bash
+pip install -U gradio
+python gradio_app.py
+```
+
+Then open the printed URL (default `http://127.0.0.1:7860`).
+
+Notes:
+- If using VLM, set the API key field in the UI or export `VLM_API_KEY`.
+- Outputs are saved under `outputs/<pdf_stem>/` and previewed in the UI.
+
+### Deploy on Hugging Face Spaces
+
+1) Create a new Space (type: Gradio, SDK: Python).
+
+2) Add these files to the Space repo:
+   - Your package code (or install from PyPI).
+   - `gradio_app.py` (entry point).
+   - `requirements.txt` with at least:
+
+```text
+doctra
+gradio
+```
+
+3) Set a secret named `VLM_API_KEY` if you want VLM features.
+
+4) In Space settings, set `python gradio_app.py` as the run command (or rely on auto-detect).
+
+The Space will build and expose the same interface for uploads and processing.