Spaces:
Sleeping
Sleeping
Upload 5 files
Browse files- Dockerfile.hf +43 -0
- README.md +121 -11
- app.py +91 -0
- main.py +342 -0
- requirements.txt +6 -0
Dockerfile.hf
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.10-slim
|
| 2 |
+
|
| 3 |
+
# Install system dependencies required for ocrmypdf and Python packages
|
| 4 |
+
RUN apt-get update && apt-get install -y --no-install-recommends \
|
| 5 |
+
ocrmypdf \
|
| 6 |
+
tesseract-ocr \
|
| 7 |
+
poppler-utils \
|
| 8 |
+
libsnappy-dev \
|
| 9 |
+
zlib1g-dev \
|
| 10 |
+
libbz2-dev \
|
| 11 |
+
libreadline-dev \
|
| 12 |
+
libsqlite3-dev \
|
| 13 |
+
wget \
|
| 14 |
+
curl \
|
| 15 |
+
llvm \
|
| 16 |
+
libncurses5-dev \
|
| 17 |
+
libncursesw5-dev \
|
| 18 |
+
xz-utils \
|
| 19 |
+
tk-dev \
|
| 20 |
+
libssl-dev \
|
| 21 |
+
libffi-dev \
|
| 22 |
+
liblzma-dev \
|
| 23 |
+
python3-dev \
|
| 24 |
+
libopenjp2-7-dev \
|
| 25 |
+
libharfbuzz-dev \
|
| 26 |
+
libfribidi-dev \
|
| 27 |
+
libxcb-composite0-dev \
|
| 28 |
+
libatlas-base-dev \
|
| 29 |
+
git \
|
| 30 |
+
gcc \
|
| 31 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 32 |
+
|
| 33 |
+
WORKDIR /app
|
| 34 |
+
|
| 35 |
+
COPY requirements.txt .
|
| 36 |
+
|
| 37 |
+
RUN pip install --no-cache-dir -r requirements.txt
|
| 38 |
+
|
| 39 |
+
COPY . .
|
| 40 |
+
|
| 41 |
+
EXPOSE 7860
|
| 42 |
+
|
| 43 |
+
CMD ["python", "app.py"]
|
README.md
CHANGED
|
@@ -1,11 +1,121 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
-
|
| 10 |
-
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Document Converter Pro
|
| 2 |
+
|
| 3 |
+
A powerful document conversion tool that combines PDF OCR conversion and Excel to CSV conversion in a single interface.
|
| 4 |
+
|
| 5 |
+
## Features
|
| 6 |
+
|
| 7 |
+
- **PDF OCR Conversion**: Convert scanned PDFs to searchable, text-based PDFs using ocrmypdf
|
| 8 |
+
- **Excel to CSV Conversion**: Convert Excel files to CSV format quickly and efficiently
|
| 9 |
+
- **Web Interface**: Built with Gradio for an intuitive user experience
|
| 10 |
+
- **Optimized Processing**: Fast conversion with optimized parameters
|
| 11 |
+
|
| 12 |
+
## Deployment to Hugging Face Spaces
|
| 13 |
+
|
| 14 |
+
This application is ready to be deployed on Hugging Face Spaces as a free, public application.
|
| 15 |
+
|
| 16 |
+
### Steps to Deploy:
|
| 17 |
+
|
| 18 |
+
1. **Sign up on Hugging Face**
|
| 19 |
+
- Go to [huggingface.co](https://huggingface.co) and create an account
|
| 20 |
+
|
| 21 |
+
2. **Create a New Space**
|
| 22 |
+
- Click on your profile icon → "New Space"
|
| 23 |
+
- Fill in the details:
|
| 24 |
+
- Name: Choose a unique name for your application
|
| 25 |
+
- License: Select appropriate license
|
| 26 |
+
- SDK: Select "Docker" (required for ocrmypdf)
|
| 27 |
+
- Visibility: Public or Private
|
| 28 |
+
|
| 29 |
+
3. **Configure the Space**
|
| 30 |
+
- Use the files provided in this repository:
|
| 31 |
+
- `app.py` - Main application file
|
| 32 |
+
- `requirements.txt` - Python dependencies
|
| 33 |
+
- `Dockerfile` - Container configuration (see below)
|
| 34 |
+
|
| 35 |
+
4. **Create a Dockerfile** (needed for ocrmypdf dependencies)
|
| 36 |
+
```Dockerfile
|
| 37 |
+
FROM huggingface/transformers-all-deepspeed:latest
|
| 38 |
+
|
| 39 |
+
RUN apt-get update && apt-get install -y \
|
| 40 |
+
ocrmypdf \
|
| 41 |
+
tesseract-ocr \
|
| 42 |
+
poppler-utils \
|
| 43 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 44 |
+
|
| 45 |
+
COPY requirements.txt .
|
| 46 |
+
RUN pip install -r requirements.txt
|
| 47 |
+
|
| 48 |
+
COPY . .
|
| 49 |
+
|
| 50 |
+
CMD ["python", "app.py"]
|
| 51 |
+
```
|
| 52 |
+
|
| 53 |
+
5. **Alternative Dockerfile** (lighter image)
|
| 54 |
+
```Dockerfile
|
| 55 |
+
FROM python:3.10-slim
|
| 56 |
+
|
| 57 |
+
# Install system dependencies
|
| 58 |
+
RUN apt-get update && apt-get install -y \
|
| 59 |
+
ocrmypdf \
|
| 60 |
+
tesseract-ocr \
|
| 61 |
+
poppler-utils \
|
| 62 |
+
gcc \
|
| 63 |
+
&& rm -rf /var/lib/apt/lists/*
|
| 64 |
+
|
| 65 |
+
WORKDIR /app
|
| 66 |
+
|
| 67 |
+
COPY requirements.txt .
|
| 68 |
+
RUN pip install -r requirements.txt
|
| 69 |
+
|
| 70 |
+
COPY . .
|
| 71 |
+
|
| 72 |
+
EXPOSE 7860
|
| 73 |
+
|
| 74 |
+
CMD ["python", "app.py"]
|
| 75 |
+
```
|
| 76 |
+
|
| 77 |
+
6. **Upload Files**
|
| 78 |
+
- Upload all files to your Space repository
|
| 79 |
+
- Or connect your GitHub repository for automatic synchronization
|
| 80 |
+
|
| 81 |
+
7. **Wait for Build**
|
| 82 |
+
- Hugging Face will automatically build and deploy your application
|
| 83 |
+
- This may take several minutes due to the size of OCR dependencies
|
| 84 |
+
|
| 85 |
+
## How to Use
|
| 86 |
+
|
| 87 |
+
### PDF OCR Conversion:
|
| 88 |
+
1. Navigate to the "PDF OCR Converter" tab
|
| 89 |
+
2. Upload your PDF file
|
| 90 |
+
3. Click "Process PDF with OCR"
|
| 91 |
+
4. Wait for processing to complete
|
| 92 |
+
5. Download the OCRed PDF when finished
|
| 93 |
+
|
| 94 |
+
### Excel to CSV Conversion:
|
| 95 |
+
1. Navigate to the "Excel to CSV Converter" tab
|
| 96 |
+
2. Upload your Excel file (.xlsx or .xls)
|
| 97 |
+
3. Click "Convert to CSV"
|
| 98 |
+
4. Download the converted CSV file when finished
|
| 99 |
+
|
| 100 |
+
## Technical Details
|
| 101 |
+
|
| 102 |
+
The application uses:
|
| 103 |
+
- **ocrmypdf**: For PDF OCR processing with parameters: `--force-ocr --deskew --rotate-pages --oversample 300`
|
| 104 |
+
- **pandas**: For fast Excel to CSV conversion with optimized settings
|
| 105 |
+
- **Gradio**: For the user interface
|
| 106 |
+
- **openpyxl**: For Excel file processing
|
| 107 |
+
|
| 108 |
+
## Requirements
|
| 109 |
+
|
| 110 |
+
- Python 3.8+
|
| 111 |
+
- System dependencies: ocrmypdf, tesseract-ocr, poppler-utils
|
| 112 |
+
|
| 113 |
+
## Local Development
|
| 114 |
+
|
| 115 |
+
To run locally:
|
| 116 |
+
```bash
|
| 117 |
+
pip install -r requirements.txt
|
| 118 |
+
python app.py
|
| 119 |
+
```
|
| 120 |
+
|
| 121 |
+
Then visit `http://localhost:7860` in your browser.
|
app.py
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import ocrmypdf
|
| 3 |
+
import pandas as pd
|
| 4 |
+
import openpyxl
|
| 5 |
+
import tempfile
|
| 6 |
+
import os
|
| 7 |
+
import uuid
|
| 8 |
+
from werkzeug.utils import secure_filename
|
| 9 |
+
|
| 10 |
+
def ocr_pdf(input_pdf):
|
| 11 |
+
"""Process PDF with OCR using ocrmypdf"""
|
| 12 |
+
if input_pdf is None:
|
| 13 |
+
return None, "Please upload a PDF file"
|
| 14 |
+
|
| 15 |
+
with tempfile.NamedTemporaryFile(suffix='.pdf', delete=False) as temp_output:
|
| 16 |
+
try:
|
| 17 |
+
# Process with ocrmypdf using the parameters you specified
|
| 18 |
+
ocrmypdf.ocr(
|
| 19 |
+
input_pdf.name,
|
| 20 |
+
temp_output.name,
|
| 21 |
+
force_ocr=True,
|
| 22 |
+
deskew=True,
|
| 23 |
+
rotate_pages=True,
|
| 24 |
+
oversample=300
|
| 25 |
+
)
|
| 26 |
+
return temp_output.name, "PDF processed successfully!"
|
| 27 |
+
except Exception as e:
|
| 28 |
+
return None, f"Error processing PDF: {str(e)}"
|
| 29 |
+
|
| 30 |
+
def convert_excel_to_csv(input_excel):
|
| 31 |
+
"""Convert Excel file to CSV"""
|
| 32 |
+
if input_excel is None:
|
| 33 |
+
return None, "Please upload an Excel file"
|
| 34 |
+
|
| 35 |
+
try:
|
| 36 |
+
# Read Excel file with optimizations for speed
|
| 37 |
+
df = pd.read_excel(
|
| 38 |
+
input_excel.name,
|
| 39 |
+
engine="openpyxl",
|
| 40 |
+
dtype=str, # Big speed boost
|
| 41 |
+
parse_dates=False, # No date parsing
|
| 42 |
+
keep_default_na=False
|
| 43 |
+
)
|
| 44 |
+
|
| 45 |
+
# Create temporary CSV file
|
| 46 |
+
with tempfile.NamedTemporaryFile(mode='w', suffix='.csv', delete=False, newline='', encoding='utf-8') as temp_csv:
|
| 47 |
+
df.to_csv(temp_csv.name, index=False, encoding="utf-8")
|
| 48 |
+
return temp_csv.name, "Excel converted to CSV successfully!"
|
| 49 |
+
except Exception as e:
|
| 50 |
+
return None, f"Error converting Excel: {str(e)}"
|
| 51 |
+
|
| 52 |
+
# Create Gradio interface with tabs
|
| 53 |
+
with gr.Blocks(title="Document Converter Pro", theme=gr.themes.Soft()) as demo:
|
| 54 |
+
gr.Markdown("# 📄 Document Converter Pro")
|
| 55 |
+
gr.Markdown("Convert your documents with ease using our powerful online tools")
|
| 56 |
+
|
| 57 |
+
with gr.Tabs():
|
| 58 |
+
with gr.TabItem("PDF OCR Converter"):
|
| 59 |
+
gr.Markdown("### Convert scanned PDFs to searchable, text-based PDFs")
|
| 60 |
+
with gr.Row():
|
| 61 |
+
with gr.Column():
|
| 62 |
+
pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
|
| 63 |
+
pdf_btn = gr.Button("Process PDF with OCR")
|
| 64 |
+
with gr.Column():
|
| 65 |
+
pdf_output = gr.File(label="Download OCRed PDF")
|
| 66 |
+
pdf_status = gr.Textbox(label="Status", interactive=False)
|
| 67 |
+
|
| 68 |
+
pdf_btn.click(
|
| 69 |
+
fn=ocr_pdf,
|
| 70 |
+
inputs=pdf_input,
|
| 71 |
+
outputs=[pdf_output, pdf_status]
|
| 72 |
+
)
|
| 73 |
+
|
| 74 |
+
with gr.TabItem("Excel to CSV Converter"):
|
| 75 |
+
gr.Markdown("### Convert Excel files to CSV format quickly and efficiently")
|
| 76 |
+
with gr.Row():
|
| 77 |
+
with gr.Column():
|
| 78 |
+
excel_input = gr.File(label="Upload Excel", file_types=[".xlsx", ".xls"])
|
| 79 |
+
excel_btn = gr.Button("Convert to CSV")
|
| 80 |
+
with gr.Column():
|
| 81 |
+
excel_output = gr.File(label="Download CSV")
|
| 82 |
+
excel_status = gr.Textbox(label="Status", interactive=False)
|
| 83 |
+
|
| 84 |
+
excel_btn.click(
|
| 85 |
+
fn=convert_excel_to_csv,
|
| 86 |
+
inputs=excel_input,
|
| 87 |
+
outputs=[excel_output, excel_status]
|
| 88 |
+
)
|
| 89 |
+
|
| 90 |
+
if __name__ == "__main__":
|
| 91 |
+
demo.launch(server_port=int(os.getenv("PORT", 7860)))
|
main.py
ADDED
|
@@ -0,0 +1,342 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from flask import Flask, render_template, request, send_file, redirect, url_for, flash, jsonify
|
| 2 |
+
import os
|
| 3 |
+
import subprocess
|
| 4 |
+
import tempfile
|
| 5 |
+
import uuid
|
| 6 |
+
from werkzeug.utils import secure_filename
|
| 7 |
+
import threading
|
| 8 |
+
import time
|
| 9 |
+
|
| 10 |
+
app = Flask(__name__)
|
| 11 |
+
app.secret_key = 'your-secret-key-here'
|
| 12 |
+
|
| 13 |
+
# Import for Excel processing
|
| 14 |
+
import pandas as pd
|
| 15 |
+
import openpyxl
|
| 16 |
+
|
| 17 |
+
# Configuration
|
| 18 |
+
UPLOAD_FOLDER = 'uploads'
|
| 19 |
+
PROCESSED_FOLDER = 'processed'
|
| 20 |
+
ALLOWED_EXTENSIONS = {'pdf'}
|
| 21 |
+
|
| 22 |
+
app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
|
| 23 |
+
app.config['PROCESSED_FOLDER'] = PROCESSED_FOLDER
|
| 24 |
+
|
| 25 |
+
# Create necessary directories
|
| 26 |
+
os.makedirs(UPLOAD_FOLDER, exist_ok=True)
|
| 27 |
+
os.makedirs(PROCESSED_FOLDER, exist_ok=True)
|
| 28 |
+
|
| 29 |
+
# Global dictionary to store processing status
|
| 30 |
+
processing_status = {}
|
| 31 |
+
|
| 32 |
+
def allowed_file(filename):
|
| 33 |
+
"""Check if the uploaded file has an allowed extension"""
|
| 34 |
+
return '.' in filename and \
|
| 35 |
+
filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
def validate_pdf_file(file_path):
|
| 39 |
+
"""Validate that the uploaded file is actually a PDF"""
|
| 40 |
+
try:
|
| 41 |
+
# Check if file exists and has PDF header
|
| 42 |
+
if not os.path.exists(file_path):
|
| 43 |
+
return False, "File does not exist"
|
| 44 |
+
|
| 45 |
+
# Read first few bytes to check PDF header
|
| 46 |
+
with open(file_path, 'rb') as f:
|
| 47 |
+
header = f.read(4)
|
| 48 |
+
if header != b'%PDF':
|
| 49 |
+
return False, "File is not a valid PDF"
|
| 50 |
+
|
| 51 |
+
# Check file size (max 50MB)
|
| 52 |
+
file_size = os.path.getsize(file_path)
|
| 53 |
+
max_size = 50 * 1024 * 1024 # 50MB
|
| 54 |
+
if file_size > max_size:
|
| 55 |
+
return False, f"File size too large. Max size is 50MB, got {file_size / (1024*1024):.1f}MB"
|
| 56 |
+
|
| 57 |
+
return True, "Valid PDF"
|
| 58 |
+
except Exception as e:
|
| 59 |
+
return False, f"Error validating PDF: {str(e)}"
|
| 60 |
+
|
| 61 |
+
def process_pdf_with_ocr(input_path, output_path, job_id=None):
|
| 62 |
+
"""Process PDF with OCR using ocrmypdf"""
|
| 63 |
+
try:
|
| 64 |
+
# Using the ocrmypdf command similar to what was provided
|
| 65 |
+
cmd = [
|
| 66 |
+
'ocrmypdf',
|
| 67 |
+
'--force-ocr',
|
| 68 |
+
'--deskew',
|
| 69 |
+
'--rotate-pages',
|
| 70 |
+
'--oversample', '300',
|
| 71 |
+
input_path,
|
| 72 |
+
output_path
|
| 73 |
+
]
|
| 74 |
+
|
| 75 |
+
import logging
|
| 76 |
+
logging.basicConfig(level=logging.INFO)
|
| 77 |
+
logger = logging.getLogger(__name__)
|
| 78 |
+
|
| 79 |
+
logger.info(f"Executing OCR command: {' '.join(cmd)}")
|
| 80 |
+
|
| 81 |
+
# Run the command and capture output in real-time
|
| 82 |
+
result = subprocess.run(
|
| 83 |
+
cmd,
|
| 84 |
+
capture_output=True,
|
| 85 |
+
text=True,
|
| 86 |
+
timeout=600 # 10 minute timeout
|
| 87 |
+
)
|
| 88 |
+
|
| 89 |
+
logger.info(f"OCR process completed with return code: {result.returncode}")
|
| 90 |
+
if result.stdout:
|
| 91 |
+
logger.info(f"OCR stdout: {result.stdout}")
|
| 92 |
+
if result.stderr:
|
| 93 |
+
logger.info(f"OCR stderr: {result.stderr}")
|
| 94 |
+
|
| 95 |
+
if result.returncode != 0:
|
| 96 |
+
raise Exception(f"OCR processing failed with return code {result.returncode}: {result.stderr}")
|
| 97 |
+
|
| 98 |
+
return True
|
| 99 |
+
except subprocess.TimeoutExpired:
|
| 100 |
+
raise Exception("Processing timed out after 10 minutes")
|
| 101 |
+
except Exception as e:
|
| 102 |
+
import logging
|
| 103 |
+
logging.basicConfig(level=logging.INFO)
|
| 104 |
+
logger = logging.getLogger(__name__)
|
| 105 |
+
logger.error(f"OCR processing error: {str(e)}")
|
| 106 |
+
raise e
|
| 107 |
+
|
| 108 |
+
# Main routes
|
| 109 |
+
@app.route('/')
|
| 110 |
+
def index():
|
| 111 |
+
"""Main page route"""
|
| 112 |
+
return render_template('index.html')
|
| 113 |
+
|
| 114 |
+
@app.route('/pdf-ocr')
|
| 115 |
+
def pdf_ocr():
|
| 116 |
+
"""PDF OCR converter page route"""
|
| 117 |
+
return render_template('pdf_ocr.html')
|
| 118 |
+
|
| 119 |
+
@app.route('/excel-to-csv')
|
| 120 |
+
def excel_to_csv():
|
| 121 |
+
"""Excel to CSV converter page route"""
|
| 122 |
+
return render_template('excel_to_csv.html')
|
| 123 |
+
|
| 124 |
+
# Routes for PDF OCR functionality
|
| 125 |
+
@app.route('/upload', methods=['POST'])
|
| 126 |
+
def upload_file():
|
| 127 |
+
"""Handle PDF file upload"""
|
| 128 |
+
if 'pdf_file' not in request.files:
|
| 129 |
+
return jsonify({'success': False, 'message': 'No file selected'})
|
| 130 |
+
|
| 131 |
+
file = request.files['pdf_file']
|
| 132 |
+
|
| 133 |
+
if file.filename == '':
|
| 134 |
+
return jsonify({'success': False, 'message': 'No file selected'})
|
| 135 |
+
|
| 136 |
+
if file and allowed_file(file.filename):
|
| 137 |
+
filename = secure_filename(file.filename)
|
| 138 |
+
unique_filename = f"{uuid.uuid4()}_{filename}"
|
| 139 |
+
filepath = os.path.join(app.config['UPLOAD_FOLDER'], unique_filename)
|
| 140 |
+
file.save(filepath)
|
| 141 |
+
|
| 142 |
+
# Validate the PDF file
|
| 143 |
+
is_valid, validation_msg = validate_pdf_file(filepath)
|
| 144 |
+
if not is_valid:
|
| 145 |
+
# Clean up invalid file
|
| 146 |
+
try:
|
| 147 |
+
os.remove(filepath)
|
| 148 |
+
except:
|
| 149 |
+
pass # Ignore cleanup errors
|
| 150 |
+
return jsonify({'success': False, 'message': f'Invalid PDF file: {validation_msg}'})
|
| 151 |
+
|
| 152 |
+
# Generate unique output filename
|
| 153 |
+
output_filename = f"ocr_{unique_filename}"
|
| 154 |
+
output_path = os.path.join(app.config['PROCESSED_FOLDER'], output_filename)
|
| 155 |
+
|
| 156 |
+
# Store processing status
|
| 157 |
+
job_id = str(uuid.uuid4())
|
| 158 |
+
processing_status[job_id] = {
|
| 159 |
+
'status': 'processing',
|
| 160 |
+
'input_file': filepath,
|
| 161 |
+
'output_file': output_path,
|
| 162 |
+
'progress': 0
|
| 163 |
+
}
|
| 164 |
+
|
| 165 |
+
# Start processing in a separate thread
|
| 166 |
+
thread = threading.Thread(target=process_file_thread, args=(job_id,))
|
| 167 |
+
thread.start()
|
| 168 |
+
|
| 169 |
+
return jsonify({
|
| 170 |
+
'success': True,
|
| 171 |
+
'job_id': job_id,
|
| 172 |
+
'message': 'File uploaded successfully. Processing started...'
|
| 173 |
+
})
|
| 174 |
+
else:
|
| 175 |
+
return jsonify({'success': False, 'message': 'Invalid file type. Please upload a PDF.'})
|
| 176 |
+
|
| 177 |
+
def process_file_thread(job_id):
|
| 178 |
+
"""Process file in a separate thread"""
|
| 179 |
+
status = processing_status.get(job_id)
|
| 180 |
+
if not status:
|
| 181 |
+
return
|
| 182 |
+
|
| 183 |
+
try:
|
| 184 |
+
# Update progress
|
| 185 |
+
processing_status[job_id]['progress'] = 5
|
| 186 |
+
time.sleep(0.5) # Brief pause
|
| 187 |
+
|
| 188 |
+
# Estimate progress based on file size
|
| 189 |
+
input_size = os.path.getsize(status['input_file'])
|
| 190 |
+
estimated_pages = max(1, input_size // (100 * 1024)) # Rough estimate: 1 page per 100KB
|
| 191 |
+
|
| 192 |
+
# Update progress incrementally
|
| 193 |
+
base_progress = 5
|
| 194 |
+
progress_increment = (90 - base_progress) / estimated_pages # 90% total for processing
|
| 195 |
+
|
| 196 |
+
for page in range(estimated_pages):
|
| 197 |
+
# In a real implementation, we would track actual OCR progress
|
| 198 |
+
# Here we simulate progress based on estimated pages
|
| 199 |
+
simulated_progress = int(base_progress + (page * progress_increment))
|
| 200 |
+
processing_status[job_id]['progress'] = min(simulated_progress, 95)
|
| 201 |
+
time.sleep(0.1) # Small delay to simulate processing time per page
|
| 202 |
+
|
| 203 |
+
# Perform OCR processing
|
| 204 |
+
process_pdf_with_ocr(status['input_file'], status['output_file'], job_id)
|
| 205 |
+
|
| 206 |
+
# Update status when complete
|
| 207 |
+
processing_status[job_id]['status'] = 'completed'
|
| 208 |
+
processing_status[job_id]['progress'] = 100
|
| 209 |
+
except Exception as e:
|
| 210 |
+
processing_status[job_id]['status'] = 'failed'
|
| 211 |
+
processing_status[job_id]['error'] = str(e)
|
| 212 |
+
processing_status[job_id]['progress'] = 0
|
| 213 |
+
|
| 214 |
+
# Clean up failed output file if it exists
|
| 215 |
+
try:
|
| 216 |
+
if os.path.exists(status['output_file']):
|
| 217 |
+
os.remove(status['output_file'])
|
| 218 |
+
except:
|
| 219 |
+
pass # Ignore cleanup errors
|
| 220 |
+
|
| 221 |
+
@app.route('/status/<job_id>')
|
| 222 |
+
def get_status(job_id):
|
| 223 |
+
"""Get processing status for a job"""
|
| 224 |
+
if job_id in processing_status:
|
| 225 |
+
return jsonify(processing_status[job_id])
|
| 226 |
+
else:
|
| 227 |
+
return jsonify({'status': 'not_found'})
|
| 228 |
+
|
| 229 |
+
@app.route('/download/<job_id>')
|
| 230 |
+
def download_file(job_id):
|
| 231 |
+
"""Download processed file"""
|
| 232 |
+
if job_id in processing_status:
|
| 233 |
+
status = processing_status[job_id]
|
| 234 |
+
if status['status'] == 'completed':
|
| 235 |
+
return send_file(status['output_file'], as_attachment=True,
|
| 236 |
+
download_name=f"ocr_processed_{os.path.basename(status['output_file'])}")
|
| 237 |
+
else:
|
| 238 |
+
flash('File is not ready for download')
|
| 239 |
+
return redirect(url_for('index'))
|
| 240 |
+
else:
|
| 241 |
+
flash('Invalid job ID')
|
| 242 |
+
return redirect(url_for('index'))
|
| 243 |
+
|
| 244 |
+
@app.route('/cleanup', methods=['POST'])
|
| 245 |
+
def cleanup():
|
| 246 |
+
"""Clean up temporary files"""
|
| 247 |
+
# Clean up uploaded files older than 1 hour
|
| 248 |
+
current_time = time.time()
|
| 249 |
+
for filename in os.listdir(app.config['UPLOAD_FOLDER']):
|
| 250 |
+
file_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
|
| 251 |
+
if os.path.isfile(file_path):
|
| 252 |
+
if current_time - os.path.getmtime(file_path) > 3600: # 1 hour
|
| 253 |
+
os.remove(file_path)
|
| 254 |
+
|
| 255 |
+
# Clean up processed files older than 1 hour
|
| 256 |
+
for filename in os.listdir(app.config['PROCESSED_FOLDER']):
|
| 257 |
+
file_path = os.path.join(app.config['PROCESSED_FOLDER'], filename)
|
| 258 |
+
if os.path.isfile(file_path):
|
| 259 |
+
if current_time - os.path.getmtime(file_path) > 3600: # 1 hour
|
| 260 |
+
os.remove(file_path)
|
| 261 |
+
|
| 262 |
+
# Clean up old processing statuses
|
| 263 |
+
expired_jobs = []
|
| 264 |
+
for job_id, status in processing_status.items():
|
| 265 |
+
if status['status'] == 'completed' and current_time - os.path.getmtime(status['output_file']) > 3600:
|
| 266 |
+
expired_jobs.append(job_id)
|
| 267 |
+
|
| 268 |
+
for job_id in expired_jobs:
|
| 269 |
+
del processing_status[job_id]
|
| 270 |
+
|
| 271 |
+
return jsonify({'success': True, 'message': 'Cleanup completed'})
|
| 272 |
+
|
| 273 |
+
|
| 274 |
+
# Routes for Excel to CSV functionality
|
| 275 |
+
@app.route('/convert-excel', methods=['POST'])
|
| 276 |
+
def convert_excel_to_csv():
|
| 277 |
+
"""Convert Excel files to CSV"""
|
| 278 |
+
try:
|
| 279 |
+
if 'excel_file' not in request.files:
|
| 280 |
+
return jsonify({'success': False, 'message': 'No file selected'})
|
| 281 |
+
|
| 282 |
+
file = request.files['excel_file']
|
| 283 |
+
|
| 284 |
+
if file.filename == '':
|
| 285 |
+
return jsonify({'success': False, 'message': 'No file selected'})
|
| 286 |
+
|
| 287 |
+
if file and file.filename.lower().endswith(('.xlsx', '.xls')):
|
| 288 |
+
filename = secure_filename(file.filename)
|
| 289 |
+
unique_filename = f"{uuid.uuid4()}_{filename}"
|
| 290 |
+
filepath = os.path.join(app.config['UPLOAD_FOLDER'], unique_filename)
|
| 291 |
+
file.save(filepath)
|
| 292 |
+
|
| 293 |
+
# Convert Excel to CSV
|
| 294 |
+
output_filename = f"{os.path.splitext(unique_filename)[0]}.csv"
|
| 295 |
+
output_path = os.path.join(app.config['PROCESSED_FOLDER'], output_filename)
|
| 296 |
+
|
| 297 |
+
# Process Excel file to CSV
|
| 298 |
+
df = pd.read_excel(
|
| 299 |
+
filepath,
|
| 300 |
+
engine="openpyxl",
|
| 301 |
+
dtype=str,
|
| 302 |
+
parse_dates=False,
|
| 303 |
+
keep_default_na=False
|
| 304 |
+
)
|
| 305 |
+
|
| 306 |
+
df.to_csv(output_path, index=False, encoding="utf-8")
|
| 307 |
+
|
| 308 |
+
# Clean up uploaded file
|
| 309 |
+
try:
|
| 310 |
+
os.remove(filepath)
|
| 311 |
+
except:
|
| 312 |
+
pass # Ignore cleanup errors
|
| 313 |
+
|
| 314 |
+
return jsonify({
|
| 315 |
+
'success': True,
|
| 316 |
+
'output_filename': output_filename,
|
| 317 |
+
'message': 'File converted successfully!'
|
| 318 |
+
})
|
| 319 |
+
else:
|
| 320 |
+
return jsonify({'success': False, 'message': 'Invalid file type. Please upload an Excel file (.xlsx or .xls)'})
|
| 321 |
+
|
| 322 |
+
except Exception as e:
|
| 323 |
+
return jsonify({'success': False, 'message': f'Conversion failed: {str(e)}'})
|
| 324 |
+
|
| 325 |
+
@app.route('/download-csv/<filename>')
|
| 326 |
+
def download_csv(filename):
|
| 327 |
+
"""Download converted CSV file"""
|
| 328 |
+
try:
|
| 329 |
+
file_path = os.path.join(app.config['PROCESSED_FOLDER'], filename)
|
| 330 |
+
if os.path.exists(file_path):
|
| 331 |
+
return send_file(file_path, as_attachment=True,
|
| 332 |
+
download_name=f"converted_{filename}")
|
| 333 |
+
else:
|
| 334 |
+
flash('File not found')
|
| 335 |
+
return redirect(url_for('excel_to_csv'))
|
| 336 |
+
except Exception as e:
|
| 337 |
+
flash(f'Download failed: {str(e)}')
|
| 338 |
+
return redirect(url_for('excel_to_csv'))
|
| 339 |
+
|
| 340 |
+
|
| 341 |
+
if __name__ == '__main__':
|
| 342 |
+
app.run(debug=True, host='0.0.0.0', port=5000)
|
requirements.txt
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Flask==2.3.3
|
| 2 |
+
Werkzeug==2.3.7
|
| 3 |
+
pandas==2.2.2
|
| 4 |
+
openpyxl==3.1.2
|
| 5 |
+
gradio==4.28.3
|
| 6 |
+
ocrmypdf==16.11.1
|