Fix 'filename does not exists' error in Docling

import os
from docling.document_converter import DocumentConverter

# Set the path to artifacts
os.environ["DOCLING_ARTIFACTS_PATH"] = "/path/to/artifacts/directory"

source = "https://arxiv.org/pdf/2408.09869"
converter = DocumentConverter()
result = converter.convert(source)
print(result.document.export_to_markdown())

# Complete library reinstall
pip uninstall docling docling-parse
pip install docling --no-cache-dir

# Or using pip cache purge
pip cache purge
pip install docling

from docling.document_converter import DocumentConverter
from docling.datamodel.pipeline_options import PdfPipelineOptions
import os

# Create artifacts directory if it doesn't exist
artifacts_path = "./docling_artifacts"
os.makedirs(artifacts_path, exist_ok=True)

# Configure options with explicit paths
pdf_pipeline_options = PdfPipelineOptions(
    artifacts_path=artifacts_path,
    generate_page_images=False,
    generate_picture_images=False,
)

# Create converter with options
converter = DocumentConverter(
    format_options={
        "pdf": {
            "pipeline_options": pdf_pipeline_options
        }
    }
)

source = "https://arxiv.org/pdf/2408.09869"
result = converter.convert(source)

import sys
from pathlib import Path

# Check installation paths
for path in sys.path:
    resource_path = Path(path) / "docling_parse" / "pdf_resources_v2" / "glyphs" / "standard" / "additional.dat"
    if resource_path.exists():
        print(f"File found: {resource_path}")
    else:
        print(f"File missing: {resource_path}")

from docling.document_converter import DocumentConverter
from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend

# Use alternative backend
converter = DocumentConverter(backend_class=PyPdfiumDocumentBackend)

source = "https://arxiv.org/pdf/2408.09869"
result = converter.convert(source)
print(result.document.export_to_markdown())

from docling.document_converter import DocumentConverter
from docling.datamodel.pipeline_options import PdfPipelineOptions, AcceleratorOptions, EasyOcrOptions

pdf_pipeline_options = PdfPipelineOptions(
    accelerator_options=AcceleratorOptions(device="auto"),
    ocr_options=EasyOcrOptions(force_full_page_ocr=False, use_gpu=False),
    do_ocr=False,  # Disable OCR if not required
    table_structure_options=None,  # Disable table recognition
    generate_page_images=False,
    generate_picture_images=False,
)

converter = DocumentConverter(
    format_options={
        "pdf": {
            "pipeline_options": pdf_pipeline_options
        }
    }
)

# Create a clean environment
python -m venv clean_docling_env
source clean_docling_env/bin/activate  # For Linux/Mac
# or
clean_docling_env\Scripts\activate  # For Windows

# Install in clean environment
pip install docling

pip install --upgrade docling

import logging
from docling.document_converter import DocumentConverter

# Configure logging
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)

converter = DocumentConverter()
try:
    result = converter.convert("https://arxiv.org/pdf/2408.09869")
    print(result.document.export_to_markdown())
except Exception as e:
    logger.error(f"Conversion error: {e}")

from docling.document_converter import DocumentConverter
import os

def test_docling_installation():
    try:
        # Check basic functionality
        converter = DocumentConverter()
        print("✓ Basic converter created successfully")
        
        # Check resource paths
        import sys
        from pathlib import Path
        
        for path in sys.path:
            resource_dir = Path(path) / "docling_parse" / "pdf_resources_v2"
            if resource_dir.exists():
                print(f"✓ Resource directory found: {resource_dir}")
                return True
        
        print("✗ Resource directory not found")
        return False
        
    except Exception as e:
        print(f"✗ Error during verification: {e}")
        return False

if __name__ == "__main__":
    test_docling_installation()