Skip to content

PdfMinerExtractor

We provide a PDF line extractor built on top of PdfMiner.

This is the most portable extractor, since it is pure-python and can therefore be run on any platform. Be sure to have a look at their documentation, especially the part providing a bird's eye view of the PDF extraction process.

Usage

from edspdf import Pipeline
from pathlib import Path

# Add the component to a new pipeline
model = Pipeline()
model.add_pipe(
    "pdfminer-extractor",
    config=dict(
        extract_style=False,
    ),
)

# Apply on a new document
model(Path("path/to/your/pdf/document").read_bytes())

Configuration

Parameter Description Default
line_overlap See PDFMiner documentation 0.5
char_margin See PDFMiner documentation 2.05
line_margin See PDFMiner documentation 0.5
word_margin See PDFMiner documentation 0.1
boxes_flow See PDFMiner documentation 0.5
detect_vertical See PDFMiner documentation False
all_texts See PDFMiner documentation False
extract_style Whether to extract style (font, size, ...) information for each line of the document. False