arXiv-dl

Command-line Paper Downloader for ArXiv, ECVA & CVF Open Access.

Disclaimer: This is a highly-opinionated command-line tool for downloading papers. It priorities ease of use for researchers. Obviously, this is not an ArXiv official project.

What does it do?

• Support downloading papers from ArXiv, ECCV, CVPR, ICCV, WACV via simple CLI.
• Support downloading speedup by using aria2.
• Retrieve the paper's metadata such as:
  • Title, Abstract, Year
  • Authors
  • Comments (Conference acceptance info)
  • Repository URLs
  • BibTeX Citation
• Automatically maintain a list of local papers and their metadata in a JSON file.
• Configure the desired download destination via an environment variable or a command-line argument.
• All downloaded papers will have standardized filename for easy browsing.

Why?

• Save time and effort to download and organize papers on your machine.
• Speedup downloading process by using multiple parallel connections.
• Local paper list would be handy for quick local lookup, making notes, and doing citations.

How to install it?

This is a command-line tool, simply use pip to install the package globally, then you are good to go!

• Pre-requisite: Python 3.x

python3 -m pip install -U arxiv-dl

NOTE: After installation, you need to ensure the installation path is included in your PATH variable. If you encounter any difficulty finding / setting the PATH, there is this recommended way of installing stand alone command line tools, kindly follow its instruction when installing arxiv-dl.

Optionally, install aria2c for download speedup.

• MacOS: brew install aria2
• Linux: sudo snap install aria2c

How to use it?

After installation, you may use the command paper in your shell to download papers. (Legacy commands arxiv-dl and getpaper are equivalent to the command paper.)

paper [OPTIONS] TARGET

Usage Examples:

# download a single TARGET
$ paper 1512.03385

# download multiple TARGETs separated by space
$ paper 2103.15538 2304.04415 https://arxiv.org/abs/1512.03385

Supported types of TARGET:

✅ Supported, 🚧 Not Yet Supported, ❌ Not Supported

• ArXiv
  • ✅ ArXiv ID: 1512.03385
  • ✅ ArXiv Abstract Page URL: https://arxiv.org/abs/1512.03385
  • ✅ ArXiv PDF Page URL: https://arxiv.org/pdf/1512.03385.pdf
• CVF Open Access (CVPR, ICCV, WACV)
  • ✅ CVF Abstract Page URL: https://openaccess.thecvf.com/content/**/html/**/*.html
  • ✅ CVF PDF Page URL: https://openaccess.thecvf.com/content/**/papers/**/*.pdf
• ECVA (ECCV)
  • ✅ ECVA Abstract Page URL: https://www.ecva.net/html/**/*.php
  • ❌ ECVA PDF Page URL: https://www.ecva.net/papers/**/*.pdf
• NeurIPS
  • 🚧 NeurIPS Abstract Page URL
  • 🚧 NeurIPS PDF Page URL
• OpenReview
  • 🚧 TODO

Supported OPTIONS:

• -v, --verbose (optional): Print paper metadata.
• -p, --pdf_only (optional): Download PDF only without creating Markdown notes
• -d, --download_dir (optional): Specify one-time download directory. This option will override the default download directory or the one specified in the environment variable ARXIV_DOWNLOAD_FOLDER.
• -n, --n_threads (optional): Specify the number of parallel connections to be used by aria2.

Configurations

Default Download Destination

• Without any configurations, all paper will be downloaded to $HOME/Downloads/ArXiv_Papers.

Set Your Custom Download Destination (Optional)

You may configure your preferred download destination once and for all via an environment variable. This will override the default download destination. To do that, include the following line in your .bashrc or .zshrc file:

export ARXIV_DOWNLOAD_FOLDER="YOUR/PATH/TO/ANY/FOLDER"

• Every time you use the paper command, the download destination will be set to the following order of priority:
  1. Command-line option -d
  2. Environment variable ARXIV_DOWNLOAD_FOLDER
  3. Default download destination

Set Custom Command Alias (Optional)

• You can always set your own preferred alias to rename the command or add more options.
• Include the following line(s) in your .bashrc or .zshrc file to set your preferred alias:

  alias dp="paper"
  alias dpv="paper -v -d '~/Documents/Papers'"

Development

Set up development environment

# create a virtual environment
python3 -m venv venv && source venv/bin/activate

# install dependencies
pip install -r requirements.txt

# install the package in editable mode & dev dependencies
pip install -e ".[dev]"

Run Tests

pytest

Build the package

make

Clean cache & build artifacts

make clean

License

MIT License - Copyright (c) 2021-2024 Mark H. Huang