Animadversio / pyarxiv

The arXiv client you deserve, upgraded

pyarxiv

pyarxiv is a wrapper for the API of Cornell University's famous repository for scientific papers.

Supports Python 2.7, 3.3-3.6+

Installation

pip install pyarxiv

Features

• Query the arXiv API (atom feed) in your code
• Use enums for arXiv categories
• Download papers in your code as PDF TODO - Do the above in the commandline

Update 2023-12-18

• Add sort by and sort order
• Add fields (all) for query, allowing multiple condition per field

Usage

CLI

pyarxiv-cli download -h
pyarxiv-cli query -h
# will download a couple of papers with given ids to folder /home/user, name them according to their titles,
# append their arxiv ids, and do not give progress feedback when each paper is downloaded
pyarxiv-cli download 1703.00001 1703.00002v1 ... --target-folder=/home/user --use-title-for-filename --append-id --silent

# Queries for papers with "Lorem" in them, maximally gets 5 papers (default 100), authors Einstein and Zweistein
# Other potential arguments are --abstract, --journalref and manualmode with --querystring
pyarxiv-cli query --title="Lorem" --max-results=5 --authors="A Einstein, B Zweistein"

You can also chain commands:

pyarxiv-cli download $(pyarxiv-cli query --title="WaveNet") --use-title-for-filename --append-id

Python

from pyarxiv import query, download_entries
from pyarxiv.arxiv_categories import ArxivCategory, arxiv_category_map
#query(max_results=100, ids=[], categories=[],
#                title='', authors='', abstract='', journal_ref='',
#                querystring='')
entries = query(title='WaveNet') 
titles = map(lambda x: x['title'], entries)
print(list(titles))


#download_entries(entries_or_ids_or_uris=[], target_folder='.',
#                     use_title_for_filename=False, append_id=False,
#                     progress_callback=(lambda x, y: id))
download_entries(entries)


entries_with_category = query([ArxivCategory.cs_AI])
print(arxiv_category_map(ArxivCategory.cs_AI))