blob: fb1a4b8e447ef4f88092b017106600f13c51af80 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
#!/usr/bin/env bash
set -e # fail on error
set -u # fail if variable not set in substitution
set -o pipefail # fail if part of a '|' command fails
# query for specific works; about 8.6 million circa 2021-04-29
fatcat-cli search scholar 'doc_type:work (fulltext.access_type:ia_file OR fulltext.access_type:wayback) (NOT biblio.arxiv_id:*) (NOT biblio.pmcid:*) (NOT biblio.publisher_type:big5) (year:<1926 OR tags:*)' --index-json --limit 0 \
| pv -l \
| jq '[.fulltext.access_type, .fulltext.access_url] | @tsv' -r \
| rg -v '^null' \
| ./transform_access_url.py \
| split --lines 20000 - sitemap-pdfs- -d -a 5 --additional-suffix .txt
|
