aboutsummaryrefslogtreecommitdiffstats
path: root/extra
diff options
context:
space:
mode:
Diffstat (limited to 'extra')
-rwxr-xr-xextra/sitemap/access_urls_query.sh3
1 files changed, 2 insertions, 1 deletions
diff --git a/extra/sitemap/access_urls_query.sh b/extra/sitemap/access_urls_query.sh
index bc7035f..284b52b 100755
--- a/extra/sitemap/access_urls_query.sh
+++ b/extra/sitemap/access_urls_query.sh
@@ -4,11 +4,12 @@ set -e # fail on error
set -u # fail if variable not set in substitution
set -o pipefail # fail if part of a '|' command fails
+SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
# 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 \
+ | $SCRIPT_DIR/transform_access_url.py \
| split --lines 20000 - sitemap-access- -d -a 5 --additional-suffix .txt