phantop/quatalog-scraper
1
0
Fork 0
mirror of https://github.com/quatalog/quatalog.git synced 2024-11-18 04:52:51 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update transfer.yml

Browse source
This commit is contained in:
powe97 2024-04-08 14:32:48 -04:00 committed by GitHub
parent 3b16ecc9bb
commit c7af001773
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
.github/workflows/transfer.yml vendored
View file

@ -13,7 +13,11 @@ jobs:
- name: Create matrix parameters - name: Create matrix parameters
id: matrix-params id: matrix-params
run: | run: |
NUM_PAGES="$(curl -s -H 'User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:122.0) Gecko/20100101 Firefox/122.0' 'https://tes.collegesource.com/publicview/TES_publicview01.aspx?rid=f080a477-bff8-46df-a5b2-25e9affdd4ed&aid=27b576bb-cd07-4e57-84d0-37475fde70ce' | grep -e 'lblInstWithEQPaginationInfo' | grep -Poie '(?<=of )[0-9]*')" UA='User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:122.0) Gecko/20100101 Firefox/122.0'
URL='https://tes.collegesource.com/publicview/TES_publicview01.aspx?rid=f080a477-bff8-46df-a5b2-25e9affdd4ed&aid=27b576bb-cd07-4e57-84d0-37475fde70ce'
HTML="$(curl -s -H "$UA" "$URL")"
echo "$HTML" | grep "<title>"
NUM_PAGES="$(echo "$HTML" | grep -e 'lblInstWithEQPaginationInfo' | grep -Poie '(?<=of )[0-9]*')"
curl "Found $NUM_PAGES pages" curl "Found $NUM_PAGES pages"
MATRIX_PARAMS="$(seq -s "," 1 "$NUM_PAGES")" MATRIX_PARAMS="$(seq -s "," 1 "$NUM_PAGES")"
MATRIX_PARAMS="\"page\": $(sed -e 's/,/}, {"page": /g' <<< "$MATRIX_PARAMS")" MATRIX_PARAMS="\"page\": $(sed -e 's/,/}, {"page": /g' <<< "$MATRIX_PARAMS")"