feat: limit to number of urls checked

Author: phette23

example.env

@@ -2,6 +2,7 @@ ACCESS_ID=cca
 API_KEY="secret..."
 HOST=api.summon.serialssolutions.com
 KOHA_DOMAIN=library.cca.edu
+LINKCHECK_LIMIT=500
 LINKCHECK_LOGFILE=data/linkcheck.log
 LINKCHECK_REPORT="https://library.cca.edu/cgi-bin/koha/svc/report?id=345"
 LINKCHECK_OPAC_URL="https://library.cca.edu/cgi-bin/koha/opac-detail.pl?biblionumber={id}"

linkcheck/linkcheck.py

@@ -42,13 +42,17 @@ def quote(list):
 
 
 def main() -> None:
+    count = 0
     report = httpx.get(config["LINKCHECK_REPORT"])
     for bib in report.json():
         # bibs are arrays like [urls string, title, biblionumber]
         urls, title, id = bib
         # urls are separated by " | "
         urls = urls.split(" | ")
         for url in urls:
+            count += 1
+            if config.get("LINKCHECK_LIMIT") and count > int(config["LINKCHECK_LIMIT"]):
+                break
             try:
                 r = httpx.get(url, follow_redirects=True)
                 status = r.status_code
@@ -108,3 +112,4 @@ def signal_handler(sig, frame) -> None:
     # TODO but the script keeps running
     signal.signal(signal.SIGINT, signal_handler)
     main()
+    summarize()

linkcheck/readme.md

@@ -6,6 +6,7 @@ Takes a public Koha report and checks each URL (`856$u`) to see if they resolve
 
 The script uses the same .env file as the root project or it can take environment variables.
 
+- `LINKCHECK_LIMIT` number of links to check (leave undefined for all of them)
 - `LINKCHECK_REPORT` URL to a Koha report that returns item URLs (see [report.sql](./report.sql)). Report must be Public.
 - `LINKCHECK_OPAC_URL` catalog link for individual records, should include `biblionumber={id}` in it (id is interpolated)
 - `LINKCHECK_LOGFILE` path to logged CSV, defaults to the data dir named "YYYY-MM-DD-linkcheck.csv" with today's date