diff --git a/lychee.toml b/lychee.toml index b35f38cd..0b66dfef 100644 --- a/lychee.toml +++ b/lychee.toml @@ -27,7 +27,8 @@ cache_exclude_status = "500.." ############################# Runtime ############################## -max_concurrency = 14 +# Reduce concurrency since lychee can check 500+ links in 1 minute with 1 worker +max_concurrency = 1 max_redirects = 10 max_retries = 2 # Wait longer between retries to give slow servers room to recover. @@ -37,8 +38,9 @@ timeout = 45 ############################# Requests ############################# -# Some sites block lychee's default UA. Use a generic browser-ish UA. -user_agent = "Mozilla/5.0 (compatible; lychee-link-checker)" +# Some sites block lychee's default user agent (curl). Use a generic browser. +user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" + # Accept 200–403 as success. 401/403 are common on bot-protected or # auth-walled pages we still want to consider "reachable". @@ -59,8 +61,6 @@ include_verbatim = false # Regex excludes. Plain-domain skips belong in .lycheeignore. exclude = [ + 'https://(www\.)?neonscience\.org/.*', 'zenodo\.org', - # Self-references to the deployed site lag the build, producing false - # 404s for new pages. Internal links are validated via relative paths. - '^https?://(www\.)?pyopensci\.org/', ]