| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
Sorry, something went wrong.
|
Could you explain how this is necessary when we can just use Sphinx's built-in -n for internal links (which only takes a few minutes) and its linkcheck builder for external links (which takes a couple tens of minutes)? |
Sorry, something went wrong.
Generated parts of HTML pages and external links also need to be checked. According to a size of gh-93853, we need either a tool like the one in this PR or a Sphinx extension. |
Sorry, something went wrong.
|
I'm afraid I'm still confused. If a link is an external link in the source (even generated by standard or most third-party roles), it will be caught by linkcheck if broken or redirected. If a link is an internal link, -n will immediately catch it at build time, far faster than scraping every link. Aside from possible very narrow corner cases (which I have yet to see conclusively illustrated), I still don't understand what this bespoke manual script usefully does that a combination of -n and linkcheck doesn't, faster, more efficiently and without having to maintain a bespoke solution. @AA-Turner is there something I'm missing here? |
Sorry, something went wrong.
My apologises, I've totally missed this part of your first comment:
|
Sorry, something went wrong.
| Back | FazBrowse Home | New Git URL |
Broken links from a parent issue were found using this tool.
C:\Users\oleg\Documents\dev\notmine\cpython>python Tools/scripts/checkhtmllinks.py -h usage: checkhtmllinks.py [-h] [-r] [-l LIMIT] path Check if specified HTML files have dead or redirected links. positional arguments: path a glob pattern of file paths to scan options: -h, --help show this help message and exit -r, --allow-redirects do not report HTTP 3xx links as kind-of-broken -l LIMIT, --limit LIMIT skip files that contain more links than specified Call this script on HTML files of the rendered documentation. Eventhough the script is multithreaded and findings for already processed pages are cached, a full run through the whole rendered documentation takes about an hour.