-
-
tried uploading 237 files, no errors, yet it didn't upload quite a few of them
-
wget-ing the public site, unfortunately, doesn't result in a list of files with links to downloads - can this be fixed?
Otherwise my Debian maintainer friends complain about not being able to upload packages. Essentially, the site should work without Javascript as well.
-
-
The first one is most definitely a bug which needs fixing. Can you maybe give some instructions how to reproduce this? Did you upload via the cli, webdav, or webui?
The second issue I guess is now there since surfer has that new file listing, which is actually a vuejs app, mostly the same as the actual surfer admin ui without edit options.
I wonder if it is possible to actually deliver a whole different page for wget and the likes, have never thought about that use-case. @mario can you also explain what your debian maintainer friend would like to have to be happy
-
The first one is most definitely a bug which needs fixing. Can you maybe give some instructions how to reproduce this? Did you upload via the cli, webdav, or webui?
The second issue I guess is now there since surfer has that new file listing, which is actually a vuejs app, mostly the same as the actual surfer admin ui without edit options.
I wonder if it is possible to actually deliver a whole different page for wget and the likes, have never thought about that use-case. @mario can you also explain what your debian maintainer friend would like to have to be happy@nebulon I used the web UI. The web UI seemed to accept everything, it finishes, but then when I go look for it there are a dozen files missing. The second time I tried to upload the same set of files, it uploads a few new ones from those 237 but doesn't upload all again.
And absolutely happy to explain - so Debian uses scripts which run every once in a while to parse a certain website for new releases. Since those scripts rely on wget and other tools which do not support javascript, they basically cannot see anything here.
So even just an index.html with a bunch of links on it that lead to files download is an improvement over the current situation.
-
Are there any special file/folder names amongst those 237 files? Also how large are they? Can you maybe also try with a different browser?
For the second thing, does anyone know how to maybe properly distinguish between an agent like wget and normal browser? I think it would make more sense to simply deliver very basic html for such agents, while keeping it as if for actual browsers.
-
Are there any special file/folder names amongst those 237 files? Also how large are they? Can you maybe also try with a different browser?
For the second thing, does anyone know how to maybe properly distinguish between an agent like wget and normal browser? I think it would make more sense to simply deliver very basic html for such agents, while keeping it as if for actual browsers.
@nebulon files are between 64B and 1.5MB each, so not big. I can try with a different browser, of course.
What I would suggest is to put a <noscript> tag in there (with links to download file, and potentially filesname too) which would basically be shown if you don't have javascript. This would resolve the problem.
