You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
My server is set to redirect all HTTP requests from http://scott.mn to https://scott.mn, and the latter is my entity URI. So if I put in http://scott.mn, python-tent-client should follow the redirect, find the Link: header at https://scott.mn, and successfully discover the profile URL (and from that, the API root).
What happens instead is that it fails to find any profile URLs.
Interestingly, http://graue.tent.isdoes work even though Tent.is redirects http to https in the same manner. But Tent.is uses both a Link: HTTP header and an HTML <link> tag, while my site uses only the HTTP header.
So I suspect what's happening is that Requests doesn't follow a redirect when you do a HEAD request, but does for a GET request. It's doing a HEAD against the http site and not finding any Link: headers. Then it's retrying as a GET but this time against the https site, but it's not noticing that the https site has Link: headers because you only check for the <link> tag in this case.
If that guess is correct, making Requests follow redirects during the initial HEAD request would fix this bug.
The text was updated successfully, but these errors were encountered:
My server is set to redirect all HTTP requests from
http://scott.mn
tohttps://scott.mn
, and the latter is my entity URI. So if I put inhttp://scott.mn
, python-tent-client should follow the redirect, find the Link: header athttps://scott.mn
, and successfully discover the profile URL (and from that, the API root).What happens instead is that it fails to find any profile URLs.
Interestingly,
http://graue.tent.is
does work even though Tent.is redirects http to https in the same manner. But Tent.is uses both a Link: HTTP header and an HTML<link>
tag, while my site uses only the HTTP header.So I suspect what's happening is that Requests doesn't follow a redirect when you do a HEAD request, but does for a GET request. It's doing a HEAD against the http site and not finding any Link: headers. Then it's retrying as a GET but this time against the https site, but it's not noticing that the https site has Link: headers because you only check for the
<link>
tag in this case.If that guess is correct, making Requests follow redirects during the initial HEAD request would fix this bug.
The text was updated successfully, but these errors were encountered: