Opened 4 years ago
Closed 4 years ago
#52542 closed defect (bug) (fixed)
HTTPS local html detection RSD check is protocol specific
Reported by: | TimothyBlynJacobs | Owned by: | flixos90 |
---|---|---|---|
Milestone: | 5.7 | Priority: | normal |
Severity: | normal | Version: | 5.7 |
Component: | Site Health | Keywords: | good-first-bug has-patch has-unit-tests dev-feedback commit |
Focuses: | Cc: |
Description
The wp_is_local_html_output
function is used to check whether the output from the https detection loopback request comes from the WordPress site itself.
The first check is done is for RSD, but it checks specifically for the http
version of the URL, whereas the other checks ignore the protocol.
In my testing environment at least, the RSD link header is output as an https
url when the site is requested over https
, and as such this check returns false. Incorrectly asserting in Site Health that I need to talk to my host about supporting HTTPS even though my site already supports it.
Tentatively milestoning for 5.7.
Change History (8)
#3
@
4 years ago
@lukecarbis Yep! Sorry for not including that in the original report! Yeah, I believe @flixos90 is planning on working on it this week.
This ticket was mentioned in PR #1023 on WordPress/wordpress-develop by felixarntz.
4 years ago
#5
- Keywords has-patch has-unit-tests added; needs-patch removed
- Updates the RSD link check in
wp_is_local_html_output()
to be protocol-agnostic as expected, like the other ones. - Updates a minor issue in the REST API link check: The removal of the protocol should happen on the complete URL as included in the HTML output, which is run through
esc_url
. Beforeesc_url
was called after removing the protocol which is not correct.
Trac ticket: https://core.trac.wordpress.org/ticket/52542
#6
@
4 years ago
- Keywords dev-feedback added
- Owner changed from flixos90 to TimothyBlynJacobs
- Status changed from assigned to reviewing
@TimothyBlynJacobs I've added PR to address this bug. Can you review please?
@TimothyBlynJacobs That sounds like a bug specific to the changes to Site Health in the 5.7 beta, is that right? Possibly introduced by [49904]. Is this something we could get fixed in the next week before RC1 is out?