Knowing the actual fetchPriority of the initial request is useful information
when computing the urgency. This allows us to not worry about the priority
being clobbered by another caller of nsISupportsPriority.
Differential Revision: https://phabricator.services.mozilla.com/D221957
Knowing the actual fetchPriority of the initial request is useful information
when computing the urgency. This allows us to not worry about the priority
being clobbered by another caller of nsISupportsPriority.
Differential Revision: https://phabricator.services.mozilla.com/D221957
Knowing the actual fetchPriority of the initial request is useful information
when computing the urgency. This allows us to not worry about the priority
being clobbered by another caller of nsISupportsPriority.
Differential Revision: https://phabricator.services.mozilla.com/D221957
To not unexpectedly downgrade requests that weren't upgraded via HTTPS-First.
From the added tests only the http redirect test cases start passing
with this patch. META and JS redirect already pass before this patch.
Differential Revision: https://phabricator.services.mozilla.com/D215214
Http2 uses a single connection for multiple requests. When ORB blocks a
request with NS_ERROR_FAILURE, it's treated as a hard error in necko,
then the entire connection is closed.
This patches converts ORB to uses NS_BINDING_ABORTED.
Differential Revision: https://phabricator.services.mozilla.com/D218643
This changes where the IsUpgradeDowngradeEndlessLoop check triggers.
Before this patch, it triggered during the redirect caused by the https
upgrade. With this patch, it triggers during the downgrade for http
redirects. META and JS redirect are still detected during upgrade.
This should be fixed as a follow up (See Bug 1896691).
Downgrade in this context means same url, except with the scheme http
instead of https.
Different query parameters normally lead to different responses by web servers.
Don't consider the '#ref' part of the uri, because it doesn't get send to
the server and therefore can't change the server response.
We can't use the redirect chain anymore, because the query parameters
are trimmed since Bug 1715785.
This also removes the config option dom.security.https_only_check_path_upgrade_downgrade_endless_loop,
because it adds unnecessary complexity. Removing it for this patch is
easier.
https-only, https-first and httpssvc_https_upgrade tests had to be
modified, because they depended on the incorrect handling of query
strings in loop detection.
Differential Revision: https://phabricator.services.mozilla.com/D193672
This variety is specifically used to make the User-Agent header be treated as
a default request header, but it wasn't used properly of the UA override.
This caused the header to be copied by CopyNonDefaultHeaderVisitor.
Differential Revision: https://phabricator.services.mozilla.com/D210478
This changes where the IsUpgradeDowngradeEndlessLoop check triggers.
Before this patch, it triggered during the redirect caused by the https
upgrade. With this patch, it triggers during the downgrade for http
redirects. META and JS redirect are still detected during upgrade.
This should be fixed as a follow up (See Bug 1896691).
Downgrade in this context means same url, except with the scheme http
instead of https.
Different query parameters normally lead to different responses by web servers.
Don't consider the '#ref' part of the uri, because it doesn't get send to
the server and therefore can't change the server response.
We can't use the redirect chain anymore, because the query parameters
are trimmed since Bug 1715785.
This also removes the config option dom.security.https_only_check_path_upgrade_downgrade_endless_loop,
because it adds unnecessary complexity. Removing it for this patch is
easier.
https-only, https-first and httpssvc_https_upgrade tests had to be
modified, because they depended on the incorrect handling of query
strings in loop detection.
Differential Revision: https://phabricator.services.mozilla.com/D193672
For WebDriver BiDi network interception we need to be able to modify the request as late as possible.
This changeset exposes RequestObserversCalled to JS so that we can update it from the webdriver bidi codebase.
Differential Revision: https://phabricator.services.mozilla.com/D208011
This patch is a no-op for observable functionality. All it does is add a bit to the partitionKey for OA, and then adds plumbing to be able to set it, and where it is required, sets it to false.
This is serialized identically to the absence of the bit, so nothing changes at all.
Differential Revision: https://phabricator.services.mozilla.com/D203155
This patch is a no-op for observable functionality. All it does is add a bit to the partitionKey for OA, and then adds plumbing to be able to set it, and where it is required, sets it to false.
This is serialized identically to the absence of the bit, so nothing changes at all.
Differential Revision: https://phabricator.services.mozilla.com/D203155
This patch is a no-op for observable functionality. All it does is add a bit to the partitionKey for OA, and then adds plumbing to be able to set it, and where it is required, sets it to false.
This is serialized identically to the absence of the bit, so nothing changes at all.
Differential Revision: https://phabricator.services.mozilla.com/D203155
This patch is a no-op for observable functionality. All it does is add a bit to the partitionKey for OA, and then adds plumbing to be able to set it, and where it is required, sets it to false.
This is serialized identically to the absence of the bit, so nothing changes at all.
Differential Revision: https://phabricator.services.mozilla.com/D203155
This notification would be sent by HttpChannelChild before
calling onStopRequest for a channel.
This patch also updates some comments regarding the on stop topic
which referenced the non-existent http-on-stop-connect
Differential Revision: https://phabricator.services.mozilla.com/D204482
